1.1. Obtaining and installing samba

Binary packages of samba are included in almost any Linux or Unix distribution. There are also some packages available at the samba homepage

If you need to compile samba from source, check the appropriate appendix chapter.

1.2. Configuring samba

Samba's configuration is stored in the smb.conf file, that usually resides in /etc/samba/smb.conf or /usr/local/samba/lib/smb.conf. You can either edit this file yourself or do it using one of the many graphical tools that are available, such as the web-based interface swat, that is included with samba.

	[global]
	   workgroup = MYGROUP

	   [homes]
	      guest ok = no
	      read only = no
	

1.3. Try listing the shares available on your server

$ smbclient -L yourhostname

You should get back a list of shares available on your server. If you don't then something is incorrectly setup. Note that this method can also be used to see what shares are available on other LanManager clients (such as WfWg).

If you choose user level security then you may find that Samba requests a password before it will list the shares. See the smbclient man page for details. (you can force it to list the shares without a password by adding the option -U% to the command line. This will not work with non-Samba servers)

1.4. Try connecting with the unix client

$ smbclient //yourhostname/aservice

Typically the yourhostname would be the name of the host where you installed smbd. The aservice is any service you have defined in the smb.conf file. Try your user name if you just have a [homes] section in smb.conf.

For example if your unix host is bambi and your login name is fred you would type:

$ smbclient //bambi/fred

1.5. Try connecting from a DOS, WfWg, Win9x, WinNT, Win2k, OS/2, etc... client

Try mounting disks. eg:

C:\WINDOWS\> net use d: \\servername\service

Try printing. eg:

C:\WINDOWS\> net use lpt1: \\servername\spoolservice

C:\WINDOWS\> print filename

Celebrate, or send me a bug report!

1.6. What If Things Don't Work?

Then you might read the file HOWTO chapter Diagnosis and the FAQ. If you are still stuck then try the mailing list or newsgroup (look in the README for details). Samba has been successfully installed at thousands of sites worldwide, so maybe someone else has hit your problem and has overcome it. You could also use the WWW site to scan back issues of the samba-digest.

When you fix the problem please send some updates of the documentation (or source code) to one of the documentation maintainers or the list.

2.1. Discussion

Firstly, all MS Windows networking is based on SMB (Server Message Block) based messaging. SMB messaging may be implemented using NetBIOS or without NetBIOS. Samba implements NetBIOS by encapsulating it over TCP/IP. MS Windows products can do likewise. NetBIOS based networking uses broadcast messaging to affect browse list management. When running NetBIOS over TCP/IP this uses UDP based messaging. UDP messages can be broadcast or unicast.

Normally, only unicast UDP messaging can be forwarded by routers. The remote announce parameter to smb.conf helps to project browse announcements to remote network segments via unicast UDP. Similarly, the remote browse sync parameter of smb.conf implements browse list collation using unicast UDP.

Secondly, in those networks where Samba is the only SMB server technology wherever possible nmbd should be configured on one (1) machine as the WINS server. This makes it easy to manage the browsing environment. If each network segment is configured with it's own Samba WINS server, then the only way to get cross segment browsing to work is by using the remote announce and the remote browse sync parameters to your smb.conf file.

If only one WINS server is used for an entire multi-segment network then the use of the remote announce and the remote browse sync parameters should NOT be necessary.

As of Samba 3 WINS replication is being worked on. The bulk of the code has been committed, but it still needs maturation.

Right now samba WINS does not support MS-WINS replication. This means that when setting up Samba as a WINS server there must only be one nmbd configured as a WINS server on the network. Some sites have used multiple Samba WINS servers for redundancy (one server per subnet) and then used remote browse sync and remote announce to affect browse list collation across all segments. Note that this means clients will only resolve local names, and must be configured to use DNS to resolve names on other subnets in order to resolve the IP addresses of the servers they can see on other subnets. This setup is not recommended, but is mentioned as a practical consideration (ie: an 'if all else fails' scenario).

Lastly, take note that browse lists are a collection of unreliable broadcast messages that are repeated at intervals of not more than 15 minutes. This means that it will take time to establish a browse list and it can take up to 45 minutes to stabilise, particularly across network segments.

2.2. How browsing functions and how to deploy stable and dependable browsing using Samba

As stated above, MS Windows machines register their NetBIOS names (i.e.: the machine name for each service type in operation) on start up. Also, as stated above, the exact method by which this name registration takes place is determined by whether or not the MS Windows client/server has been given a WINS server address, whether or not LMHOSTS lookup is enabled, or if DNS for NetBIOS name resolution is enabled, etc.

In the case where there is no WINS server all name registrations as well as name lookups are done by UDP broadcast. This isolates name resolution to the local subnet, unless LMHOSTS is used to list all names and IP addresses. In such situations Samba provides a means by which the samba server name may be forcibly injected into the browse list of a remote MS Windows network (using the remote announce parameter).

Where a WINS server is used, the MS Windows client will use UDP unicast to register with the WINS server. Such packets can be routed and thus WINS allows name resolution to function across routed networks.

During the startup process an election will take place to create a local master browser if one does not already exist. On each NetBIOS network one machine will be elected to function as the domain master browser. This domain browsing has nothing to do with MS security domain control. Instead, the domain master browser serves the role of contacting each local master browser (found by asking WINS or from LMHOSTS) and exchanging browse list contents. This way every master browser will eventually obtain a complete list of all machines that are on the network. Every 11-15 minutes an election is held to determine which machine will be the master browser. By the nature of the election criteria used, the machine with the highest uptime, or the most senior protocol version, or other criteria, will win the election as domain master browser.

Clients wishing to browse the network make use of this list, but also depend on the availability of correct name resolution to the respective IP address/addresses.

Any configuration that breaks name resolution and/or browsing intrinsics will annoy users because they will have to put up with protracted inability to use the network services.

Samba supports a feature that allows forced synchonisation of browse lists across routed networks using the remote browse sync parameter in the smb.conf file. This causes Samba to contact the local master browser on a remote network and to request browse list synchronisation. This effectively bridges two networks that are separated by routers. The two remote networks may use either broadcast based name resolution or WINS based name resolution, but it should be noted that the remote browse sync parameter provides browse list synchronisation - and that is distinct from name to address resolution, in other words, for cross subnet browsing to function correctly it is essential that a name to address resolution mechanism be provided. This mechanism could be via DNS, /etc/hosts, and so on.

2.3. Use of the Remote Announce parameter

The remote announce parameter of smb.conf can be used to forcibly ensure that all the NetBIOS names on a network get announced to a remote network. The syntax of the remote announce parameter is:

	remote announce = a.b.c.d [e.f.g.h] ...

	remote announce = a.b.c.d/WORKGROUP [e.f.g.h/WORKGROUP] ...

2.4. Use of the Remote Browse Sync parameter

The remote browse sync parameter of smb.conf is used to announce to another LMB that it must synchronise it's NetBIOS name list with our Samba LMB. It works ONLY if the Samba server that has this option is simultaneously the LMB on it's network segment.

The syntax of the remote browse sync parameter is:

remote browse sync = a.b.c.d

2.5. Use of WINS

Use of WINS (either Samba WINS _or_ MS Windows NT Server WINS) is highly recommended. Every NetBIOS machine registers it's name together with a name_type value for each of of several types of service it has available. eg: It registers it's name directly as a unique (the type 0x03) name. It also registers it's name if it is running the lanmanager compatible server service (used to make shares and printers available to other users) by registering the server (the type 0x20) name.

All NetBIOS names are up to 15 characters in length. The name_type variable is added to the end of the name - thus creating a 16 character name. Any name that is shorter than 15 characters is padded with spaces to the 15th character. ie: All NetBIOS names are 16 characters long (including the name_type information).

WINS can store these 16 character names as they get registered. A client that wants to log onto the network can ask the WINS server for a list of all names that have registered the NetLogon service name_type. This saves broadcast traffic and greatly expedites logon processing. Since broadcast name resolution can not be used across network segments this type of information can only be provided via WINS _or_ via statically configured lmhosts files that must reside on all clients in the absence of WINS.

WINS also serves the purpose of forcing browse list synchronisation by all LMB's. LMB's must synchronise their browse list with the DMB (domain master browser) and WINS helps the LMB to identify it's DMB. By definition this will work only within a single workgroup. Note that the domain master browser has NOTHING to do with what is referred to as an MS Windows NT Domain. The later is a reference to a security environment while the DMB refers to the master controller for browse list information only.

Use of WINS will work correctly only if EVERY client TCP/IP protocol stack has been configured to use the WINS server/s. Any client that has not been configured to use the WINS server will continue to use only broadcast based name registration so that WINS may NEVER get to know about it. In any case, machines that have not registered with a WINS server will fail name to address lookup attempts by other clients and will therefore cause workstation access errors.

To configure Samba as a WINS server just add wins support = yes to the smb.conf file [globals] section.

To configure Samba to register with a WINS server just add "wins server = a.b.c.d" to your smb.conf file [globals] section.

DO NOT EVER use both "wins support = yes" together with "wins server = a.b.c.d" particularly not using it's own IP address. Specifying both will cause nmbd to refuse to start!

2.6. Do NOT use more than one (1) protocol on MS Windows machines

A very common cause of browsing problems results from installing more than one protocol on an MS Windows machine.

Every NetBIOS machine takes part in a process of electing the LMB (and DMB) every 15 minutes. A set of election criteria is used to determine the order of precidence for winning this election process. A machine running Samba or Windows NT will be biased so that the most suitable machine will predictably win and thus retain it's role.

The election process is "fought out" so to speak over every NetBIOS network interface. In the case of a Windows 9x machine that has both TCP/IP and IPX installed and has NetBIOS enabled over both protocols the election will be decided over both protocols. As often happens, if the Windows 9x machine is the only one with both protocols then the LMB may be won on the NetBIOS interface over the IPX protocol. Samba will then lose the LMB role as Windows 9x will insist it knows who the LMB is. Samba will then cease to function as an LMB and thus browse list operation on all TCP/IP only machines will fail.

Windows 95, 98, 98se, Me are referred to generically as Windows 9x. The Windows NT4, 2000, XP and 2003 use common protocols. These are roughly referred to as the WinNT family, but it should be recognised that 2000 and XP/2003 introduce new protocol extensions that cause them to behave differently from MS Windows NT4. Generally, where a server does NOT support the newer or extended protocol, these will fall back to the NT4 protocols.

The safest rule of all to follow it this - USE ONLY ONE PROTOCOL!

2.7. Name Resolution Order

Resolution of NetBIOS names to IP addresses can take place using a number of methods. The only ones that can provide NetBIOS name_type information are:

Alternative means of name resolution includes:

Many sites want to restrict DNS lookups and want to avoid broadcast name resolution traffic. The "name resolve order" parameter is of great help here. The syntax of the "name resolve order" parameter is:

name resolve order = wins lmhosts bcast host

name resolve order = wins lmhosts  	(eliminates bcast and host)

name  resolve order = host lmhost wins bcast

3.1. Introduction

Old windows clients send plain text passwords over the wire. Samba can check these passwords by crypting them and comparing them to the hash stored in the unix user database.

Newer windows clients send encrypted passwords (so-called Lanman and NT hashes) over the wire, instead of plain text passwords. The newest clients will only send encrypted passwords and refuse to send plain text passwords, unless their registry is tweaked.

These passwords can't be converted to unix style encrypted passwords. Because of that you can't use the standard unix user database, and you have to store the Lanman and NT hashes somewhere else.

Next to a differently encrypted passwords, windows also stores certain data for each user that is not stored in a unix user database, e.g. workstations the user may logon from, the location where his/her profile is stored, etc. Samba retrieves and stores this information using a "passdb backend". Commonly available backends are LDAP, plain text file, MySQL and nisplus. For more information, see the documentation about the passdb backend = parameter.

3.2. Important Notes About Security

The unix and SMB password encryption techniques seem similar on the surface. This similarity is, however, only skin deep. The unix scheme typically sends clear text passwords over the network when logging in. This is bad. The SMB encryption scheme never sends the cleartext password over the network but it does store the 16 byte hashed values on disk. This is also bad. Why? Because the 16 byte hashed values are a "password equivalent". You cannot derive the user's password from them, but they could potentially be used in a modified client to gain access to a server. This would require considerable technical knowledge on behalf of the attacker but is perfectly possible. You should thus treat the data stored in whatever passdb backend you use (smbpasswd file, ldap, mysql) as though it contained the cleartext passwords of all your users. Its contents must be kept secret, and the file should be protected accordingly.

Ideally we would like a password scheme which neither requires plain text passwords on the net or on disk. Unfortunately this is not available as Samba is stuck with being compatible with other SMB systems (WinNT, WfWg, Win95 etc).

Note that Windows NT 4.0 Service pack 3 changed the default for permissible authentication so that plaintext passwords are never sent over the wire. The solution to this is either to switch to encrypted passwords with Samba or edit the Windows NT registry to re-enable plaintext passwords. See the document WinNT.txt for details on how to do this. Other Microsoft operating systems which also exhibit this behavior includes These versions of MS Windows do not support full domain security protocols, although they may log onto a domain environment. Of these Only MS Windows XP Home does NOT support domain logons. MS DOS Network client 3.0 with the basic network redirector installed Windows 95 with the network redirector update installed Windows 98 [se] Windows Me Windows XP Home The following versions of MS Windows fully support domain security protocols. Windows NT 3.5x Windows NT 4.0 Windows 2000 Professional Windows 200x Server/Advanced Server Windows XP Professional Note :All current release of Microsoft SMB/CIFS clients support authentication via the SMB Challenge/Response mechanism described here. Enabling clear text authentication does not disable the ability of the client to participate in encrypted authentication. MS Windows clients will cache the encrypted password alone. Even when plain text passwords are re-enabled, through the appropriate registry change, the plain text password is NEVER cached. This means that in the event that a network connections should become disconnected (broken) only the cached (encrypted) password will be sent to the resource server to affect a auto-reconnect. If the resource server does not support encrypted passwords the auto-reconnect will fail. USE OF ENCRYPTED PASSWORDS IS STRONGLY ADVISED.

3.3. The smbpasswd Command

The smbpasswd utility is a utility similar to the passwd or yppasswd programs. It maintains the two 32 byte password fields in the passdb backend.

smbpasswd works in a client-server mode where it contacts the local smbd to change the user's password on its behalf. This has enormous benefits - as follows.

smbpasswd has the capability to change passwords on Windows NT servers (this only works when the request is sent to the NT Primary Domain Controller if you are changing an NT Domain user's password).

To run smbpasswd as a normal user just type :

$ smbpasswd

Old SMB password: <type old value here - or hit return if there was no old password>

New SMB Password: <type new value>

Repeat New SMB Password: <re-type new value

If the old value does not match the current value stored for that user, or the two new values do not match each other, then the password will not be changed.

If invoked by an ordinary user it will only allow the user to change his or her own Samba password.

If run by the root user smbpasswd may take an optional argument, specifying the user name whose SMB password you wish to change. Note that when run as root smbpasswd does not prompt for or check the old password value, thus allowing root to set passwords for users who have forgotten their passwords.

smbpasswd is designed to work in the same way and be familiar to UNIX users who use the passwd or yppasswd commands.

For more details on using smbpasswd refer to the man page which will always be the definitive reference.

3.4. Plain text

Older versions of samba retrieved user information from the unix user database and eventually some other fields from the file /etc/samba/smbpasswd or /etc/smbpasswd. When password encryption is disabled, no data is stored at all.

3.5. TDB

Samba can also store the user data in a "TDB" (Trivial Database). Using this backend doesn't require any additional configuration. This backend is recommended for new installations who don't require LDAP.

3.6. LDAP

objectclass ( 1.3.1.5.1.4.1.7165.2.2.2 NAME 'sambaAccount' SUP top STRUCTURAL
     DESC 'Samba Account'
     MUST ( uid $ rid )
     MAY  ( cn $ lmPassword $ ntPassword $ pwdLastSet $ logonTime $
            logoffTime $ kickoffTime $ pwdCanChange $ pwdMustChange $ acctFlags $
            displayName $ smbHome $ homeDrive $ scriptPath $ profilePath $
            description $ userWorkstations $ primaryGroupID $ domain ))

## /etc/openldap/slapd.conf

## schema files (core.schema is required by default)
include	           /etc/openldap/schema/core.schema

## needed for sambaAccount
include            /etc/openldap/schema/cosine.schema
include            /etc/openldap/schema/inetorgperson.schema
include            /etc/openldap/schema/samba.schema

## uncomment this line if you want to support the RFC2307 (NIS) schema
## include         /etc/openldap/schema/nis.schema

....

# Indices to maintain
## required by OpenLDAP 2.0
index objectclass   eq

## support pb_getsampwnam()
index uid           pres,eq
## support pdb_getsambapwrid()
index rid           eq

## uncomment these if you are storing posixAccount and
## posixGroup entries in the directory as well
##index uidNumber     eq
##index gidNumber     eq
##index cn            eq
##index memberUid     eq

## /usr/local/samba/lib/smb.conf
[global]
     security = user
     encrypt passwords = yes

     netbios name = TASHTEGO
     workgroup = NARNIA

     # ldap related parameters

     # define the DN to use when binding to the directory servers
     # The password for this DN is not stored in smb.conf.  Rather it
     # must be set by using 'smbpasswd -w secretpw' to store the
     # passphrase in the secrets.tdb file.  If the "ldap admin dn" values
     # changes, this password will need to be reset.
     ldap admin dn = "cn=Samba Manager,ou=people,dc=samba,dc=org"

     #  specify the LDAP server's hostname (defaults to locahost)
     ldap server = ahab.samba.org

     # Define the SSL option when connecting to the directory
     # ('off', 'start tls', or 'on' (default))
     ldap ssl = start tls

     # define the port to use in the LDAP session (defaults to 636 when
     # "ldap ssl = on")
     ldap port = 389

     # specify the base DN to use when searching the directory
     ldap suffix = "ou=people,dc=samba,dc=org"

     # generally the default ldap search filter is ok
     # ldap filter = "(&(uid=%u)(objectclass=sambaAccount))"

## allow the "ldap admin dn" access, but deny everyone else
access to attrs=lmPassword,ntPassword
     by dn="cn=Samba Admin,ou=people,dc=plainjoe,dc=org" write
     by * none

dn: uid=guest2, ou=people,dc=plainjoe,dc=org
ntPassword: 878D8014606CDA29677A44EFA1353FC7
pwdMustChange: 2147483647
primaryGroupID: 1201
lmPassword: 552902031BEDE9EFAAD3B435B51404EE
pwdLastSet: 1010179124
logonTime: 0
objectClass: sambaAccount
uid: guest2
kickoffTime: 2147483647
acctFlags: [UX         ]
logoffTime: 2147483647
rid: 19006
pwdCanChange: 0

dn: uid=gcarter, ou=people,dc=plainjoe,dc=org
logonTime: 0
displayName: Gerald Carter
lmPassword: 552902031BEDE9EFAAD3B435B51404EE
primaryGroupID: 1201
objectClass: posixAccount
objectClass: sambaAccount
acctFlags: [UX         ]
userPassword: {crypt}BpM2ej8Rkzogo
uid: gcarter
uidNumber: 9000
cn: Gerald Carter
loginShell: /bin/bash
logoffTime: 2147483647
gidNumber: 100
kickoffTime: 2147483647
pwdLastSet: 1010179230
rid: 19000
homeDirectory: /home/tashtego/gcarter
pwdCanChange: 0
pwdMustChange: 2147483647
ntPassword: 878D8014606CDA29677A44EFA1353FC7

3.7. MySQL

---

3.7.2. Configuring

This plugin lacks some good documentation, but here is some short info:

Add a the following to the passdb backend variable in your smb.conf:

passdb backend = [other-plugins] mysql:identifier [other-plugins]

The identifier can be any string you like, as long as it doesn't collide with the identifiers of other plugins or other instances of pdb_mysql. If you specify multiple pdb_mysql.so entries in 'passdb backend', you also need to use different identifiers!

Additional options can be given thru the smb.conf file in the [global] section.

identifier:mysql host                     - host name, defaults to 'localhost'
identifier:mysql password
identifier:mysql user                     - defaults to 'samba'
identifier:mysql database                 - defaults to 'samba'
identifier:mysql port                     - defaults to 3306
identifier:table                          - Name of the table containing users

Since the password for the mysql user is stored in the smb.conf file, you should make the the smb.conf file readable only to the user that runs samba. This is considered a security bug and will be fixed soon.

Names of the columns in this table(I've added column types those columns should have first):

identifier:logon time column             - int(9)
identifier:logoff time column            - int(9)
identifier:kickoff time column           - int(9)
identifier:pass last set time column     - int(9)
identifier:pass can change time column   - int(9)
identifier:pass must change time column  - int(9)
identifier:username column               - varchar(255) - unix username
identifier:domain column                 - varchar(255) - NT domain user is part of
identifier:nt username column            - varchar(255) - NT username
identifier:fullname column            - varchar(255) - Full name of user
identifier:home dir column               - varchar(255) - Unix homedir path
identifier:dir drive column              - varchar(2) - Directory drive path (eg: 'H:')
identifier:logon script column           - varchar(255) - Batch file to run on client side when logging on
identifier:profile path column           - varchar(255) - Path of profile
identifier:acct desc column              - varchar(255) - Some ASCII NT user data
identifier:workstations column           - varchar(255) - Workstations user can logon to (or NULL for all)
identifier:unknown string column         - varchar(255) - unknown string
identifier:munged dial column            - varchar(255) - ?
identifier:uid column                    - int(9) - Unix user ID (uid)
identifier:gid column                    - int(9) - Unix user group (gid)
identifier:user sid column               - varchar(255) - NT user SID
identifier:group sid column              - varchar(255) - NT group ID
identifier:lanman pass column            - varchar(255) - encrypted lanman password
identifier:nt pass column                - varchar(255) - encrypted nt passwd
identifier:plain pass column             - varchar(255) - plaintext password
identifier:acct control column           - int(9) - nt user data
identifier:unknown 3 column              - int(9) - unknown
identifier:logon divs column             - int(9) - ?
identifier:hours len column              - int(9) - ?
identifier:unknown 5 column              - int(9) - unknown
identifier:unknown 6 column              - int(9) - unknown

Eventually, you can put a colon (:) after the name of each column, which should specify the column to update when updating the table. You can also specify nothing behind the colon - then the data from the field will not be updated.

3.8. XML

This module requires libxml2 to be installed.

The usage of pdb_xml is pretty straightforward. To export data, use: pdbedit -e xml:filename (where filename is the name of the file to put the data in)

To import data, use: pdbedit -i xml:filename -e current-pdb Where filename is the name to read the data from and current-pdb to put it in.

4.1. Stand Alone Server

The term stand alone server means that the server will provide local authentication and access control for all resources that are available from it. In general this means that there will be a local user database. In more technical terms, it means that resources on the machine will either be made available in either SHARE mode or in USER mode. SHARE mode and USER mode security are documented under discussions regarding "security mode". The smb.conf configuration parameters that control security mode are: "security = user" and "security = share".

No special action is needed other than to create user accounts. Stand-alone servers do NOT provide network logon services, meaning that machines that use this server do NOT perform a domain logon but instead make use only of the MS Windows logon which is local to the MS Windows workstation/server.

Samba tends to blur the distinction a little in respect of what is a stand alone server. This is because the authentication database may be local or on a remote server, even if from the samba protocol perspective the samba server is NOT a member of a domain security context.

Through the use of PAM (Pluggable Authentication Modules) and nsswitch (the name service switcher) the source of authentication may reside on another server. We would be inclined to call this the authentication server. This means that the samba server may use the local Unix/Linux system password database (/etc/passwd or /etc/shadow), may use a local smbpasswd file (/etc/samba/smbpasswd or /usr/local/samba/lib/private/smbpasswd), or may use an LDAP back end, or even via PAM and Winbind another CIFS/SMB server for authentication.

4.2. Domain Member Server

This mode of server operation involves the samba machine being made a member of a domain security context. This means by definition that all user authentication will be done from a centrally defined authentication regime. The authentication regime may come from an NT3/4 style (old domain technology) server, or it may be provided from an Active Directory server (ADS) running on MS Windows 2000 or later.

Of course it should be clear that the authentication back end itself could be from any distributed directory architecture server that is supported by Samba. This can be LDAP (from OpenLDAP), or Sun's iPlanet, of NetWare Directory Server, etc.

Please refer to the section on Howto configure Samba as a Primary Domain Controller and for more information regarding how to create a domain machine account for a domain member server as well as for information regading how to enable the samba domain member machine to join the domain and to be fully trusted by it.

4.3. Domain Controller

Over the years public perceptions of what Domain Control really is has taken on an almost mystical nature. Before we branch into a brief overview of what Domain Control is the following types of controller are known:

5.1. User and Share security level

A SMB server tells the client at startup what "security level" it is running. There are two options "share level" and "user level". Which of these two the client receives affects the way the client then tries to authenticate itself. It does not directly affect (to any great extent) the way the Samba server does security. I know this is strange, but it fits in with the client/server approach of SMB. In SMB everything is initiated and controlled by the client, and the server can only tell the client what is available and whether an action is allowed.

	passsword level = integer
	username level = integer

	encrypt passwords = Yes
	security = server
	password server = "NetBIOS_name_of_PDC"

	encrypt passwords = Yes
	security = domain
	workgroup = "name of NT domain"
	password server = *

6.1. Prerequisite Reading

Before you continue reading in this chapter, please make sure that you are comfortable with configuring basic files services in smb.conf and how to enable and administer password encryption in Samba. Theses two topics are covered in the smb.conf(5) manpage.

6.2. Background

This article outlines the steps necessary for configuring Samba as a PDC. It is necessary to have a working Samba server prior to implementing the PDC functionality.

• Domain logons for Windows NT 4.0 / 200x / XP Professional clients.

• Placing Windows 9x / Me clients in user level security

• Retrieving a list of users and groups from a Samba PDC to Windows 9x / Me / NT / 200x / XP Professional clients

• Roaming Profiles

• Network/System Policies

Roaming Profiles and System/Network policies are advanced network administration topics that are covered separately in this document.

The following functionalities are new to the Samba 3.0 release:

• Windows NT 4 domain trusts

• Adding users via the User Manager for Domains

The following functionalities are NOT provided by Samba 3.0:

• SAM replication with Windows NT 4.0 Domain Controllers (i.e. a Samba PDC and a Windows NT BDC or vice versa)

• Acting as a Windows 2000 Domain Controller (i.e. Kerberos and Active Directory)

Please note that Windows 9x / Me / XP Home clients are not true members of a domain for reasons outlined in this article. Therefore the protocol for support Windows 9x-style domain logons is completely different from NT4 / Win2k type domain logons and has been officially supported for some time.

MS Windows XP Home edition is NOT able to join a domain and does not permit the use of domain logons.

Implementing a Samba PDC can basically be divided into 3 broad steps.

1. Configuring the Samba PDC

2. Creating machine trust accounts and joining clients to the domain

3. Adding and managing domain user accounts

There are other minor details such as user profiles, system policies, etc... However, these are not necessarily specific to a Samba PDC as much as they are related to Windows NT networking concepts.

6.3. Configuring the Samba Domain Controller

The first step in creating a working Samba PDC is to understand the parameters necessary in smb.conf. Here we attempt to explain the parameters that are covered in the smb.conf man page.

Here is an example smb.conf for acting as a PDC:

[global]
    ; Basic server settings
    netbios name = POGO
    workgroup = NARNIA

    ; we should act as the domain and local master browser
    os level = 64
    preferred master = yes
    domain master = yes
    local master = yes
    
    ; security settings (must user security = user)
    security = user
    
    ; encrypted passwords are a requirement for a PDC
    encrypt passwords = yes
    
    ; support domain logons
    domain logons = yes
    
    ; where to store user profiles?
    logon path = \\%N\profiles\%u
    
    ; where is a user's home directory and where should it be mounted at?
    logon drive = H:
    logon home = \\homeserver\%u
    
    ; specify a generic logon script for all users
    ; this is a relative **DOS** path to the [netlogon] share
    logon script = logon.cmd

; necessary share for domain controller
[netlogon]
    path = /usr/local/samba/lib/netlogon
    read only = yes
    write list = ntadmin
    
; share for storing user profiles
[profiles]
    path = /export/smb/ntprofile
    read only = no
    create mask = 0600
    directory mask = 0700

There are a couple of points to emphasize in the above configuration.

• Encrypted passwords must be enabled. For more details on how to do this, refer to ENCRYPTION.html.

• The server must support domain logons and a [netlogon] share

• The server must be the domain master browser in order for Windows client to locate the server as a DC. Please refer to the various Network Browsing documentation included with this distribution for details.

Samba 3.0 offers a complete implementation of group mapping between Windows NT groups and Unix groups (this is really quite complicated to explain in a short space).

6.4. Creating Machine Trust Accounts and Joining Clients to the Domain

A machine trust account is a Samba account that is used to authenticate a client machine (rather than a user) to the Samba server. In Windows terminology, this is known as a "Computer Account."

The password of a machine trust account acts as the shared secret for secure communication with the Domain Controller. This is a security feature to prevent an unauthorized machine with the same NetBIOS name from joining the domain and gaining access to domain user/group accounts. Windows NT, 200x, XP Professional clients use machine trust accounts, but Windows 9x / Me / XP Home clients do not. Hence, a Windows 9x / Me / XP Home client is never a true member of a domain because it does not possess a machine trust account, and thus has no shared secret with the domain controller.

A Windows PDC stores each machine trust account in the Windows Registry. A Samba-3 PDC also has to stoe machine trust account information in a suitable back-end data store. With Samba-3 there can be multiple back-ends for this including:

• smbpaswd - the plain ascii file stored used by earlier versions of Samba. This file configuration option requires a Unix/Linux system account for EVERY entry (ie: both for user and for machine accounts). This file will be located in the private directory (default is /usr/local/samba/lib/private or on linux /etc/samba).

• smbpasswd_nua - This file is independant of the system wide user accounts. The use of this back-end option requires specification of the "non unix account range" option also. It is called smbpasswd and will be located in the private directory.

• tdbsam - a binary database backend that will be stored in the private directory in a file called passwd.tdb. The key benefit of this binary format file is that it can store binary objects that can not be accomodated in the traditional plain text smbpasswd file.

• tdbsam_nua like the smbpasswd_nua option above, this file allows the creation of arbitrary user and machine accounts without requiring that account to be added to the system (/etc/passwd) file. It too requires the specification of the "non unix account range" option in the [globals] section of the smb.conf file.

• ldapsam - An LDAP based back-end. Permits the LDAP server to be specified. eg: ldap://localhost or ldap://frodo.murphy.com

• ldapsam_nua - LDAP based back-end with no unix account requirement, like smbpasswd_nua and tdbsam_nua above.

A Samba PDC, however, stores each machine trust account in two parts, as follows:

• A Samba account, stored in the same location as user LanMan and NT password hashes (currently smbpasswd). The Samba account possesses and uses only the NT password hash.

• A corresponding Unix account, typically stored in /etc/passwd. (Future releases will alleviate the need to create /etc/passwd entries.)

There are two ways to create machine trust accounts:

• Manual creation. Both the Samba and corresponding Unix account are created by hand.

• "On-the-fly" creation. The Samba machine trust account is automatically created by Samba at the time the client is joined to the domain. (For security, this is the recommended method.) The corresponding Unix account may be created automatically or manually.

---

6.4.1. Manual Creation of Machine Trust Accounts

The first step in manually creating a machine trust account is to manually create the corresponding Unix account in /etc/passwd. This can be done using vipw or other 'add user' command that is normally used to create new Unix accounts. The following is an example for a Linux based Samba server:

root# /usr/sbin/useradd -g 100 -d /dev/null -c "machine nickname" -s /bin/false machine_name$

root# passwd -l machine_name$

On *BSD systems, this can be done using the 'chpass' utility:

root# chpass -a "machine_name$:*:101:100::0:0:Workstation machine_name:/dev/null:/sbin/nologin"

The /etc/passwd entry will list the machine name with a "$" appended, won't have a password, will have a null shell and no home directory. For example a machine named 'doppy' would have an /etc/passwd entry like this:

doppy$:x:505:501:machine_nickname:/dev/null:/bin/false

Above, machine_nickname can be any descriptive name for the client, i.e., BasementComputer. machine_name absolutely must be the NetBIOS name of the client to be joined to the domain. The "$" must be appended to the NetBIOS name of the client or Samba will not recognize this as a machine trust account.

Now that the corresponding Unix account has been created, the next step is to create the Samba account for the client containing the well-known initial machine trust account password. This can be done using the smbpasswd(8) command as shown here:

root# smbpasswd -a -m machine_name

where machine_name is the machine's NetBIOS name. The RID of the new machine account is generated from the UID of the corresponding Unix account.

Join the client to the domain immediately

Manually creating a machine trust account using this method is the equivalent of creating a machine trust account on a Windows NT PDC using the "Server Manager". From the time at which the account is created to the time which the client joins the domain and changes the password, your domain is vulnerable to an intruder joining your domain using a a machine with the same NetBIOS name. A PDC inherently trusts members of the domain and will serve out a large degree of user information to such clients. You have been warned!

[global]
   # <...remainder of parameters...>
   add user script = /usr/sbin/useradd -d /dev/null -g 100 -s /bin/false -M %u 

6.5. Common Problems and Errors

• I cannot include a '$' in a machine name.

  A 'machine name' in (typically) /etc/passwd of the machine name with a '$' appended. FreeBSD (and other BSD systems?) won't create a user with a '$' in their name.

  The problem is only in the program used to make the entry, once made, it works perfectly. So create a user without the '$' and use vipw to edit the entry, adding the '$'. Or create the whole entry with vipw if you like, make sure you use a unique User ID !

• I get told "You already have a connection to the Domain...." or "Cannot join domain, the credentials supplied conflict with an existing set.." when creating a machine trust account.

  This happens if you try to create a machine trust account from the machine itself and already have a connection (e.g. mapped drive) to a share (or IPC$) on the Samba PDC. The following command will remove all network drive connections:

  C:\WINNT\> net use * /d

  Further, if the machine is a already a 'member of a workgroup' that is the same name as the domain you are joining (bad idea) you will get this message. Change the workgroup name to something else, it does not matter what, reboot, and try again.

• The system can not log you on (C000019B)....

  I joined the domain successfully but after upgrading to a newer version of the Samba code I get the message, "The system can not log you on (C000019B), Please try again or consult your system administrator" when attempting to logon.

  This occurs when the domain SID stored in the secrets.tdb database is changed. The most common cause of a change in domain SID is when the domain name and/or the server name (netbios name) is changed. The only way to correct the problem is to restore the original domain SID or remove the domain client from the domain and rejoin. The domain SID may be reset using either the smbpasswd or rpcclient utilities.

• The machine trust account for this computer either does not exist or is not accessible.

  When I try to join the domain I get the message "The machine account for this computer either does not exist or is not accessible". What's wrong?

  This problem is caused by the PDC not having a suitable machine trust account. If you are using the add user script method to create accounts then this would indicate that it has not worked. Ensure the domain admin user system is working.

  Alternatively if you are creating account entries manually then they have not been created correctly. Make sure that you have the entry correct for the machine trust account in smbpasswd file on the Samba PDC. If you added the account using an editor rather than using the smbpasswd utility, make sure that the account name is the machine NetBIOS name with a '$' appended to it ( i.e. computer_name$ ). There must be an entry in both /etc/passwd and the smbpasswd file. Some people have reported that inconsistent subnet masks between the Samba server and the NT client have caused this problem. Make sure that these are consistent for both client and server.

• When I attempt to login to a Samba Domain from a NT4/W2K workstation, I get a message about my account being disabled.

  This problem is caused by a PAM related bug in Samba 2.2.0. This bug is fixed in 2.2.1. Other symptoms could be unaccessible shares on NT/W2K member servers in the domain or the following error in your smbd.log: passdb/pampass.c:pam_account(268) PAM: UNKNOWN ERROR for User: %user%

  At first be ensure to enable the useraccounts with smbpasswd -e %user%, this is normally done, when you create an account.

  In order to work around this problem in 2.2.0, configure the account control flag in /etc/pam.d/samba file as follows:

  	account required        pam_permit.so
  	

  If you want to remain backward compatibility to samba 2.0.x use pam_permit.so, it's also possible to use pam_pwdb.so. There are some bugs if you try to use pam_unix.so, if you need this, be ensure to use the most recent version of this file.

6.6. What other help can I get?

There are many sources of information available in the form of mailing lists, RFC's and documentation. The docs that come with the samba distribution contain very good explanations of general SMB topics such as browsing.

• What are some diagnostics tools I can use to debug the domain logon process and where can I find them?

  One of the best diagnostic tools for debugging problems is Samba itself. You can use the -d option for both smbd and nmbd to specify what 'debug level' at which to run. See the man pages on smbd, nmbd and smb.conf for more information on debugging options. The debug level can range from 1 (the default) to 10 (100 for debugging passwords).

  Another helpful method of debugging is to compile samba using the gcc -g flag. This will include debug information in the binaries and allow you to attach gdb to the running smbd / nmbd process. In order to attach gdb to an smbd process for an NT workstation, first get the workstation to make the connection. Pressing ctrl-alt-delete and going down to the domain box is sufficient (at least, on the first time you join the domain) to generate a 'LsaEnumTrustedDomains'. Thereafter, the workstation maintains an open connection, and therefore there will be an smbd process running (assuming that you haven't set a really short smbd idle timeout) So, in between pressing ctrl alt delete, and actually typing in your password, you can gdb attach and continue.

  Some useful samba commands worth investigating:

  • testparam | more

  • smbclient -L //{netbios name of server}

  An SMB enabled version of tcpdump is available from http://www.tcpdup.org/. Ethereal, another good packet sniffer for Unix and Win32 hosts, can be downloaded from http://www.ethereal.com.

  For tracing things on the Microsoft Windows NT, Network Monitor (aka. netmon) is available on the Microsoft Developer Network CD's, the Windows NT Server install CD and the SMS CD's. The version of netmon that ships with SMS allows for dumping packets between any two computers (i.e. placing the network interface in promiscuous mode). The version on the NT Server install CD will only allow monitoring of network traffic directed to the local NT box and broadcasts on the local subnet. Be aware that Ethereal can read and write netmon formatted files.

• How do I install 'Network Monitor' on an NT Workstation or a Windows 9x box?

  Installing netmon on an NT workstation requires a couple of steps. The following are for installing Netmon V4.00.349, which comes with Microsoft Windows NT Server 4.0, on Microsoft Windows NT Workstation 4.0. The process should be similar for other version of Windows NT / Netmon. You will need both the Microsoft Windows NT Server 4.0 Install CD and the Workstation 4.0 Install CD.

  Initially you will need to install 'Network Monitor Tools and Agent' on the NT Server. To do this

  • Goto Start - Settings - Control Panel - Network - Services - Add

  • Select the 'Network Monitor Tools and Agent' and click on 'OK'.

  • Click 'OK' on the Network Control Panel.

  • Insert the Windows NT Server 4.0 install CD when prompted.

  At this point the Netmon files should exist in %SYSTEMROOT%\System32\netmon\*.*. Two subdirectories exist as well, parsers\ which contains the necessary DLL's for parsing the netmon packet dump, and captures\.

  In order to install the Netmon tools on an NT Workstation, you will first need to install the 'Network Monitor Agent' from the Workstation install CD.

  • Goto Start - Settings - Control Panel - Network - Services - Add

  • Select the 'Network Monitor Agent' and click on 'OK'.

  • Click 'OK' on the Network Control Panel.

  • Insert the Windows NT Workstation 4.0 install CD when prompted.

  Now copy the files from the NT Server in %SYSTEMROOT%\System32\netmon\*.* to %SYSTEMROOT%\System32\netmon\*.* on the Workstation and set permissions as you deem appropriate for your site. You will need administrative rights on the NT box to run netmon.

  To install Netmon on a Windows 9x box install the network monitor agent from the Windows 9x CD (\admin\nettools\netmon). There is a readme file located with the netmon driver files on the CD if you need information on how to do this. Copy the files from a working Netmon installation.

• The following is a list if helpful URLs and other links:

  • Home of Samba site http://samba.org. We have a mirror near you !

  • The Development document on the Samba mirrors might mention your problem. If so, it might mean that the developers are working on it.

  • See how Scott Merrill simulates a BDC behavior at http://www.skippy.net/linux/smb-howto.html.

  • Although 2.0.7 has almost had its day as a PDC, David Bannon will keep the 2.0.7 PDC pages at http://bioserve.latrobe.edu.au/samba going for a while yet.

  • Misc links to CIFS information http://samba.org/cifs/

  • NT Domains for Unix http://mailhost.cb1.com/~lkcl/ntdom/

  • FTP site for older SMB specs: ftp://ftp.microsoft.com/developr/drg/CIFS/

• How do I get help from the mailing lists?

  There are a number of Samba related mailing lists. Go to http://samba.org, click on your nearest mirror and then click on Support and then click on Samba related mailing lists.

  For questions relating to Samba TNG go to http://www.samba-tng.org/ It has been requested that you don't post questions about Samba-TNG to the main stream Samba lists.

  If you post a message to one of the lists please observe the following guide lines :

  • Always remember that the developers are volunteers, they are not paid and they never guarantee to produce a particular feature at a particular time. Any time lines are 'best guess' and nothing more.

  • Always mention what version of samba you are using and what operating system its running under. You should probably list the relevant sections of your smb.conf file, at least the options in [global] that affect PDC support.

  • In addition to the version, if you obtained Samba via CVS mention the date when you last checked it out.

  • Try and make your question clear and brief, lots of long, convoluted questions get deleted before they are completely read ! Don't post html encoded messages (if you can select colour or font size its html).

  • If you run one of those nifty 'I'm on holidays' things when you are away, make sure its configured to not answer mailing lists.

  • Don't cross post. Work out which is the best list to post to and see what happens, i.e. don't post to both samba-ntdom and samba-technical. Many people active on the lists subscribe to more than one list and get annoyed to see the same message two or more times. Often someone will see a message and thinking it would be better dealt with on another, will forward it on for you.

  • You might include partial log files written at a debug level set to as much as 20. Please don't send the entire log but enough to give the context of the error messages.

  • (Possibly) If you have a complete netmon trace ( from the opening of the pipe to the error ) you can send the *.CAP file as well.

  • Please think carefully before attaching a document to an email. Consider pasting the relevant parts into the body of the message. The samba mailing lists go to a huge number of people, do they all need a copy of your smb.conf in their attach directory?

• How do I get off the mailing lists?

  To have your name removed from a samba mailing list, go to the same place you went to to get on it. Go to http://lists.samba.org, click on your nearest mirror and then click on Support and then click on Samba related mailing lists. Or perhaps see here

  Please don't post messages to the list asking to be removed, you will just be referred to the above address (unless that process failed in some way...)

6.7. Domain Control for Windows 9x/ME

A domain and a workgroup are exactly the same thing in terms of network browsing. The difference is that a distributable authentication database is associated with a domain, for secure login access to a network. Also, different access rights can be granted to users if they successfully authenticate against a domain logon server. Samba-3 does this now in the same way that MS Windows NT/2K.

The SMB client logging on to a domain has an expectation that every other server in the domain should accept the same authentication information. Network browsing functionality of domains and workgroups is identical and is explained in this documentation under the browsing discussions. It should be noted, that browsing is totally orthogonal to logon support.

Issues related to the single-logon network model are discussed in this section. Samba supports domain logons, network logon scripts, and user profiles for MS Windows for workgroups and MS Windows 9X/ME clients which are the focus of this section.

When an SMB client in a domain wishes to logon it broadcast requests for a logon server. The first one to reply gets the job, and validates its password using whatever mechanism the Samba administrator has installed. It is possible (but very stupid) to create a domain where the user database is not shared between servers, i.e. they are effectively workgroup servers advertising themselves as participating in a domain. This demonstrates how authentication is quite different from but closely involved with domains.

Using these features you can make your clients verify their logon via the Samba server; make clients run a batch file when they logon to the network and download their preferences, desktop and start menu.

Before launching into the configuration instructions, it is worthwhile lookingat how a Windows 9x/ME client performs a logon:

1. The client broadcasts (to the IP broadcast address of the subnet it is in) a NetLogon request. This is sent to the NetBIOS name DOMAIN<1c> at the NetBIOS layer. The client chooses the first response it receives, which contains the NetBIOS name of the logon server to use in the format of \\SERVER.

2. The client then connects to that server, logs on (does an SMBsessetupX) and then connects to the IPC$ share (using an SMBtconX).

3. The client then does a NetWkstaUserLogon request, which retrieves the name of the user's logon script.

4. The client then connects to the NetLogon share and searches for this and if it is found and can be read, is retrieved and executed by the client. After this, the client disconnects from the NetLogon share.

5. The client then sends a NetUserGetInfo request to the server, to retrieve the user's home share, which is used to search for profiles. Since the response to the NetUserGetInfo request does not contain much more the user's home share, profiles for Win9X clients MUST reside in the user home directory.

6. The client then connects to the user's home share and searches for the user's profile. As it turns out, you can specify the user's home share as a sharename and path. For example, \\server\fred\.profile. If the profiles are found, they are implemented.

7. The client then disconnects from the user's home share, and reconnects to the NetLogon share and looks for CONFIG.POL, the policies file. If this is found, it is read and implemented.

---

6.7.1. Configuration Instructions: Network Logons

The main difference between a PDC and a Windows 9x logon server configuration is that

• Password encryption is not required for a Windows 9x logon server.

• Windows 9x/ME clients do not possess machine trust accounts.

Therefore, a Samba PDC will also act as a Windows 9x logon server.

security mode and master browsers

There are a few comments to make in order to tie up some loose ends. There has been much debate over the issue of whether or not it is ok to configure Samba as a Domain Controller in security modes other than USER. The only security mode which will not work due to technical reasons is SHARE mode security. DOMAIN and SERVER mode security is really just a variation on SMB user level security. Actually, this issue is also closely tied to the debate on whether or not Samba must be the domain master browser for its workgroup when operating as a DC. While it may technically be possible to configure a server as such (after all, browsing and domain logons are two distinctly different functions), it is not a good idea to so. You should remember that the DC must register the DOMAIN#1b NetBIOS name. This is the name used by Windows clients to locate the DC. Windows clients do not distinguish between the DC and the DMB. For this reason, it is very wise to configure the Samba DC as the DMB. Now back to the issue of configuring a Samba DC to use a mode other than "security = user". If a Samba host is configured to use another SMB server or DC in order to validate user connection requests, then it is a fact that some other machine on the network (the "password server") knows more about user than the Samba host. 99% of the time, this other host is a domain controller. Now in order to operate in domain mode security, the "workgroup" parameter must be set to the name of the Windows NT domain (which already has a domain controller, right?) Therefore configuring a Samba box as a DC for a domain that already by definition has a PDC is asking for trouble. Therefore, you should always configure the Samba DC to be the DMB for its domain.

7.1. Prerequisite Reading

Before you continue reading in this chapter, please make sure that you are comfortable with configuring a Samba PDC as described in the Samba-PDC-HOWTO.

7.2. Background

What is a Domain Controller? It is a machine that is able to answer logon requests from workstations in a Windows NT Domain. Whenever a user logs into a Windows NT Workstation, the workstation connects to a Domain Controller and asks him whether the username and password the user typed in is correct. The Domain Controller replies with a lot of information about the user, for example the place where the users profile is stored, the users full name of the user. All this information is stored in the NT user database, the so-called SAM.

There are two kinds of Domain Controller in a NT 4 compatible Domain: A Primary Domain Controller (PDC) and one or more Backup Domain Controllers (BDC). The PDC contains the master copy of the SAM. Whenever the SAM has to change, for example when a user changes his password, this change has to be done on the PDC. A Backup Domain Controller is a machine that maintains a read-only copy of the SAM. This way it is able to reply to logon requests and authenticate users in case the PDC is not available. During this time no changes to the SAM are possible. Whenever changes to the SAM are done on the PDC, all BDC receive the changes from the PDC.

Since version 2.2 Samba officially supports domain logons for all current Windows Clients, including Windows 2000 and XP. This text assumes the domain to be named SAMBA. To be able to act as a PDC, some parameters in the [global]-section of the smb.conf have to be set:

workgroup = SAMBA
domain master = yes
domain logons = yes

Several other things like a [homes] and a [netlogon] share also may be set along with settings for the profile path, the users home drive and others. This will not be covered in this document.

7.3. What qualifies a Domain Controller on the network?

Every machine that is a Domain Controller for the domain SAMBA has to register the NetBIOS group name SAMBA#1c with the WINS server and/or by broadcast on the local network. The PDC also registers the unique NetBIOS name SAMBA#1b with the WINS server. The name type #1b is normally reserved for the domain master browser, a role that has nothing to do with anything related to authentication, but the Microsoft Domain implementation requires the domain master browser to be on the same machine as the PDC.

7.4. Can Samba be a Backup Domain Controller to an NT PDC?

With version 2.2, no. The native NT SAM replication protocols have not yet been fully implemented. The Samba Team is working on understanding and implementing the protocols, but this work has not been finished for version 2.2.

With version 3.0, the work on both the replication protocols and a suitable storage mechanism has progressed, and some form of NT4 BDC support is expected soon.

Can I get the benefits of a BDC with Samba? Yes. The main reason for implementing a BDC is availability. If the PDC is a Samba machine, a second Samba machine can be set up to service logon requests whenever the PDC is down.

7.5. How do I set up a Samba BDC?

Several things have to be done:

• The domain SID has to be the same on the PDC and the BDC. This used to be stored in the file private/MACHINE.SID. This file is not created anymore since Samba 2.2.5 or even earlier. Nowadays the domain SID is stored in the file private/secrets.tdb. Simply copying the secrets.tdb from the PDC to the BDC does not work, as the BDC would generate a new SID for itself and override the domain SID with this new BDC SID.

  To retrieve the domain SID from the PDC or an existing BDC and store it in the secrets.tdb, execute 'net rpc getsid' on the BDC.

• The Unix user database has to be synchronized from the PDC to the BDC. This means that both the /etc/passwd and /etc/group have to be replicated from the PDC to the BDC. This can be done manually whenever changes are made, or the PDC is set up as a NIS master server and the BDC as a NIS slave server. To set up the BDC as a mere NIS client would not be enough, as the BDC would not be able to access its user database in case of a PDC failure.

• The Samba password database in the file private/smbpasswd has to be replicated from the PDC to the BDC. This is a bit tricky, see the next section.

• Any netlogon share has to be replicated from the PDC to the BDC. This can be done manually whenever login scripts are changed, or it can be done automatically together with the smbpasswd synchronization.

Finally, the BDC has to be found by the workstations. This can be done by setting

workgroup = samba
domain master = no
domain logons = yes

in the [global]-section of the smb.conf of the BDC. This makes the BDC only register the name SAMBA#1c with the WINS server. This is no problem as the name SAMBA#1c is a NetBIOS group name that is meant to be registered by more than one machine. The parameter 'domain master = no' forces the BDC not to register SAMBA#1b which as a unique NetBIOS name is reserved for the Primary Domain Controller.

8.1. Setup your smb.conf

You must use at least the following 3 options in smb.conf:

  realm = YOUR.KERBEROS.REALM
  security = ADS
  encrypt passwords = yes

In case samba can't figure out your ads server using your realm name, use the ads server option in smb.conf:

  ads server = your.kerberos.server

You do *not* need a smbpasswd file, and older clients will be authenticated as if security = domain, although it won't do any harm and allows you to have local users not in the domain. I expect that the above required options will change soon when we get better active directory integration.

8.2. Setup your /etc/krb5.conf

The minimal configuration for krb5.conf is:

[realms]
    YOUR.KERBEROS.REALM = {
	kdc = your.kerberos.server
    }

Test your config by doing a kinit USERNAME@REALM and making sure that your password is accepted by the Win2000 KDC.

The realm must be uppercase.

You also must ensure that you can do a reverse DNS lookup on the IP address of your KDC. Also, the name that this reverse lookup maps to must either be the netbios name of the KDC (ie. the hostname with no domain attached) or it can alternatively be the netbios name followed by the realm.

The easiest way to ensure you get this right is to add a /etc/hosts entry mapping the IP address of your KDC to its netbios name. If you don't get this right then you will get a "local error" when you try to join the realm.

If all you want is kerberos support in smbclient then you can skip straight to Test with smbclient now. Creating a computer account and testing your servers is only needed if you want kerberos support for smbd and winbindd.

8.3. Create the computer account

As a user that has write permission on the Samba private directory (usually root) run: net ads join

8.4. Test your server setup

On a Windows 2000 client try net use * \\server\share. You should be logged in with kerberos without needing to know a password. If this fails then run klist tickets. Did you get a ticket for the server? Does it have an encoding type of DES-CBC-MD5 ?

8.5. Testing with smbclient

On your Samba server try to login to a Win2000 server or your Samba server using smbclient and kerberos. Use smbclient as usual, but specify the -k option to choose kerberos authentication.

8.6. Notes

You must change administrator password at least once after DC install, to create the right encoding types

w2k doesn't seem to create the _kerberos._udp and _ldap._tcp in their defaults DNS setup. Maybe fixed in service packs?

9.1. Joining an NT Domain with Samba 3.0

Assume you have a Samba 3.0 server with a NetBIOS name of SERV1 and are joining an or Win2k NT domain called DOM, which has a PDC with a NetBIOS name of DOMPDC and two backup domain controllers with NetBIOS names DOMBDC1 and DOMBDC2 .

Firstly, you must edit your smb.conf(5) file to tell Samba it should now use domain security.

Change (or add) your security = line in the [global] section of your smb.conf to read:

security = domain

Next change the workgroup = line in the [global] section to read:

workgroup = DOM

as this is the name of the domain we are joining.

You must also have the parameter encrypt passwords set to yes in order for your users to authenticate to the NT PDC.

Finally, add (or modify) a password server = line in the [global] section to read:

password server = DOMPDC DOMBDC1 DOMBDC2

These are the primary and backup domain controllers Samba will attempt to contact in order to authenticate users. Samba will try to contact each of these servers in order, so you may want to rearrange this list in order to spread out the authentication load among domain controllers.

Alternatively, if you want smbd to automatically determine the list of Domain controllers to use for authentication, you may set this line to be :

password server = *

This method, allows Samba to use exactly the same mechanism that NT does. This method either broadcasts or uses a WINS database in order to find domain controllers to authenticate against.

In order to actually join the domain, you must run this command:

root# net rpc join -S DOMPDC -UAdministrator%password

as we are joining the domain DOM and the PDC for that domain (the only machine that has write access to the domain SAM database) is DOMPDC. The Administrator%password is the login name and password for an account which has the necessary privilege to add machines to the domain. If this is successful you will see the message:

Joined domain DOM. or Joined 'SERV1' to realm 'MYREALM'

in your terminal window. See the net(8) man page for more details.

This process joins the server to thedomain without having to create the machine trust account on the PDC beforehand.

This command goes through the machine account password change protocol, then writes the new (random) machine account password for this Samba server into a file in the same directory in which an smbpasswd file would be stored - normally :

/usr/local/samba/private/secrets.tdb

This file is created and owned by root and is not readable by any other user. It is the key to the domain-level security for your system, and should be treated as carefully as a shadow password file.

Finally, restart your Samba daemons and get ready for clients to begin using domain security!

9.2. Why is this better than security = server?

Currently, domain security in Samba doesn't free you from having to create local Unix users to represent the users attaching to your server. This means that if domain user DOM\fred attaches to your domain security Samba server, there needs to be a local Unix user fred to represent that user in the Unix filesystem. This is very similar to the older Samba security mode security = server, where Samba would pass through the authentication request to a Windows NT server in the same way as a Windows 95 or Windows 98 server would.

Please refer to the Winbind paper for information on a system to automatically assign UNIX uids and gids to Windows NT Domain users and groups. This code is available in development branches only at the moment, but will be moved to release branches soon.

The advantage to domain-level security is that the authentication in domain-level security is passed down the authenticated RPC channel in exactly the same way that an NT server would do it. This means Samba servers now participate in domain trust relationships in exactly the same way NT servers do (i.e., you can add Samba servers into a resource domain and have the authentication passed on from a resource domain PDC to an account domain PDC.

In addition, with security = server every Samba daemon on a server has to keep a connection open to the authenticating server for as long as that daemon lasts. This can drain the connection resources on a Microsoft NT server and cause it to run out of available connections. With security = domain, however, the Samba daemons connect to the PDC/BDC only for as long as is necessary to authenticate the user, and then drop the connection, thus conserving PDC connection resources.

And finally, acting in the same manner as an NT server authenticating to a PDC means that as part of the authentication reply, the Samba server gets the user identification information such as the user SID, the list of NT groups the user belongs to, etc.

Much of the text of this document was first published in the Web magazine LinuxWorld as the article Doing the NIS/NT Samba.

10.1. Viewing and changing UNIX permissions using the NT security dialogs

Windows NT clients can use their native security settings dialog box to view and modify the underlying UNIX permissions.

Note that this ability is careful not to compromise the security of the UNIX host Samba is running on, and still obeys all the file permission rules that a Samba administrator can set.

All access to Unix/Linux system file via Samba is controlled at the operating system file access control level. When trying to figure out file access problems it is vitally important to identify the identity of the Windows user as it is presented by Samba at the point of file access. This can best be determined from the Samba log files.

10.2. How to view file security on a Samba share

From an NT4/2000/XP client, single-click with the right mouse button on any file or directory in a Samba mounted drive letter or UNC path. When the menu pops-up, click on the Properties entry at the bottom of the menu. This brings up the file properties dialog box. Click on the tab Security and you will see three buttons, Permissions, Auditing, and Ownership. The Auditing button will cause either an error message A requested privilege is not held by the client to appear if the user is not the NT Administrator, or a dialog which is intended to allow an Administrator to add auditing requirements to a file if the user is logged on as the NT Administrator. This dialog is non-functional with a Samba share at this time, as the only useful button, the Add button will not currently allow a list of users to be seen.

10.3. Viewing file ownership

Clicking on the "Ownership" button brings up a dialog box telling you who owns the given file. The owner name will be of the form :

"SERVER\user (Long name)"

Where SERVER is the NetBIOS name of the Samba server, user is the user name of the UNIX user who owns the file, and (Long name) is the descriptive string identifying the user (normally found in the GECOS field of the UNIX password database). Click on the Close button to remove this dialog.

If the parameter nt acl support is set to false then the file owner will be shown as the NT user "Everyone".

The Take Ownership button will not allow you to change the ownership of this file to yourself (clicking on it will display a dialog box complaining that the user you are currently logged onto the NT client cannot be found). The reason for this is that changing the ownership of a file is a privileged operation in UNIX, available only to the root user. As clicking on this button causes NT to attempt to change the ownership of a file to the current user logged into the NT client this will not work with Samba at this time.

There is an NT chown command that will work with Samba and allow a user with Administrator privilege connected to a Samba server as root to change the ownership of files on both a local NTFS filesystem or remote mounted NTFS or Samba drive. This is available as part of the Seclib NT security library written by Jeremy Allison of the Samba Team, available from the main Samba ftp site.

10.4. Viewing file or directory permissions

The third button is the "Permissions" button. Clicking on this brings up a dialog box that shows both the permissions and the UNIX owner of the file or directory. The owner is displayed in the form :

"SERVER\user (Long name)"

Where SERVER is the NetBIOS name of the Samba server, user is the user name of the UNIX user who owns the file, and (Long name) is the descriptive string identifying the user (normally found in the GECOS field of the UNIX password database).

If the parameter nt acl support is set to false then the file owner will be shown as the NT user "Everyone" and the permissions will be shown as NT "Full Control".

The permissions field is displayed differently for files and directories, so I'll describe the way file permissions are displayed first.

10.5. Modifying file or directory permissions

Modifying file and directory permissions is as simple as changing the displayed permissions in the dialog box, and clicking the OK button. However, there are limitations that a user needs to be aware of, and also interactions with the standard Samba permission masks and mapping of DOS attributes that need to also be taken into account.

If the parameter nt acl support is set to false then any attempt to set security permissions will fail with an "Access Denied" message.

The first thing to note is that the "Add" button will not return a list of users in Samba (it will give an error message of "The remote procedure call failed and did not execute"). This means that you can only manipulate the current user/group/world permissions listed in the dialog box. This actually works quite well as these are the only permissions that UNIX actually has.

If a permission triple (either user, group, or world) is removed from the list of permissions in the NT dialog box, then when the "OK" button is pressed it will be applied as "no permissions" on the UNIX side. If you then view the permissions again the "no permissions" entry will appear as the NT "O" flag, as described above. This allows you to add permissions back to a file or directory once you have removed them from a triple component.

As UNIX supports only the "r", "w" and "x" bits of an NT ACL then if other NT security attributes such as "Delete access" are selected then they will be ignored when applied on the Samba server.

When setting permissions on a directory the second set of permissions (in the second set of parentheses) is by default applied to all files within that directory. If this is not what you want you must uncheck the "Replace permissions on existing files" checkbox in the NT dialog before clicking "OK".

If you wish to remove all permissions from a user/group/world component then you may either highlight the component and click the "Remove" button, or set the component to only have the special "Take Ownership" permission (displayed as "O" ) highlighted.

10.6. Interaction with the standard Samba create mask parameters

There are four parameters to control interaction with the standard Samba create mask parameters. These are :

security mask

force security mode

directory security mask

force directory security mode

Once a user clicks "OK" to apply the permissions Samba maps the given permissions into a user/group/world r/w/x triple set, and then will check the changed permissions for a file against the bits set in the security mask parameter. Any bits that were changed that are not set to '1' in this parameter are left alone in the file permissions.

Essentially, zero bits in the security mask mask may be treated as a set of bits the user is not allowed to change, and one bits are those the user is allowed to change.

If not set explicitly this parameter is set to the same value as the create mask parameter. To allow a user to modify all the user/group/world permissions on a file, set this parameter to 0777.

Next Samba checks the changed permissions for a file against the bits set in the force security mode parameter. Any bits that were changed that correspond to bits set to '1' in this parameter are forced to be set.

Essentially, bits set in the force security mode parameter may be treated as a set of bits that, when modifying security on a file, the user has always set to be 'on'.

If not set explicitly this parameter is set to the same value as the force create mode parameter. To allow a user to modify all the user/group/world permissions on a file with no restrictions set this parameter to 000.

The security mask and force security mode parameters are applied to the change request in that order.

For a directory Samba will perform the same operations as described above for a file except using the parameter directory security mask instead of security mask, and force directory security mode parameter instead of force security mode .

The directory security mask parameter by default is set to the same value as the directory mask parameter and the force directory security mode parameter by default is set to the same value as the force directory mode parameter.

In this way Samba enforces the permission restrictions that an administrator can set on a Samba share, whilst still allowing users to modify the permission bits within that restriction.

If you want to set up a share that allows users full control in modifying the permission bits on their files and directories and doesn't force any particular bits to be set 'on', then set the following parameters in the smb.conf(5) file in that share specific section :

security mask = 0777

force security mode = 0

directory security mask = 0777

force directory security mode = 0

10.7. Interaction with the standard Samba file attribute mapping

Samba maps some of the DOS attribute bits (such as "read only") into the UNIX permissions of a file. This means there can be a conflict between the permission bits set via the security dialog and the permission bits set by the file attribute mapping.

One way this can show up is if a file has no UNIX read access for the owner it will show up as "read only" in the standard file attributes tabbed dialog. Unfortunately this dialog is the same one that contains the security info in another tab.

What this can mean is that if the owner changes the permissions to allow themselves read access using the security dialog, clicks "OK" to get back to the standard attributes tab dialog, and then clicks "OK" on that dialog, then NT will set the file permissions back to read-only (as that is what the attributes still say in the dialog). This means that after setting permissions and clicking "OK" to get back to the attributes dialog you should always hit "Cancel" rather than "OK" to ensure that your changes are not overridden.

12.1. Introduction

Beginning with the 2.2.0 release, Samba supports the native Windows NT printing mechanisms implemented via MS-RPC (i.e. the SPOOLSS named pipe). Previous versions of Samba only supported LanMan printing calls.

The additional functionality provided by the new SPOOLSS support includes:

• Support for downloading printer driver files to Windows 95/98/NT/2000 clients upon demand.

• Uploading of printer drivers via the Windows NT Add Printer Wizard (APW) or the Imprints tool set (refer to http://imprints.sourceforge.net).

• Support for the native MS-RPC printing calls such as StartDocPrinter, EnumJobs(), etc... (See the MSDN documentation at http://msdn.microsoft.com/ for more information on the Win32 printing API)

• Support for NT Access Control Lists (ACL) on printer objects

• Improved support for printer queue manipulation through the use of an internal databases for spooled job information

There has been some initial confusion about what all this means and whether or not it is a requirement for printer drivers to be installed on a Samba host in order to support printing from Windows clients. As a side note, Samba does not use these drivers in any way to process spooled files. They are utilized entirely by the clients.

The following MS KB article, may be of some help if you are dealing with Windows 2000 clients: How to Add Printers with No User Interaction in Windows 2000

http://support.microsoft.com/support/kb/articles/Q189/1/05.ASP

12.2. Configuration

[print$] vs. [printer$]

Previous versions of Samba recommended using a share named [printer$]. This name was taken from the printer$ service created by Windows 9x clients when a printer was shared. Windows 9x printer servers always have a printer$ service which provides read-only access via no password in order to support printer driver downloads. However, the initial implementation allowed for a parameter named printer driver location to be used on a per share basis to specify the location of the driver files associated with that printer. Another parameter named printer driver provided a means of defining the printer driver name to be sent to the client.

---

12.2.1. Creating [print$]

In order to support the uploading of printer driver files, you must first configure a file share named [print$]. The name of this share is hard coded in Samba's internals so the name is very important (print$ is the service used by Windows NT print servers to provide support for printer driver download).

You should modify the server's smb.conf file to add the global parameters and to create the following file share (of course, some of the parameter values, such as 'path' are arbitrary and should be replaced with appropriate values for your site):

[global]
    ; members of the ntadmin group should be able
    ; to add drivers and set printer properties
    ; root is implicitly a 'printer admin'
    printer admin = @ntadmin

[print$]
    path = /usr/local/samba/printers
    guest ok = yes
    browseable = yes
    read only = yes
    ; since this share is configured as read only, then we need
    ; a 'write list'.  Check the file system permissions to make
    ; sure this account can copy files to the share.  If this
    ; is setup to a non-root account, then it should also exist
    ; as a 'printer admin'
    write list = @ntadmin,root

The write list is used to allow administrative level user accounts to have write access in order to update files on the share. See the smb.conf(5) man page for more information on configuring file shares.

The requirement for guest ok = yes depends upon how your site is configured. If users will be guaranteed to have an account on the Samba host, then this is a non-issue.

Author's Note

The non-issue is that if all your Windows NT users are guaranteed to be authenticated by the Samba server (such as a domain member server and the NT user has already been validated by the Domain Controller in order to logon to the Windows NT console), then guest access is not necessary. Of course, in a workgroup environment where you just want to be able to print without worrying about silly accounts and security, then configure the share for guest access. You'll probably want to add map to guest = Bad User in the [global] section as well. Make sure you understand what this parameter does before using it though. --jerry

In order for a Windows NT print server to support the downloading of driver files by multiple client architectures, it must create subdirectories within the [print$] service which correspond to each of the supported client architectures. Samba follows this model as well.

Next create the directory tree below the [print$] share for each architecture you wish to support.

[print$]----- |-W32X86 ; "Windows NT x86" |-WIN40 ; "Windows 95/98" |-W32ALPHA ; "Windows NT Alpha_AXP" |-W32MIPS ; "Windows NT R4000" |-W32PPC ; "Windows NT PowerPC"

ATTENTION! REQUIRED PERMISSIONS

In order to currently add a new driver to you Samba host, one of two conditions must hold true: The account used to connect to the Samba host must have a uid of 0 (i.e. a root account) The account used to connect to the Samba host must be a member of the printer admin list. Of course, the connected account must still possess access to add files to the subdirectories beneath [print$]. Remember that all file shares are set to 'read only' by default.

Once you have created the required [print$] service and associated subdirectories, simply log onto the Samba server using a root (or printer admin) account from a Windows NT 4.0/2k client. Open "Network Neighbourhood" or "My Network Places" and browse for the Samba host. Once you have located the server, navigate to the "Printers..." folder. You should see an initial listing of printers that matches the printer shares defined on your Samba host.

 
Domain=[NARNIA] OS=[Unix] Server=[Samba 2.2.0-alpha3]
 
[Windows NT x86]
Printer Driver Info 1:
     Driver Name: [HP LaserJet 4000 Series PS]
 
Printer Driver Info 1:
     Driver Name: [HP LaserJet 2100 Series PS]
 
Printer Driver Info 1:
     Driver Name: [HP LaserJet 4Si/4SiMX PS]

Domain=[NARNIA] OS=[Unix] Server=[Samba 2.2.0-alpha3]
     flags:[0x800000]
     name:[\\POGO\hp-print]
     description:[POGO\\POGO\hp-print,NO DRIVER AVAILABLE FOR THIS PRINTER,]
     comment:[]
				  

Domain=[NARNIA] OS=[Unix] Server=[Samba 2.2.0-alpha3]
Successfully set hp-print to driver HP LaserJet 4000 Series PS.

#!/bin/sh

# Script to insert a new printer entry into printcap.local
#
# $1, printer name, used as the descriptive name
# $2, share name, used as the printer name for Linux
# $3, port name
# $4, driver name
# $5, location, used for the device file of the printer
# $6, win9x location

#
# Make sure we use the location that RedHat uses for local printer defs
PRINTCAP=/etc/printcap.local
DATE=`date +%Y%m%d-%H%M%S`
LP=lp
RESTART="service lpd restart"

# Keep a copy
cp $PRINTCAP $PRINTCAP.$DATE
# Add the printer to $PRINTCAP
echo ""				 			>> $PRINTCAP
echo "$2|$1:\\" 					>> $PRINTCAP
echo "  :sd=/var/spool/lpd/$2:\\" 			>> $PRINTCAP
echo "  :mx=0:ml=0:sh:\\" 				>> $PRINTCAP
echo "  :lp=/usr/local/samba/var/print/$5.prn:" 	>> $PRINTCAP

touch "/usr/local/samba/var/print/$5.prn" >> /tmp/printadd.$$ 2>&1
chown $LP "/usr/local/samba/var/print/$5.prn" >> /tmp/printadd.$$ 2>&1

mkdir /var/spool/lpd/$2
chmod 700 /var/spool/lpd/$2
chown $LP /var/spool/lpd/$2
#echo $1 >> "/usr/local/samba/var/print/$5.prn"
#echo $2 >> "/usr/local/samba/var/print/$5.prn"
#echo $3 >> "/usr/local/samba/var/print/$5.prn"
#echo $4 >> "/usr/local/samba/var/print/$5.prn"
#echo $5 >> "/usr/local/samba/var/print/$5.prn"
#echo $6 >> "/usr/local/samba/var/print/$5.prn"
$RESTART >> "/usr/local/samba/var/print/$5.prn"
# Not sure if this is needed
touch /usr/local/samba/lib/smb.conf
#
# You need to return a value, but I am not sure what it means.
#
echo "Done"
exit 0

12.3. The Imprints Toolset

The Imprints tool set provides a UNIX equivalent of the Windows NT Add Printer Wizard. For complete information, please refer to the Imprints web site at http://imprints.sourceforge.net/ as well as the documentation included with the imprints source distribution. This section will only provide a brief introduction to the features of Imprints.

	
foreach (supported architecture for a given driver)
{
     1.  rpcclient: Get the appropriate upload directory 
         on the remote server
     2.  smbclient: Upload the driver files
     3.  rpcclient: Issues an AddPrinterDriver() MS-RPC
}
	
4.  rpcclient: Issue an AddPrinterEx() MS-RPC to actually
    create the printer

12.4. Diagnosis

      [global]
        print command     - send a file to a spooler
        lpq command       - get spool queue status
        lprm command      - remove a job
      [printers]
        path = /var/spool/lpd/samba

        queuepause command   - stop a printer or print queue
        queueresume command  - start a printer or print queue

        print command = /usr/bin/lpr -r -P%p %s
        lpq command   = /usr/bin/lpq    -P%p %s
        lprm command  = /usr/bin/lprm   -P%p %j
        queuepause command = /usr/sbin/lpc -P%p stop
        queuepause command = /usr/sbin/lpc -P%p start

	print command = /tmp/saveprint %p %s

    #!/bin/saveprint
    # we make sure that we are the right user
    /usr/bin/id -p >/tmp/tmp.print
    # we run the command and save the error messages
    # replace the command with the one appropriate for your system
    /usr/bin/lpr -r -P$1 $2 2>>&/tmp/tmp.print

h4: {42} % echo hi >/tmp/hi
h4: {43} % smbclient //localhost/lw4
added interface ip=10.0.0.4 bcast=10.0.0.255 nmask=255.255.255.0
Password: 
Domain=[ASTART] OS=[Unix] Server=[Samba 2.0.7]
smb: \> print /tmp/hi
putting file /tmp/hi as hi-17534 (0.0 kb/s) (average 0.0 kb/s)
smb: \> queue
1049     3            hi-17534
smb: \> cancel 1049
Error cancelling job 1049 : code 0
smb: \> cancel 1049
Job 1049 cancelled
smb: \> queue
smb: \> exit

    testprns printer /etc/printcap

    testprns -a printer /etc/printcap

    testprns -a printer '|/bin/cat printcap'

  name|alias1|alias2...:option=value:...

  lpc -Pprinter stop

    cd /var/spool/lpd/printer   # spool directory of print jobs
    ls                          # find job files
    file dfA001myhost

   printer: ... :sh

  Printers|Printer Name|(Right Click)Properties|Postscript|Advanced|

13.1. Introduction

The Common Unix Print System (CUPS) has become very popular, but to many it is a very mystical tool. There is a great deal of uncertainty regarding CUPS and how it works. The result is seen in a large number of posting on the samba mailing lists expressing frustration when MS Windows printers appear not to work with a CUPS backr-end. /para>

This is a good time to point out how CUPS can be used and what it does. CUPS is more than just a print spooling system - it is a complete printer management system that complies with HTTP and IPP protocols. It can be managed remotely via a web browser and it can print using http and ipp protocols.

CUPS allows to creation of RAW printers (ie: NO file format translation) as well as SMART printers (ie: CUPS does file format conversion as required for the printer). In many ways this gives CUPS similar capabilities to the MS Windows print monitoring system. Of course, if you are a CUPS advocate, you would agrue that CUPS is better! In any case, let us now move on to explore how one may configure CUPS for interfacing with MS Windows print clients via Samba.

CUPS is a newcomer in the UNIX printing scene, which has convinced many people upon first trial already. However, it has quite a few new features, which make it different from other, more traditional printing systems.

13.2. Configuring smb.conf for CUPS

Printing with CUPS in the most basic smb.conf setup in Samba-3 only needs two settings: printing = cups and printcap = cups. While CUPS itself doesn't need a printcap anymore, the cupsd.conf configuration file knows two directives (example: Printcap /etc/printcap and PrintcapFormat BSD), which control if such a file should be created for the convenience of third party applications. Make sure it is set! For details see man cupsd.conf and other CUPS-related documentation.

If SAMBA is compiled against libcups, then printcap = cups uses the CUPS API to list printers, submit jobs, etc. Otherwise it maps to the System V commands with an additional -oraw option for printing. On a Linux system, you can use the ldd command to find out details (ldd may not be present on other OS platforms, or its function may be embodied by a different command):

transmeta:/home/kurt # ldd `which smbd`
        libssl.so.0.9.6 => /usr/lib/libssl.so.0.9.6 (0x4002d000)
        libcrypto.so.0.9.6 => /usr/lib/libcrypto.so.0.9.6 (0x4005a000)
        libcups.so.2 => /usr/lib/libcups.so.2 (0x40123000)
        libdl.so.2 => /lib/libdl.so.2 (0x401e8000)
        libnsl.so.1 => /lib/libnsl.so.1 (0x401ec000)
        libpam.so.0 => /lib/libpam.so.0 (0x40202000)
        libc.so.6 => /lib/libc.so.6 (0x4020b000)
        /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)

The line "libcups.so.2 => /usr/lib/libcups.so.2 (0x40123000)" shows there is CUPS support compiled into this version of Samba. If this is the case, and printing = cups is set, then any otherwise manually set print command in smb.conf is ignored.

13.3. CUPS - RAW Print Through Mode

When used in raw print through mode is will be necessary to use the printer vendor's drivers in each Windows client PC.

When CUPS printers are configured for RAW print-through mode operation it is the responsibility of the Samba client to fully render the print job (file) in a format that is suitable for direct delivery to the printer. In this case CUPS will NOT do any print file format conversion work.

The CUPS files that need to be correctly set for RAW mode printers to work are:

• /etc/cups/mime.types

• /etc/cups/mime.convs

Firstly, to enable CUPS based printing from Samba the following options must be enabled in your smb.conf file [globals] section:

• printing = CUPS

• printcap = CUPS

Cupsomatic (an enhanced printing utility that is part of some CUPS implementations) on the Samba/CUPS server does *not* add any features if a file is really printed "raw". However, if you have loaded the driver for the Windows client from the CUPS server, using the "cupsaddsmb" utility, and if this driver is one using a "Foomatic" PPD, the PJL header in question is already added on the Windows client, at the time when the driver initially generated the PostScript data and CUPS in true "-oraw" manner doesn't remove this PJL header and passes the file "as is" to its printer communication backend.

NOTE: editing in the "mime.convs" and the "mime.types" file does not *enforce* "raw" printing, it only *allows* it.

Print files that arrive from MS Windows printing are "auto-typed" by CUPS. This aids the process of determining proper treatment while in the print queue system.

• Files generated by PCL drivers and directed at PCK printers get auto-typed as application/octet-stream. Unknown file format types also get auto-typed with this tag.

• Files generated by a Postscript driver and directed at a Postscript printer are auto-typed depending on the auto-detected most suitable MIME type as:

  • * application/postscript

  • * application/vnd.cups-postscript

"application/postscript" first goes thru the "pstops" filter (where the page counting and accounting takes place). The outcome will be of MIME type "application/vnd.cups-postscript". The pstopsfilter reads and uses information from the PPD and inserts user-provided options into the PostScript file. As a consequence, the filtered file could possibly have an unwanted PJL header.

"application/postscript" will be all files with a ".ps", ".ai", ".eps" suffix or which have as their first character string one of "%!" or "<04>%".

"application/vnd.cups-postscript" will files which contain the string "LANGUAGE=POSTSCRIPT" (or similar variations with different capitalization) in the first 512 bytes, and also contain the "PJL super escape code" in the first 128 bytes ("<1B>%-12345X"). Very likely, most PostScript files generated on Windows using a CUPS or other PPD, will have to be auto-typed as "vnd.cups-postscript". A file produced with a "Generic PostScript driver" will just be tagged "application/postscript".

Once the file is in "application/vnd.cups-postscript" format, either "pstoraster" or "cupsomatic" will take over (depending on the printer configuration, as determined by the PPD in use).

A printer queue with *no* PPD associated to it is a "raw" printer and all files will go directly there as received by the spooler. The exeptions are file types "application/octet-stream" which need "passthrough feature" enabled. "Raw" queues don't do any filtering at all, they hand the file directly to the CUPS backend. This backend is responsible for the sending of the data to the device (as in the "device URI" notation as lpd://, socket://, smb://, ipp://, http://, parallel:/, serial:/, usb:/ etc.)

"cupsomatic"/Foomatic are *not* native CUPS drivers and they don't ship with CUPS. They are a Third Party add-on, developed at Linuxprinting.org. As such, they are a brilliant hack to make all models (driven by Ghostscript drivers/filters in traditional spoolers) also work via CUPS, with the same (good or bad!) quality as in these other spoolers. "cupsomatic" is only a vehicle to execute a ghostscript commandline at that stage in the CUPS filtering chain, where "normally" the native CUPS "pstoraster" filter would kick in. cupsomatic by-passes pstoraster, "kidnaps" the printfile from CUPS away and re-directs it to go through Ghostscipt. CUPS accepts this, because the associated CUPS-O-Matic-/Foomatic-PPD specifies: *cupsFilter: "application/vnd.cups-postscript 0 cupsomatic" This line persuades CUPS to hand the file to cupsomatic, once it has successfully converted it to the MIME type "application/vnd.cups-postscript". This conversion will not happen for Jobs arriving from Windows which are auto-typed "application/octet-stream", with the according changes in "/etc/cups/mime.types" in place.

CUPS is widely configurable and flexible, even regarding its filtering mechanism. Another workaround in some situations would be to have in "/etc/cups/mime.types" entries as follows:

   application/postscript           application/vnd.cups-raw  0  -
   application/vnd.cups-postscript  application/vnd.cups-raw  0  -

This would prevent all Postscript files from being filtered (rather, they will go thru the virtual "nullfilter" denoted with "-"). This could only be useful for PS printers. If you want to print PS code on non-PS printers an entry as follows could be useful:

   */*           application/vnd.cups-raw  0  -

and would effectively send *all* files to the backend without further processing.

Lastly, you could have the following entry:

   application/vnd.cups-postscript  application/vnd.cups-raw  0  my_PJL_stripping_filter

You will need to write a "my_PJL_stripping_filter" (could be a shellscript) that parses the PostScript and removes the unwanted PJL. This would need to conform to CUPS filter design (mainly, receive and pass the parameters printername, job-id, username, jobtitle, copies, print options and possibly the filename). It would be installed as world executable into "/usr/lib/cups/filters/" and will be called by CUPS if it encounters a MIME type "application/vnd.cups-postscript".

CUPS can handle "-o job-hold-until=indefinite". This keeps the job in the queue "on hold". It will only be printed upon manual release by the printer operator. This is a requirement in many "central reproduction departments", where a few operators manage the jobs of hundreds of users on some big machine, where no user is allowed to have direct access. (The operators often need to load the proper paper type before running the 10.000 page job requested by marketing for the mailing, etc.).

13.4. CUPS as a network PostScript RIP -- CUPS drivers working on server, Adobe PostScript driver with CUPS-PPDs downloaded to clients

CUPS is perfectly able to use PPD files (PostScript Printer Descriptions). PPDs can control all print device options. They are usually provided by the manufacturer -- if you own a PostSript printer, that is. PPD files are always a component of PostScript printer drivers on MS Windows or Apple Mac OS systems. They are ASCII files containing user-selectable print options, mapped to appropriate PostScript, PCL or PJL commands for the target printer. Printer driver GUI dialogs translate these options "on-the-fly" into buttons and drop-down lists for the user to select.

CUPS can load, without any conversions, the PPD file from any Windows (NT is recommended) PostScript driver and handle the options. There is a web browser interface to the print options (select http://localhost:631/printers/ and click on one "Configure Printer" button to see it), a commandline interface (see man lpoptions or try if you have lphelp on your system) plus some different GUI frontends on Linux UNIX, which can present PPD options to the users. PPD options are normally meant to become evaluated by the PostScript RIP on the real PostScript printer.

CUPS doesn't stop at "real" PostScript printers in its usage of PPDs. The CUPS developers have extended the PPD concept, to also describe available device and driver options for non-PostScript printers through CUPS-PPDs.

This is logical, as CUPS includes a fully featured PostScript interpreter (RIP). This RIP is based on Ghostscript. It can process all received PostScript (and additionally many other file formats) from clients. All CUPS-PPDs geared to non-PostScript printers contain an additional line, starting with the keyword *cupsFilter. This line tells the CUPS print system which printer-specific filter to use for the interpretation of the accompanying PostScript. Thus CUPS lets all its printers appear as PostScript devices to its clients, because it can act as a PostScript RIP for those printers, processing the received PostScript code into a proper raster print format.

CUPS-PPDs can also be used on Windows-Clients, on top of a PostScript driver (recommended is the Adobe one).

This feature enables CUPS to do a few tricks no other spooler can do:

• act as a networked PostScript RIP (Raster Image Processor), handling printfiles from all client platforms in a uniform way;

• act as a central accounting and billing server, as all files are passed through the pstops Filter and are therefor logged in the CUPS page_log. - NOTE: this can not happen with "raw" print jobs, which always remain unfiltered per definition;

• enable clients to consolidate on a single PostScript driver, even for many different target printers.

13.5. Windows Terminal Servers (WTS) as CUPS clients

This setup may be of special interest to people experiencing major problems in WTS environments. WTS need often a multitude of non-PostScript drivers installed to run their clients' variety of different printer models. This often imposes the price of much increased instability. In many cases, in an attempt to overcome this problem, site administrators have resorted to restrict the allowed drivers installed on their WTS to one generic PCL- and one PostScript driver. This however restricts the clients in the amount of printer options available for them -- often they can't get out more then simplex prints from one standard paper tray, while their devices could do much better, if driven by a different driver!

Using an Adobe PostScript driver, enabled with a CUPS-PPD, seems to be a very elegant way to overcome all these shortcomings. The PostScript driver is not known to cause major stability problems on WTS (even if used with many different PPDs). The clients will be able to (again) chose paper trays, duplex printing and other settings. However, there is a certain price for this too: a CUPS server acting as a PostScript RIP for its clients requires more CPU and RAM than just to act as a "raw spooling" device. Plus, this setup is not yet widely tested, although the first feedbacks look very promising...

13.6. Setting up CUPS for driver download

The cupsadsmb utility (shipped with all current CUPS versions) makes the sharing of any (or all) installed CUPS printers very easy. Prior to using it, you need the following settings in smb.conf:

[global]
		 load printers = yes
		 printing = cups
		 printcap name = cups

	[printers]
		 comment = All Printers
		 path = /var/spool/samba
		 browseable = no
		 public = yes
		 guest ok = yes
		 writable = no
		 printable = yes
		 printer admin = root

	[print$]
		 comment = Printer Drivers
		 path = /etc/samba/drivers
		 browseable = yes
		 guest ok = no
		 read only = yes
		 write list = root
	

For licensing reasons the necessary files of the Adobe Postscript driver can not be distributed with either Samba or CUPS. You need to download them yourself from the Adobe website. Once extracted, create a drivers directory in the CUPS data directory (usually /usr/share/cups/). Copy the Adobe files using UPPERCASE filenames, to this directory as follows:

		ADFONTS.MFM
		ADOBEPS4.DRV
		ADOBEPS4.HLP
		ADOBEPS5.DLL
		ADOBEPSU.DLL
		ADOBEPSU.HLP
		DEFPRTR2.PPD
		ICONLIB.DLL
	

Users of the ESP Print Pro software are able to install their "Samba Drivers" package for this purpose with no problem.

13.7. Sources of CUPS drivers / PPDs

On the internet you can find now many thousand CUPS-PPD files (with their companion filters), in many national languages, supporting more than 1.000 non-PostScript models.

• ESP PrintPro (http://wwwl.easysw.com/printpro/) (commercial, non-Free) is packaged with more than 3.000 PPDs, ready for successful usage "out of the box" on Linux, IBM-AIX, HP-UX, Sun-Solaris, SGI-IRIX, Compaq Tru64, Digital Unix and some more commercial Unices (it is written by the CUPS developers themselves and its sales help finance the further development of CUPS, as they feed their creators)

• the Gimp-Print-Project (http://gimp-print.sourceforge.net/) (GPL, Free Software) provides around 120 PPDs (supporting nearly 300 printers, many driven to photo quality output), to be used alongside the Gimp-Print CUPS filters;

• TurboPrint (http://www.turboprint.com/) (Shareware, non-Freee) supports roughly the same amount of printers in excellent quality;

• OMNI (http://www-124.ibm.com/developerworks/oss/linux/projects/omni/) (LPGL, Free) is a package made by IBM, now containing support for more than 400 printers, stemming from the inheritance of IBM OS/2 KnowHow ported over to Linux (CUPS support is in a Beta-stage at present);

• HPIJS (http://hpinkjet.sourceforge.net/) (BSD-style licnes, Free) supports around 120 of HP's own printers and is also providing excellent print quality now;

• Foomatic/cupsomatic (http://www.linuxprinting.org/) (LPGL, Free) from Linuxprinting.org are providing PPDs for practically every Ghostscript filter known to the world, now usable with CUPS.

NOTE: the cupsomatic trick from Linuxprinting.org is working different from the other drivers. While the other drivers take the generic CUPS raster (produced by CUPS' own pstoraster PostScript RIP) as their input, cupsomatic "kidnaps" the PostScript inside CUPS, before RIP-ping, deviates it to an external Ghostscript installation (which now becomes the RIP) and gives it back to a CUPS backend once Ghostscript is finished. -- CUPS versions from 1.1.15 and later will provide their pstoraster PostScript RIP function again inside a system-wide Ghostscript installation rather than in "their own" pstoraster filter. (This CUPS-enabling Ghostscript version may be installed either as a patch to GNU or AFPL Ghostscript, or as a complete ESP Ghostscript package). However, this will not change the cupsomatic approach of guiding the printjob along a different path through the filtering system than the standard CUPS way...

Once you installed a printer inside CUPS with one of the recommended methods (the lpadmin command, the web browser interface or one of the available GUI wizards), you can use cupsaddsmb to share the printer via Samba. cupsaddsmb prepares the driver files for comfortable client download and installation upon their first contact with this printer share.

root#  cupsaddsmb -U root infotec_IS2027
Password for root required to access localhost via SAMBA: [type in password 'secret']

Note: The following line shave been wrapped so that information is not lost.
 
root#  cupsaddsmb -v -U root infotec_IS2027
    Password for root required to access localhost via SAMBA:
    Running command: smbclient //localhost/print\$ -N -U'root%secret' -c 'mkdir W32X86;put
       /var/spool/cups/tmp/3cd1cc66376c0 W32X86/infotec_IS2027.PPD;put /usr/share/cups/drivers/
       ADOBEPS5.DLL W32X86/ADOBEPS5.DLL;put /usr/share/cups/drivers/ADOBEPSU.DLLr
       W32X86/ADOBEPSU.DLL;put /usr/share/cups/drivers/ADOBEPSU.HLP W32X86/ADOBEPSU.HLP'
    added interface ip=10.160.16.45 bcast=10.160.31.255 nmask=255.255.240.0
    added interface ip=192.168.182.1 bcast=192.168.182.255 nmask=255.255.255.0
    added interface ip=172.16.200.1 bcast=172.16.200.255 nmask=255.255.255.0
    Domain=[TUX-NET] OS=[Unix] Server=[Samba 2.2.3a.200204262025cvs]
    NT_STATUS_OBJECT_NAME_COLLISION making remote directory \W32X86
    putting file /var/spool/cups/tmp/3cd1cc66376c0 as \W32X86/infotec_IS2027.PPD (17394.6 kb/s)
      (average 17395.2 kb/s)
    putting file /usr/share/cups/drivers/ADOBEPS5.DLL as \W32X86/ADOBEPS5.DLL (10877.4 kb/s)
      (average 11343.0 kb/s)
    putting file /usr/share/cups/drivers/ADOBEPSU.DLL as \W32X86/ADOBEPSU.DLL (5095.2 kb/s)
      (average 9260.4 kb/s)
    putting file /usr/share/cups/drivers/ADOBEPSU.HLP as \W32X86/ADOBEPSU.HLP (8828.7 kb/s)
      (average 9247.1 kb/s)

    Running command: smbclient //localhost/print\$ -N -U'root%secret' -c 'mkdir WIN40;put
      /var/spool/cups/tmp/3cd1cc66376c0 WIN40/infotec_IS2027.PPD;put
      /usr/share/cups/drivers/ADFONTS.MFM WIN40/ADFONTS.MFM;put
      /usr/share/cups/drivers/ADOBEPS4.DRV WIN40/ADOBEPS4.DRV;put
      /usr/share/cups/drivers/ADOBEPS4.HLP WIN40/ADOBEPS4.HLP;put
      /usr/share/cups/drivers/DEFPRTR2.PPD WIN40/DEFPRTR2.PPD;put
      /usr/share/cups/drivers/ICONLIB.DLL WIN40/ICONLIB.DLL;put
      /usr/share/cups/drivers/PSMON.DLL WIN40/PSMON.DLL;'
    added interface ip=10.160.16.45 bcast=10.160.31.255 nmask=255.255.240.0
    added interface ip=192.168.182.1 bcast=192.168.182.255 nmask=255.255.255.0
    added interface ip=172.16.200.1 bcast=172.16.200.255 nmask=255.255.255.0
    Domain=[TUX-NET] OS=[Unix] Server=[Samba 2.2.3a.200204262025cvs]
    NT_STATUS_OBJECT_NAME_COLLISION making remote directory \WIN40
    putting file /var/spool/cups/tmp/3cd1cc66376c0 as \WIN40/infotec_IS2027.PPD (26091.5 kb/s)
      (average 26092.8 kb/s)
    putting file /usr/share/cups/drivers/ADFONTS.MFM as \WIN40/ADFONTS.MFM (11241.6 kb/s)
      (average 11812.9 kb/s)
    putting file /usr/share/cups/drivers/ADOBEPS4.DRV as \WIN40/ADOBEPS4.DRV (16640.6 kb/s)
      (average 14679.3 kb/s)
    putting file /usr/share/cups/drivers/ADOBEPS4.HLP as \WIN40/ADOBEPS4.HLP (11285.6 kb/s)
      (average 14281.5 kb/s)
    putting file /usr/share/cups/drivers/DEFPRTR2.PPD as \WIN40/DEFPRTR2.PPD (823.5 kb/s)
      (average 12944.0 kb/s)
    putting file /usr/share/cups/drivers/ICONLIB.DLL as \WIN40/ICONLIB.DLL (19226.2 kb/s)
      (average 13169.7 kb/s)
    putting file /usr/share/cups/drivers/PSMON.DLL as \WIN40/PSMON.DLL (18666.1 kb/s)
      (average 13266.7 kb/s)

    Running command: rpcclient localhost -N -U'root%secret' -c 'adddriver "Windows NT x86"
       "infotec_IS2027:ADOBEPS5.DLL:infotec_IS2027.PPD:ADOBEPSU.DLL:ADOBEPSU.HLP:NULL:RAW:NULL"'
    cmd = adddriver "Windows NT x86" "infotec_IS2027:ADOBEPS5.DLL:infotec_IS2027.PPD:ADOBEPSU.DLL:
       ADOBEPSU.HLP:NULL:RAW:NULL"
    Printer Driver infotec_IS2027 successfully installed.

    Running command: rpcclient localhost -N -U'root%secret' -c 'adddriver "Windows 4.0"
       "infotec_IS2027:ADOBEPS4.DRV:infotec_IS2027.PPD:NULL:ADOBEPS4.HLP:PSMON.DLL:RAW:
       ADFONTS.MFM,DEFPRTR2.PPD,ICONLIB.DLL"'
    cmd = adddriver "Windows 4.0" "infotec_IS2027:ADOBEPS4.DRV:infotec_IS2027.PPD:NULL:
       ADOBEPS4.HLP:PSMON.DLL:RAW:ADFONTS.MFM,DEFPRTR2.PPD,ICONLIB.DLL"
    Printer Driver infotec_IS2027 successfully installed.

    Running command: rpcclient localhost -N -U'root%secret'
       -c 'setdriver infotec_IS2027 infotec_IS2027'
    cmd = setdriver infotec_IS2027 infotec_IS2027
    Succesfully set infotec_IS2027 to driver infotec_IS2027.

    root# 

13.8. The CUPS Filter Chains

The following diagrams reveal how CUPS handles print jobs.

#########################################################################
#
# CUPS in and of itself has this (general) filter chain (CAPITAL
# letters are FILE-FORMATS or MIME types, other are filters (this is
# true for pre-1.1.15 of pre-4.3 versions of CUPS and ESP PrintPro):
#
# SOMETHNG-FILEFORMAT
#      |
#      |
#      V
#     somethingtops
#      |
#      |
#      V
# APPLICATION/POSTSCRIPT
#      |
#      |
#      V
#     pstops
#      |
#      |
#      V
# APPLICATION/VND.CUPS-POSTSCRIPT
#      |
#      |
#      V
#     pstoraster   # as shipped with CUPS, independent from any Ghostscipt
#      |           # installation on the system
#      |  (= "postscipt interpreter")
#      |
#      V
# APPLICATION/VND.CUPS-RASTER
#      |
#      |
#      V
#     rastertosomething  (f.e. Gimp-Print filters may be plugged in here)
#      |   (= "raster driver")
#      |
#      V
# SOMETHING-DEVICE-SPECIFIC
#      |
#      |
#      V
#     backend
#
#
# ESP PrintPro has some enhanced "rastertosomething" filters as compared to
# CUPS, and also a somewhat improved "pstoraster" filter.
#
# NOTE: Gimp-Print and some other 3rd-Party-Filters (like TurboPrint) to
#       CUPS and ESP PrintPro plug-in where rastertosomething is noted.
#
#########################################################################

#########################################################################
#
# This is how "cupsomatic" comes into play:
# =========================================
#
# SOMETHNG-FILEFORMAT
#      |
#      |
#      V
#    somethingtops
#      |
#      |
#      V
# APPLICATION/POSTSCRIPT
#      |
#      |
#      V
#    pstops
#      |
#      |
#      V
# APPLICATION/VND.CUPS-POSTSCRIPT ----------------+
#      |                                          |
#      |                                          V
#      V                                         cupsomatic
#    pstoraster                                  (constructs complicated
#      |  (= "postscipt interpreter")            Ghostscript commandline
#      |                                         to let the file be
#      V                                         processed by a
# APPLICATION/VND.CUPS-RASTER                    "-sDEVICE=s.th."
#      |                                         call...)
#      |                                          |
#      V                                          |
#    rastertosomething                          V
#      |    (= "raster driver")     +-------------------------+
#      |                            | Ghostscript at work.... |
#      V                            |                         |
# SOMETHING-DEVICE-SPECIFIC         *-------------------------+
#      |                                          |
#      |                                          |
#      V                                          |
#    backend >------------------------------------+
#      |
#      |
#      V
#    THE PRINTER
#
#
# Note, that cupsomatic "kidnaps" the printfile after the
# "APPLICATION/VND.CUPS-POSTSCRPT" stage and deviates it through
# the CUPS-external, systemwide Ghostscript installation, bypassing the
# "pstoraster" filter (therefor also bypassing the CUPS-raster-drivers
# "rastertosomething", and hands the rasterized file directly to the CUPS
# backend...
#
# cupsomatic is not made by the CUPS developers. It is an independent
# contribution to printing development, made by people from
# Linuxprinting.org. (see also http://www.cups.org/cups-help.html)
#
# NOTE: Gimp-Print and some other 3rd-Party-Filters (like TurboPrint) to
#       CUPS and ESP PrintPro plug-in where rastertosomething is noted.
#
#########################################################################

#########################################################################
#
# And this is how it works for ESP PrintPro from 4.3:
# ===================================================
#
# SOMETHNG-FILEFORMAT
#      |
#      |
#      V
#     somethingtops
#      |
#      |
#      V
# APPLICATION/POSTSCRIPT
#      |
#      |
#      V
#     pstops
#      |
#      |
#      V
# APPLICATION/VND.CUPS-POSTSCRIPT
#      |
#      |
#      V
#     gsrip
#      |  (= "postscipt interpreter")
#      |
#      V
# APPLICATION/VND.CUPS-RASTER
#      |
#      |
#      V
#     rastertosomething  (f.e. Gimp-Print filters may be plugged in here)
#      |   (= "raster driver")
#      |
#      V
# SOMETHING-DEVICE-SPECIFIC
#      |
#      |
#      V
#     backend
#
# NOTE: Gimp-Print and some other 3rd-Party-Filters (like TurboPrint) to
#       CUPS and ESP PrintPro plug-in where rastertosomething is noted.
#
#########################################################################

#########################################################################
#
# This is how "cupsomatic" would come into play with ESP PrintPro:
# ================================================================
#
#
# SOMETHNG-FILEFORMAT
#      |
#      |
#      V
#    somethingtops
#      |
#      |
#      V
# APPLICATION/POSTSCRIPT
#      |
#      |
#      V
#    pstops
#      |
#      |
#      V
# APPLICATION/VND.CUPS-POSTSCRIPT ----------------+
#      |                                          |
#      |                                          V
#      V                                         cupsomatic
#    gsrip                                       (constructs complicated
#      |  (= "postscipt interpreter")            Ghostscript commandline
#      |                                         to let the file be
#      V                                         processed by a
# APPLICATION/VND.CUPS-RASTER                    "-sDEVICE=s.th."
#      |                                         call...)
#      |                                          |
#      V                                          |
#    rastertosomething                          V
#      |   (= "raster driver")      +-------------------------+
#      |                            | Ghostscript at work.... |
#      V                            |                         |
# SOMETHING-DEVICE-SPECIFIC         *-------------------------+
#      |                                          |
#      |                                          |
#      V                                          |
#    backend >------------------------------------+
#      |
#      |
#      V
#    THE PRINTER
#
# NOTE: Gimp-Print and some other 3rd-Party-Filters (like TurboPrint) to
#       CUPS and ESP PrintPro plug-in where rastertosomething is noted.
#
#########################################################################

#########################################################################
#
# And this is how it works for CUPS from 1.1.15:
# ==============================================
#
# SOMETHNG-FILEFORMAT
#      |
#      |
#      V
#     somethingtops
#      |
#      |
#      V
# APPLICATION/POSTSCRIPT
#      |
#      |
#      V
#     pstops
#      |
#      |
#      V
# APPLICATION/VND.CUPS-POSTSCRIPT-----+
#                                     |
#                  +------------------v------------------------------+
#                  | Ghostscript                                     |
#                  | at work...                                      |
#                  | (with                                           |
#                  | "-sDEVICE=cups")                                |
#                  |                                                 |
#                  |         (= "postscipt interpreter")             |
#                  |                                                 |
#                  +------------------v------------------------------+
#                                     |
#                                     |
# APPLICATION/VND.CUPS-RASTER >-------+
#      |
#      |
#      V
#     rastertosomething
#      |   (= "raster driver")
#      |
#      V
# SOMETHING-DEVICE-SPECIFIC
#      |
#      |
#      V
#     backend
#
#
# NOTE: since version 1.1.15 CUPS "outsourced" the pstoraster process to
#       Ghostscript. GNU Ghostscript needs to be patched to handle the
#       CUPS requirement; ESP Ghostscript has this builtin. In any case,
#       "gs -h" needs to show up a "cups" device. pstoraster is now a
#       calling an appropriate "gs -sDEVICE=cups..." commandline to do
#       the job. It will output "application/vnd.cup-raster", which will
#       be finally processed by a CUPS raster driver "rastertosomething"
#       Note the difference to "cupsomatic", which will *not* output
#       CUPS-raster, but a final version of the printfile, ready to be
#       sent to the printer. cupsomatic also doesn't use the "cups"
#       devicemode in Ghostscript, but one of the classical devicemodes....
#
# NOTE: Gimp-Print and some other 3rd-Party-Filters (like TurboPrint) to
#       CUPS and ESP PrintPro plug-in where rastertosomething is noted.
#
#########################################################################

#########################################################################
#
# And this is how it works for CUPS from 1.1.15, with cupsomatic included:
# ========================================================================
#
# SOMETHNG-FILEFORMAT
#      |
#      |
#      V
#     somethingtops
#      |
#      |
#      V
# APPLICATION/POSTSCRIPT
#      |
#      |
#      V
#     pstops
#      |
#      |
#      V
# APPLICATION/VND.CUPS-POSTSCRIPT-----+
#                                     |
#                  +------------------v------------------------------+
#                  | Ghostscript        . Ghostscript at work....    |
#                  | at work...         . (with "-sDEVICE=           |
#                  | (with              .            s.th."        |
#                  | "-sDEVICE=cups")   .                            |
#                  |                    .                            |
#                  | (CUPS standard)    .      (cupsomatic)          |
#                  |                    .                            |
#                  |          (= "postscript interpreter")           |
#                  |                    .                            |
#                  +------------------v--------------v---------------+
#                                     |              |
#                                     |              |
# APPLICATION/VND.CUPS-RASTER >-------+              |
#      |                                             |
#      |                                             |
#      V                                             |
#     rastertosomething                            |
#      |   (= "raster driver")                       |
#      |                                             |
#      V                                             |
# SOMETHING-DEVICE-SPECIFIC >------------------------+
#      |
#      |
#      V
#     backend
#
#
# NOTE: Gimp-Print and some other 3rd-Party-Filters (like TurboPrint) to
#       CUPS and ESP PrintPro plug-in where rastertosomething is noted.
#
##########################################################################

13.9. CUPS Print Drivers and Devices

CUPS ships with good support for HP LaserJet type printers. You can install the driver as follows:

• lpadmin -p laserjet4plus -v parallel:/dev/lp0 -E -m laserjet.ppd

---

13.9.1. Further printing steps

Always also consult the database on linuxprinting.org for all recommendations about which driver is best used for each printer:

http://www.linuxprinting.org/printer_list.cgi

There select your model and click on "Show". You'll arrive at a page listing all drivers working with your model. There will always be *one* recommended one. Try this one first. In your case ("HP LaserJet 4 Plus"), you'll arrive here:

http://www.linuxprinting.org/show_printer.cgi?recnum=75104

The recommended driver is "ljet4". It has a link to the page for the ljet4 driver too:

http://www.linuxprinting.org/show_driver.cgi?driver=ljet4

On the driver's page, you'll find important and detailed info about how to use that driver within the various available spoolers. You can generate a PPD for CUPS. The PPD contains all the info about how to use your model and the driver; this is, once installed, working transparently for the user -- you'll only need to choose resolution, paper size etc. from the web-based menu or from the print dialog GUI or from the commandline...

On the driver's page, choose to use the "PPD-O-Matic" online PPD generator program. Select your model and click "Generate PPD file". When you safe the appearing ASCII text file, don't use "cut'n'past" (as it could possiblly corrupt line endings and tabs), but use "Save as..." in your browser's menu. Save it at "/some/path/on/your/filesystem/somewhere/my-name-for-my-printer.ppd"

Then install the printer:

    "lpadmin -p laserjet4plus -v parallel:/dev/lp0 -E \
          -P /some/path/on/your/filesystem/somewhere/my-name-for-my-printer.ppd"

Note, that for all the "Foomatic-PPDs" from Linuxprinting.org, you also need a special "CUPS filter" named "cupsomatic". Get the latest version of "cupsomatic" from:

http://www.linuxprinting.org/cupsomatic

This needs to be copied to /usr/lib/cups/filter/cupsomatic and be made world executable. This filter is needed to read and act upon the specially encoded Foomatic comments, embedded in the printfile, which in turn are used to construct (transparently for you, the user) the complicated ghostscript command line needed for your printer/driver combo.

You can have a look at all the options for the Ghostscript commandline supported by your printer and the ljet4 driver by going to the section "Execution details", selecting your model (Laserjet 4 Plus) and clicking on "Show execution details". This will bring up this web page:

http://www.linuxprinting.org/execution.cgi?driver=ljet4&printer=75104&.submit=Show+execution+details

The ingenious thing is that the database is kept current. If there is a bug fix and an improvement somewhere in the database, you will always get the most current and stable and feature-rich driver by following the steps described above.

Till Kamppeter from MandrakeSoft is doing an excellent job here that too few people are aware of. (So if you use it often, please send him a note showing your appreciation).

The latest and greatest improvement now is support for "custom page sizes" for all those printers which support it.

"cupsomatic" is documented here:

http://www.linuxprinting.org/cups-doc.html

More printing tutorial info may be found here:

http://www.linuxprinting.org/kpfeifle/LinuxKongress2002/Tutorial/

Note, that *all* the Foomatic drivers listed on Linuxprinting.org (now approaching the "all-time high" number of 1.000 for the supported models) are using a special filtering chain involving Ghostscript, as described in this document.

Summary - You need:

A "foomatic+something" PPD is not enough to print with CUPS (but it is *one* important component)

The "cupsomatic" filter script (Perl) in /usr/lib/cups/filters/

Perl to make cupsomatic run

Ghostscript (because it is called and controlled by the PPD/cupsomatic combo in a way to fit your printermodel/driver combo.

Ghostscript *must*, depending on the driver/model, contain support for a certain "device" (as shown by "gs -h")

In the case of the "hpijs" driver, you need a Ghostscript version, which has "ijs" amongst its supported devices in "gs -h". In the case of "hpijs+foomatic", a valid ghostscript commandline would be reading like this:

       gs -q -dBATCH -dPARANOIDSAFER -dQUIET -dNOPAUSE -sDEVICE=ijs       \
             -sIjsServer=hpijsPageSize -dDuplex=Duplex Model        \
             -rResolution,PS:MediaPosition=InputSlot -dIjsUseOutputFD \
             -sOutputFile=- -

Note, that with CUPS and the "hpijs+foomatic" PPD (plus Perl and cupsomatic) you don't need to remember this. You can choose the available print options thru a GUI print command (like "glp" from ESP's commercially supported PrintPro software, or KDE's "kprinter", or GNOME's "gtklp" or the independent "xpp") or the CUPS web interface via human-readable drop-down selection menus.

If you use "ESP Ghostscript" (also under the GPL, provided by Easy Software Products, the makers of CUPS, downloadable from http://www.cups.org/software.html, co-maintained by the developers of linuxprinting.org), you are guaranteed to have in use the most uptodate, bug-fixed, enhanced and stable version of a Free Ghostscript. It contains support for ~300 devices, whereas plain vanilla GNU Ghostscript 7.05 only has ~200.

If you print only one CUPS test page, from the web interface and when you try to print a windows test page, it acts like the job was never sent:

Can you print "standard" jobs from the CUPS machine?

Are the jobs from Windows visible in the Web interface on CUPS (http://localhost:631/)?

Most important: What kind of printer driver are you using on the Windows clients?

You can try to get a more detailed debugging info by setting "LogLevel debug" in /etc/cups/cupsd.conf, re-start cupsd and investigate /var/log/cups/error_log for the whereabouts of your Windows-originating printjobs:

what does the "auto-typing" line say? which is the "MIME type" CUPS thinks is arriving from the Windows clients?

are there "filter" available for this MIME type?

are there "filter rules" defined in "/etc/cups/mime.convs" for this MIME type?

13.10. Limiting the number of pages users can print

The feature you want is dependent on the real print subsystem you're using. Samba's part is always to receive the job files from the clients (filtered *or* unfiltered) and hand it over to this printing subsystem.

Of course one could "hack" things with one's own scripts.

But there is CUPS (Common Unix Printing System). CUPS supports "quotas". Quotas can be based on sizes of jobs or on the number of pages or both, and are spanning any time period you want.

This is an example command how root would set a print quota in CUPS, assuming an existing printer named "quotaprinter":

  lpadmin -p quotaprinter -o job-quota-period=604800 -o job-k-limit=1024 \
       -o job-page-limit=100

This would limit every single user to print 100 pages or 1024 KB of data (whichever comes first) within the last 604.800 seconds ( = 1 week).

For CUPS to count correctly, the printfile needs to pass the CUPS "pstops" filter, otherwise it uses a "dummy" count of "1". Some printfiles don't pass it (eg: image files) but then those are mostly 1 page jobs anyway. This also means, proprietary drivers for the target printer running on the client computers and CUPS/Samba then spooling these files as "raw" (i.e. leaving them untouched, not filtering them), will be counted as "1-pagers" too!

You need to send PostScript from the clients (i.e. run a PostScript driver there) for having the chance to get accounting done. If the printer is a non-PostScript model, you need to let CUPS do the job to convert the file to a print-ready format for the target printer. This will be working for currently ~1.000 different printer models, see

     http://www.linuxprinting.org/printer_list.cgi

Before CUPS-1.1.16 your only option was to use the Adobe PostScript Driver on the Windows clients. The output of this driver was not always passed thru the "pstops" filter on the CUPS/Samba side, and therefor was not counted correctly (the reason is that it often --- depending on the "PPD" being used --- did write a "PJL"-header in front of the real PostScript which made CUPS to skip the pstops and go directy to the "pstoraster" stage).

From CUPS-1.1.16 onward you can use the "CUPS PostScript Driver for Windows NT/2K/XP clients" (it is tagged in the download area of http://www.cups.org/ as the "cups-samba-1.1.16.tar.gz" package). It is *not* working for Win9x/ME clients. But it:

You can read more about the setup of this combination in the manpage for "cupsaddsmb" (only present with CUPS installed, only current with CUPS 1.1.16).

These are the items CUPS logs in the "page_log" for every single *page* of a job:

Here is an extract of my CUPS server's page_log file to illustrate the format and included items:

infotec_IS2027 kurt 40 [22/Nov/2002:13:18:03 +0100] 1 2 #marketing infotec_IS2027 kurt 40 [22/Nov/2002:13:18:03 +0100] 2 2 #marketing infotec_IS2027 kurt 40 [22/Nov/2002:13:18:03 +0100] 3 2 #marketing infotec_IS2027 kurt 40 [22/Nov/2002:13:18:03 +0100] 4 2 #marketing infotec_IS2027 kurt 40 [22/Nov/2002:13:18:03 +0100] 5 2 #marketing infotec_IS2027 kurt 40 [22/Nov/2002:13:18:03 +0100] 6 2 #marketing

This was Job ID "40", printed on "infotec_IS2027" by user "kurt", a 6-page job printed in 2 copies and billed to "#marketing"...

What flaws or shortcomings are there?

But this is the best system out there currently. And there are huge improvements under development:

Other than the current stage of the CUPS development, I don't know any other ready-to-use tool which you could consider.

You can download the driver files from http://www.cups.org/software.html. It is a separate package from the CUPS base software files, tagged as "CUPS 1.1.16 Windows NT/2k/XP Printer Driver for SAMBA (tar.gz, 192k)". The filename to download is "cups-samba-1.1.16.tar.gz". Upon untar-/unzip-ping it will reveal the files:

cups-samba.install cups-samba.license cups-samba.readme cups-samba.remove cups-samba.ss

These have been packaged with the ESP meta packager software "EPM". The *.install and *.remove files are simple shell script, which untars the *.ss (which is nothing else than a tar-archive) and puts its contents into /usr/share/cups/drivers/. Its contents are 3 files:

cupsdrvr.dll cupsui.dll cups.hlp

Due to a bug one CUPS release puts the cups.hlp into /usr/share/drivers/ instead of /usr/share/cups/drivers/. To work around this, copy/move the file after running the "./cups-samba.install" script manually to the right place: cp /usr/share/drivers/cups.hlp /usr/share/cups/drivers/

This new CUPS PostScript driver is currently binary-only, but free no source code is provided (yet). The reason is this: it has been developed with the help of the Microsoft Driver Developer Kit (DDK) and compiled with Microsoft Visual Studio 6. It is not clear to the driver developers if they are allowed to distribute the whole of the source code as Free Software. However, they will likely release the "diff" in source code under the GPL, so anybody with a license of Visual Studio and a DDK will be able to compile for him/herself. Once you have run the install script (and possibly manually moved the "cups.hlp" file to "/usr/share/cups/drivers/"), the driver is ready to be put into Samba's [print$] share (which often maps to "/etc/samba/drivers/" and contains a subdir tree with WIN40 and W32X86 branches), by running "cupsaddsmb" (see also "man cupsaddsmb" for CUPS 1.1.16). [Don't forget to put root into the smbpasswd file by running "smbpasswd" should you run this whole procedure for the first time.] Once the driver files are in the [print$] share, they are ready to be downloaded and installed by the Win NT/2k/XP clients.

Win 9x/ME clients won't work with this driver. For these you'd still need to use the ADOBE*.* drivers as previously.

It is not harming if you've still the ADOBE*.* driver files from previous installations in the "/usr/share/cups/drivers/" directory. The new cupsaddsmb (from 1.1.16) will automatically use the "newest" installed driver (which here then is the CUPS drivers).

Should your Win clients have had the old ADOBE*.* files and the Adobe PostScript drivers installed, the download and installation of the new CUPS PostScript driver for Windows NT/2k/XP will fail at first. It is not enough to "delete" the printer (as the driver files will still be kept by the clients and re-used if you try to re-install the printer). To really get rid of the Adobe driver files on the clients, open the "Printers" folder (possibly via "Start --> Settings --> Control Panel --> Printers"), right-click onto the folder background and select "Server Properties". A new dialog opens; select the "Drivers" tab; on the list select the driver you want to delete and click on the "Delete" button. (This will only work if there is no single printer left which uses that particular driver -- you need to "delete" all printers using this driver in the "Printers" folder first.)

Once you have successfully downloaded the CUPS PostScript driver to a client, you can easily switch all printers to this one by proceeding as described elsewhere in the "Samba HOWTO Collection" to change a driver for an existing printer.

What are the benefits with the "CUPS PostScript driver for Windows NT/2k/XP" as compared to the Adobe drivers?

• no hassle with the Adobe EULA

• no hassle with the question "where do I get the ADOBE*.* driver files from?"

• the Adobe drivers (depending on the printer PPD associated with them) often put a PJL header in front of the core PostScript part of the print file (thus the file starts with "1B%-12345X" or "escape%-12345X" instead of "%!PS"). This leads to the CUPS daemon autotyping the arriving file as a print-ready file, not requiring a pass thru the "pstops" filter (to speak more technical, it is not regarded as the generic MIME type "application/postscript", but as the more special MIME type "application/cups.vnd-postscript"), which therefore also leads to the page accounting in "/var/log/cups/page_log" not receiving the exact mumber of pages; instead the dummy page number of "1" is logged in a standard setup)

• the Adobe driver has more options to "mis-configure" the PostScript generated by it (like setting it inadvertedly to "Optimize for Speed", instead of "Optimize for Portability", which could lead to CUPS being unable to process it)

• the CUPS PostScript driver output sent by Windows clients to the CUPS server will be guaranteed to be auto-typed as generic MIME type "application/postscript", thusly passing thru the CUPS "pstops" filter and logging the correct number of pages in the page_log for accounting and quota purposes

• the CUPS PostScript driver supports the sending of additional print options by the Win NT/2k/XP clients, such as naming the CUPS standard banner pages (or the custom ones, should they be installed at the time of driver download), using the CUPS "page-label" option, setting a job-priority and setting the scheduled time of printing (with the option to support additional useful IPP job attributes in the future).

• the CUPS PostScript driver supports the inclusion of the new "*cupsJobTicket" comments at the beginnig of the PostScript file (which could be used in the future for all sort of beneficial extensions on the CUPS side, but which will not disturb any other application as those will regard it as a comment and simply ignore it).

• the CUPS PostScript driver will be the heart of the fully fledged CUPS IPP client for Windows NT/2k/XP to be released soon (probably alongside the first Beta release for CUPS 1.2).

13.11. Advanced Postscript Printing from MS Windows

Let the Windows Clients use a PostScript driver to deliver poistscript to the samba print server (just like any Linux or Unix Client would also use PostScript to send to the server)

Make the Unix printing subsystem to which Samba sends the job convert the incoming PostScript files to the native print format of the target printers (would be PCL if you have an HP printer)

Now if you are afraid that this would just mean using a *Generic* PostScript driver for the clients that has no Simplex/Duplex selection, and no paper tray choice, but you need them to be able to set up print jobs, with all the bells and whistles of your printers:-

CUPS PPDs are working perfectly on Windows clients who use Adobe PostScript drivers (or the new CUPS PostScript driver for Windows NT/2K/XP). Clients can use them to setup the job to their liking and CUPS will use the received job options to make the (PCL-, ESC/P- or PostScript-) printer behave as required.

If you want to have the additional benefit of page count logging and accounting then the CUPS PostScript driver is the best choice (better than the Adobe one).

If you want to make the drivers downloadable for the clients then "cupsaddsmb" is your friend. It will setup the [print$] share on the Samba host to be ready to serve the clients for a "point and print" driver installation.

What strings are attached?

There are some. But, given the sheer CPU power you can buy nowadays, these can be overcome easily. The strings:

Well, if the CUPS/Samba side will have to print to many printers serving many users, you probably will need to set up a second server (which can do automatic load balancing with the first one, plus a degree of fail-over mechanism). Converting the incoming PostScript jobs, "interpreting" them for non-PostScript printers, amounts to the work of a "RIP" (Raster Image Processor) done in software. This requires more CPU and RAM than for the mere "raw spooling" task your current setup is solving. It all depends on the avarage and peak printing load the server should be able to handle.

13.12. Auto-Deletion of CUPS spool files

Samba print files pass thru two "spool" directories. One the incoming directory managed by Samba, (set eg: in the "path = /var/spool/samba" directive in the [printers] section of "smb.conf"). Second is the spool directory of your UNIX print subsystem. For CUPS it is normally "/var/spool/cups/", as set by the cupsd.conf directive "RequestRoot /var/spool/cups".

I am not sure, which one of your directories keeps the files. From what you say, it is most likely the Samba part.

For the CUPS part, you may want to consult:

There are the settings described for your CUPS daemon, which could lead to completed job files not being deleted.

"PreserveJobHistory Yes" -- keeps some details of jobs in cupsd's mind (well it keeps the "c12345", "c12346" etc. files in the CUPS spool directory, which do a similar job as the old-fashioned BSD-LPD control files). This is set to "Yes" as a default.

"PreserveJobFiles Yes" -- keeps the job files themselves in cupsd's mind (well it keeps the "d12345", "d12346" etc. files in the CUPS spool directory...). This is set to "No" as the CUPS default.

"MaxJobs 500" -- this directive controls the maximum number of jobs that are kept in memory. Once the number of jobs reaches the limit, the oldest completed job is automatically purged from the system to make room for the new one. If all of the known jobs are still pending or active then the new job will be rejected. Setting the maximum to 0 disables this functionality. The default setting is 0.

(There are also additional settings for "MaxJobsPerUser" and "MaxJobsPerPrinter"...)

For everything to work as announced, you need to have three things:

Note, that in this case all other manually set printing-related commands (like "print command", "lpq command", "lprm command", "lppause command" or "lpresume command") are ignored and they should normally have no influence what-so-ever on your printing.

If you want to do things manually, replace the "printing = cups" by "printing = bsd". Then your manually set commands may work (haven't tested this), and a "print command = lp -d %P %s; rm %s" may do what you need.

You forgot to mention the CUPS version you're using. If you did set things up as described in the man pages, then the Samba spool files should be deleted. Otherwise it may be a bug. On the CUPS side, you can control the behaviour as described above.

If you have more problems, post the output of these commands:

grep -v ^# /etc/cups/cupsd.conf | grep -v ^$ grep -v ^# /etc/samba/smb.conf | grep -v ^$ | grep -v "^;"

(adapt paths as needed). These commands sanitize the files and cut out the empty lines and lines with comments, providing the "naked settings" in a compact way.

14.1. Abstract

Integration of UNIX and Microsoft Windows NT through a unified logon has been considered a "holy grail" in heterogeneous computing environments for a long time. We present winbind, a component of the Samba suite of programs as a solution to the unified logon problem. Winbind uses a UNIX implementation of Microsoft RPC calls, Pluggable Authentication Modules, and the Name Service Switch to allow Windows NT domain users to appear and operate as UNIX users on a UNIX machine. This paper describes the winbind system, explaining the functionality it provides, how it is configured, and how it works internally.

14.2. Introduction

It is well known that UNIX and Microsoft Windows NT have different models for representing user and group information and use different technologies for implementing them. This fact has made it difficult to integrate the two systems in a satisfactory manner.

One common solution in use today has been to create identically named user accounts on both the UNIX and Windows systems and use the Samba suite of programs to provide file and print services between the two. This solution is far from perfect however, as adding and deleting users on both sets of machines becomes a chore and two sets of passwords are required both of which can lead to synchronization problems between the UNIX and Windows systems and confusion for users.

We divide the unified logon problem for UNIX machines into three smaller problems:

• Obtaining Windows NT user and group information

• Authenticating Windows NT users

• Password changing for Windows NT users

Ideally, a prospective solution to the unified logon problem would satisfy all the above components without duplication of information on the UNIX machines and without creating additional tasks for the system administrator when maintaining users and groups on either system. The winbind system provides a simple and elegant solution to all three components of the unified logon problem.

14.3. What Winbind Provides

Winbind unifies UNIX and Windows NT account management by allowing a UNIX box to become a full member of a NT domain. Once this is done the UNIX box will see NT users and groups as if they were native UNIX users and groups, allowing the NT domain to be used in much the same manner that NIS+ is used within UNIX-only environments.

The end result is that whenever any program on the UNIX machine asks the operating system to lookup a user or group name, the query will be resolved by asking the NT domain controller for the specified domain to do the lookup. Because Winbind hooks into the operating system at a low level (via the NSS name resolution modules in the C library) this redirection to the NT domain controller is completely transparent.

Users on the UNIX machine can then use NT user and group names as they would use "native" UNIX names. They can chown files so that they are owned by NT domain users or even login to the UNIX machine and run a UNIX X-Window session as a domain user.

The only obvious indication that Winbind is being used is that user and group names take the form DOMAIN\user and DOMAIN\group. This is necessary as it allows Winbind to determine that redirection to a domain controller is wanted for a particular lookup and which trusted domain is being referenced.

Additionally, Winbind provides an authentication service that hooks into the Pluggable Authentication Modules (PAM) system to provide authentication via a NT domain to any PAM enabled applications. This capability solves the problem of synchronizing passwords between systems since all passwords are stored in a single location (on the domain controller).

14.4. How Winbind Works

The winbind system is designed around a client/server architecture. A long running winbindd daemon listens on a UNIX domain socket waiting for requests to arrive. These requests are generated by the NSS and PAM clients and processed sequentially.

The technologies used to implement winbind are described in detail below.

14.5. Installation and Configuration

Many thanks to John Trostel jtrostel@snapserver.com for providing the HOWTO for this section.

This HOWTO describes how to get winbind services up and running to control access and authenticate users on your Linux box using the winbind services which come with SAMBA 2.2.2.

root# autoconf
root# make clean
root# rm config.cache
root# ./configure
root# make
root# make install

	passwd:     files winbind
	shadow:     files 
	group:      files winbind

[global]
     <...>
     # separate domain and username with '+', like DOMAIN+username
     winbind separator = +
     # use uids from 10000 to 20000 for domain users
     winbind uid = 10000-20000
     # use gids from 10000 to 20000 for domain groups
     winbind gid = 10000-20000
     # allow enumeration of winbind users and groups
     winbind enum users = yes
     winbind enum groups = yes
     # give winbind users a real shell (only needed if they have telnet access)
     template homedir = /home/winnt/%D/%U
     template shell = /bin/bash

CEO+Administrator
CEO+burdell
CEO+Guest
CEO+jt-ad
CEO+krbtgt
CEO+TsInternetUser

root# /usr/local/samba/bin/wbinfo -g
CEO+Domain Admins
CEO+Domain Users
CEO+Domain Guests
CEO+Domain Computers
CEO+Domain Controllers
CEO+Cert Publishers
CEO+Schema Admins
CEO+Enterprise Admins
CEO+Group Policy Creator Owners

start() {
        KIND="SMB"
        echo -n $"Starting $KIND services: "
        daemon /usr/local/samba/bin/smbd $SMBDOPTIONS
        RETVAL=$?
        echo
        KIND="NMB"
        echo -n $"Starting $KIND services: "
        daemon /usr/local/samba/bin/nmbd $NMBDOPTIONS
        RETVAL2=$?
        echo
        KIND="Winbind"
        echo -n $"Starting $KIND services: "
        daemon /usr/local/samba/bin/winbindd
        RETVAL3=$?
        echo
        [ $RETVAL -eq 0 -a $RETVAL2 -eq 0 -a $RETVAL3 -eq 0 ] && touch /var/lock/subsys/smb || \
           RETVAL=1
        return $RETVAL
}

        daemon /usr/local/samba/bin/winbindd

        daemon /usr/local/samba/bin/winbindd -B

stop() {
        KIND="SMB"
        echo -n $"Shutting down $KIND services: "
        killproc smbd
        RETVAL=$?
        echo
        KIND="NMB"
        echo -n $"Shutting down $KIND services: "
        killproc nmbd
        RETVAL2=$?
        echo
        KIND="Winbind"
        echo -n $"Shutting down $KIND services: "
        killproc winbindd
        RETVAL3=$?
        [ $RETVAL -eq 0 -a $RETVAL2 -eq 0 -a $RETVAL3 -eq 0 ] && rm -f /var/lock/subsys/smb
        echo ""
        return $RETVAL
}

##
## samba.server
##

if [ ! -d /usr/bin ]
then                    # /usr not mounted
        exit
fi

killproc() {            # kill the named process(es)
        pid=`/usr/bin/ps -e |
             /usr/bin/grep -w $1 |
             /usr/bin/sed -e 's/^  *//' -e 's/ .*//'`
        [ "$pid" != "" ] && kill $pid
}
 
# Start/stop processes required for samba server

case "$1" in

'start')
#
# Edit these lines to suit your installation (paths, workgroup, host)
#
echo Starting SMBD
   /usr/local/samba/bin/smbd -D -s \
	/usr/local/samba/smb.conf

echo Starting NMBD
   /usr/local/samba/bin/nmbd -D -l \
	/usr/local/samba/var/log -s /usr/local/samba/smb.conf

echo Starting Winbind Daemon
   /usr/local/samba/bin/winbindd
   ;;

'stop')
   killproc nmbd
   killproc smbd
   killproc winbindd
   ;;

*)
   echo "Usage: /etc/init.d/samba.server { start | stop }"
   ;;
esac

   /usr/local/samba/bin/winbindd

   /usr/local/samba/bin/winbindd -B

auth    required        /lib/security/pam_stack.so service=system-auth
account required        /lib/security/pam_stack.so service=system-auth

enable = no

enable = yes

auth       required     /lib/security/pam_listfile.so item=user sense=deny file=/etc/ftpusers onerr=succeed
auth       sufficient   /lib/security/pam_winbind.so
auth       required     /lib/security/pam_stack.so service=system-auth
auth       required     /lib/security/pam_shells.so
account    sufficient   /lib/security/pam_winbind.so
account    required     /lib/security/pam_stack.so service=system-auth
session    required     /lib/security/pam_stack.so service=system-auth

auth       required     /lib/security/pam_securetty.so
auth       sufficient   /lib/security/pam_winbind.so
auth       sufficient   /lib/security/pam_unix.so use_first_pass
auth       required     /lib/security/pam_stack.so service=system-auth
auth       required     /lib/security/pam_nologin.so
account    sufficient   /lib/security/pam_winbind.so
account    required     /lib/security/pam_stack.so service=system-auth
password   required     /lib/security/pam_stack.so service=system-auth
session    required     /lib/security/pam_stack.so service=system-auth
session    optional     /lib/security/pam_console.so

#
#ident	"@(#)pam.conf	1.14	99/09/16 SMI"
#
# Copyright (c) 1996-1999, Sun Microsystems, Inc.
# All Rights Reserved.
#
# PAM configuration
#
# Authentication management
#
login   auth required   /usr/lib/security/pam_winbind.so
login	auth required 	/usr/lib/security/$ISA/pam_unix.so.1 try_first_pass 
login	auth required 	/usr/lib/security/$ISA/pam_dial_auth.so.1 try_first_pass 
#
rlogin  auth sufficient /usr/lib/security/pam_winbind.so
rlogin  auth sufficient /usr/lib/security/$ISA/pam_rhosts_auth.so.1
rlogin	auth required 	/usr/lib/security/$ISA/pam_unix.so.1 try_first_pass
#
dtlogin auth sufficient /usr/lib/security/pam_winbind.so
dtlogin	auth required 	/usr/lib/security/$ISA/pam_unix.so.1 try_first_pass
#
rsh	auth required	/usr/lib/security/$ISA/pam_rhosts_auth.so.1
other   auth sufficient /usr/lib/security/pam_winbind.so
other	auth required	/usr/lib/security/$ISA/pam_unix.so.1 try_first_pass
#
# Account management
#
login   account sufficient      /usr/lib/security/pam_winbind.so
login	account requisite	/usr/lib/security/$ISA/pam_roles.so.1 
login	account required	/usr/lib/security/$ISA/pam_unix.so.1 
#
dtlogin account sufficient      /usr/lib/security/pam_winbind.so
dtlogin	account requisite	/usr/lib/security/$ISA/pam_roles.so.1 
dtlogin	account required	/usr/lib/security/$ISA/pam_unix.so.1 
#
other   account sufficient      /usr/lib/security/pam_winbind.so
other	account requisite	/usr/lib/security/$ISA/pam_roles.so.1 
other	account required	/usr/lib/security/$ISA/pam_unix.so.1 
#
# Session management
#
other	session required	/usr/lib/security/$ISA/pam_unix.so.1 
#
# Password management
#
#other   password sufficient     /usr/lib/security/pam_winbind.so
other	password required	/usr/lib/security/$ISA/pam_unix.so.1 
dtsession auth required	/usr/lib/security/$ISA/pam_unix.so.1
#
# Support for Kerberos V5 authentication (uncomment to use Kerberos)
#
#rlogin	auth optional	/usr/lib/security/$ISA/pam_krb5.so.1 try_first_pass
#login	auth optional	/usr/lib/security/$ISA/pam_krb5.so.1 try_first_pass
#dtlogin	auth optional	/usr/lib/security/$ISA/pam_krb5.so.1 try_first_pass
#other	auth optional	/usr/lib/security/$ISA/pam_krb5.so.1 try_first_pass
#dtlogin	account optional /usr/lib/security/$ISA/pam_krb5.so.1
#other	account optional /usr/lib/security/$ISA/pam_krb5.so.1
#other	session optional /usr/lib/security/$ISA/pam_krb5.so.1
#other	password optional /usr/lib/security/$ISA/pam_krb5.so.1 try_first_pass

14.6. Limitations

Winbind has a number of limitations in its current released version that we hope to overcome in future releases:

• Winbind is currently only available for the Linux, Solaris and IRIX operating systems, although ports to other operating systems are certainly possible. For such ports to be feasible, we require the C library of the target operating system to support the Name Service Switch and Pluggable Authentication Modules systems. This is becoming more common as NSS and PAM gain support among UNIX vendors.

• The mappings of Windows NT RIDs to UNIX ids is not made algorithmically and depends on the order in which unmapped users or groups are seen by winbind. It may be difficult to recover the mappings of rid to UNIX id mapping if the file containing this information is corrupted or destroyed.

• Currently the winbind PAM module does not take into account possible workstation and logon time restrictions that may be been set for Windows NT users, this is instead up to the PDC to enforce.

14.7. Conclusion

The winbind system, through the use of the Name Service Switch, Pluggable Authentication Modules, and appropriate Microsoft RPC calls have allowed us to provide seamless integration of Microsoft Windows NT domain users on a UNIX system. The result is a great reduction in the administrative cost of running a mixed UNIX and NT network.

15.1. Configuring Samba Share Access Controls

This section deals with how to configure Samba per share access control restrictions. By default samba sets no restrictions on the share itself. Restrictions on the share itself can be set on MS Windows NT4/200x/XP shares. This can be a very effective way to limit who can connect to a share. In the absence of specific restrictions the default setting is to allow the global user Everyone Full Control (ie: Full control, Change and Read).

At this time Samba does NOT provide a tool for configuring access control setting on the Share itself. Samba does have the capacity to store and act on access control settings, but the only way to create those settings is to use either the NT4 Server Manager or the Windows 200x MMC for Computer Management.

Samba stores the per share access control settings in a file called share_info.tdb. The location of this file on your system will depend on how samba was compiled. The default location for samba's tdb files is under /usr/local/samba/var. If the tdbdump utility has been compiled and installed on your system then you can examine the contents of this file by: tdbdump share_info.tdb.

---

15.1.1. Share Permissions Management

The best tool for the task is platform dependant. Choose the best tool for your environmemt.

---

15.1.1.1. Windows NT4 Workstation/Server

The tool you need to use to manage share permissions on a Samba server is the NT Server Manager. Server Manager is shipped with Windows NT4 Server products but not with Windows NT4 Workstation. You can obtain the NT Server Manager for MS Windows NT4 Workstation from Microsoft - see details below.

Instructions

1. Launch the NT4 Server Manager, click on the Samba server you want to administer, then from the menu select Computer, then click on the Shared Directories entry.

2. Now click on the share that you wish to manage, then click on the Properties tab, next click on the Permissions tab. Now you can Add or change access control settings as you wish.

---

15.1.1.2. Windows 200x/XP

On MS Windows NT4/200x/XP system access control lists on the share itself are set using native tools, usually from filemanager. For example, in Windows 200x: right click on the shared folder, then select 'Sharing', then click on 'Permissions'. The default Windows NT4/200x permission allows Everyone Full Control on the Share.

MS Windows 200x and later all comes with a tool called the 'Computer Management' snap-in for the Microsoft Management Console (MMC). This tool is located by clicking on Control Panel -> Administrative Tools -> Computer Management.

Instructions

1. After launching the MMC with the Computer Management snap-in, click on the menu item 'Action', select 'Connect to another computer'. If you are not logged onto a domain you will be prompted to enter a domain login user identifier and a password. This will authenticate you to the domain. If you where already logged in with administrative privilidge this step is not offered.

2. If the Samba server is not shown in the Select Computer box, then type in the name of the target Samba server in the field 'Name:'. Now click on the [+] next to 'System Tools', then on the [+] next to 'Shared Folders' in the left panel.

3. Now in the right panel, double-click on the share you wish to set access control permissions on. Then click on the tab 'Share Permissions'. It is now possible to add access control entities to the shared folder. Do NOT forget to set what type of access (full control, change, read) you wish to assign for each entry.

Be careful. If you take away all permissions from the Everyone user without removing this user then effectively no user will be able to access the share. This is a result of what is known as ACL precidence. ie: Everyone with NO ACCESS means that MaryK who is part of the group Everyone will have no access even if this user is given explicit full control access.

15.2. Remote Server Administration

How do I get 'User Manager' and 'Server Manager'?

Since I don't need to buy an NT4 Server, how do I get the 'User Manager for Domains', the 'Server Manager'?

Microsoft distributes a version of these tools called nexus for installation on Windows 9x / Me systems. The tools set includes:

• Server Manager

• User Manager for Domains

• Event Viewer

Click here to download the archived file ftp://ftp.microsoft.com/Softlib/MSLFILES/NEXUS.EXE

The Windows NT 4.0 version of the 'User Manager for Domains' and 'Server Manager' are available from Microsoft via ftp from ftp://ftp.microsoft.com/Softlib/MSLFILES/SRVTOOLS.EXE

15.3. Network Logon Script Magic

This section needs work. Volunteer contributions most welcome. Please send your patches or updates to John Terpstra.

16.1. Creating and Managing System Policies

Under MS Windows platforms, particularly those following the release of MS Windows NT4 and MS Windows 95) it is possible to create a type of file that would be placed in the NETLOGON share of a domain controller. As the client logs onto the network this file is read and the contents initiate changes to the registry of the client machine. This file allows changes to be made to those parts of the registry that affect users, groups of users, or machines.

For MS Windows 9x/Me this file must be called Config.POL and may be generated using a tool called poledit.exe, better known as the Policy Editor. The policy editor was provided on the Windows 98 installation CD, but dissappeared again with the introduction of MS Windows Me (Millenium Edition). From comments from MS Windows network administrators it would appear that this tool became a part of the MS Windows Me Resource Kit.

MS Windows NT4 Server products include the System Policy Editor under the Start -> Programs -> Administrative Tools menu item. For MS Windows NT4 and later clients this file must be called NTConfig.POL.

New with the introduction of MS Windows 2000 was the Microsoft Management Console or MMC. This tool is the new wave in the ever changing landscape of Microsoft methods for management of network access and security. Every new Microsoft product or technology seems to obsolete the old rules and to introduce newer and more complex tools and methods. To Microsoft's credit though, the MMC does appear to be a step forward, but improved functionality comes at a great price.

Before embarking on the configuration of network and system policies it is highly advisable to read the documentation available from Microsoft's web site regarding Implementing Profiles and Policies in Windows NT 4.0 from http://www.microsoft.com/ntserver/management/deployment/planguide/prof_policies.asp available from Microsoft. There are a large number of documents in addition to this old one that should also be read and understood. Try searching on the Microsoft web site for "Group Policies".

What follows is a very discussion with some helpful notes. The information provided here is incomplete - you are warned.

---

16.1.3. MS Windows 200x / XP Professional Policies

Windows NT4 System policies allows setting of registry parameters specific to users, groups and computers (client workstations) that are members of the NT4 style domain. Such policy file will work with MS Windows 2000 / XP clients also.

New to MS Windows 2000 Microsoft introduced a new style of group policy that confers a superset of capabilities compared with NT4 style policies. Obviously, the tool used to create them is different, and the mechanism for implementing them is much changed.

The older NT4 style registry based policies are known as Administrative Templates in MS Windows 2000/XP Group Policy Objects (GPOs). The later includes ability to set various security configurations, enforce Internet Explorer browser settings, change and redirect aspects of the users' desktop (including: the location of My Documents files (directory), as well as intrinsics of where menu items will appear in the Start menu). An additional new feature is the ability to make available particular software Windows applications to particular users and/or groups.

Remember: NT4 policy files are named NTConfig.POL and are stored in the root of the NETLOGON share on the domain controllers. A Windows NT4 user enters a username, a password and selects the domain name to which the logon will attempt to take place. During the logon process the client machine reads the NTConfig.POL file from the NETLOGON share on the authenticating server, modifies the local registry values according to the settings in this file.

Windows 2K GPOs are very feature rich. They are NOT stored in the NETLOGON share, rather part of a Windows 200x policy file is stored in the Active Directory itself and the other part is stored in a shared (and replicated) volume called the SYSVOL folder. This folder is present on all Active Directory domain controllers. The part that is stored in the Active Directory itself is called the group policy container (GPC), and the part that is stored in the replicated share called SYSVOL is known as the group policy template (GPT).

With NT4 clients the policy file is read and executed upon only aas each user log onto the network. MS Windows 200x policies are much more complex - GPOs are processed and applied at client machine startup (machine specific part) and when the user logs onto the network the user specific part is applied. In MS Windows 200x style policy management each machine and/or user may be subject to any number of concurently applicable (and applied) policy sets (GPOs). Active Directory allows the administrator to also set filters over the policy settings. No such equivalent capability exists with NT4 style policy files.

---

16.1.3.1. Administration of Win2K / XP Policies

Instructions

Instead of using the tool called "The System Policy Editor", commonly called Poledit (from the executable name poledit.exe), GPOs are created and managed using a Microsoft Management Console (MMC) snap-in as follows:

1. Go to the Windows 200x / XP menu Start->Programs->Administrative Tools and select the MMC snap-in called "Active Directory Users and Computers"

2. Select the domain or organizational unit (OU) that you wish to manage, then right click to open the context menu for that object, select the properties item.

3. Now left click on the Group Policy tab, then left click on the New tab. Type a name for the new policy you will create.

4. Now left click on the Edit tab to commence the steps needed to create the GPO.

All policy configuration options are controlled through the use of policy administrative templates. These files have a .adm extension, both in NT4 as well as in Windows 200x / XP. Beware however, since the .adm files are NOT interchangible across NT4 and Windows 200x. The later introduces many new features as well as extended definition capabilities. It is well beyond the scope of this documentation to explain how to program .adm files, for that the adminsitrator is referred to the Microsoft Windows Resource Kit for your particular version of MS Windows.

The MS Windows 2000 Resource Kit contains a tool called gpolmig.exe. This tool can be used to migrate an NT4 NTConfig.POL file into a Windows 200x style GPO. Be VERY careful how you use this powerful tool. Please refer to the resource kit manuals for specific usage information.

16.2. Managing Account/User Policies

Policies can define a specific user's settings or the settings for a group of users. The resulting policy file contains the registry settings for all users, groups, and computers that will be using the policy file. Separate policy files for each user, group, or computer are not not necessary.

If you create a policy that will be automatically downloaded from validating domain controllers, you should name the file NTconfig.POL. As system administrator, you have the option of renaming the policy file and, by modifying the Windows NT-based workstation, directing the computer to update the policy from a manual path. You can do this by either manually changing the registry or by using the System Policy Editor. This path can even be a local path such that each machine has its own policy file, but if a change is necessary to all machines, this change must be made individually to each workstation.

When a Windows NT4/200x/XP machine logs onto the network the NETLOGON share on the authenticating domain controller for the presence of the NTConfig.POL file. If one exists it is downloaded, parsed and then applied to the user's part of the registry.

MS Windows 200x/XP clients that log onto an MS Windows Active Directory security domain may additionally, acquire policy settings through Group Policy Objects (GPOs) that are defined and stored in Active Directory itself. The key benefit of using AS GPOs is that they impose no registry tatooing effect. This has considerable advanage compared with the use of NTConfig.POL (NT4) style policy updates.

Inaddition to user access controls that may be imposed or applied via system and/or group policies in a manner that works in conjunction with user profiles, the user management environment under MS Windows NT4/200x/XP allows per domain as well as per user account restrictions to be applied. Common restrictions that are frequently used includes:

17.1. Roaming Profiles

NOTE! Roaming profiles support is different for Win9x / Me and Windows NT4/200x.

Before discussing how to configure roaming profiles, it is useful to see how Windows 9x / Me and Windows NT4/200x clients implement these features.

Windows 9x / Me clients send a NetUserGetInfo request to the server to get the user's profiles location. However, the response does not have room for a separate profiles location field, only the user's home share. This means that Win9X/Me profiles are restricted to being stored in the user's home directory.

Windows NT4/200x clients send a NetSAMLogon RPC request, which contains many fields, including a separate field for the location of the user's profiles.

---

17.1.1. Samba Configuration for Profile Handling

This section documents how to configure Samba for MS Windows client profile support.

---

17.1.1.1. NT4/200x User Profiles

To support Windowns NT4/200x clients, in the [global] section of smb.conf set the following (for example):

	logon path = \\profileserver\profileshare\profilepath\%U\moreprofilepath

	This is typically implemented like:

		logon path = \\%L\Profiles\%u

	where:
		%L translates to the name of the Samba server
		%u translates to the user name

The default for this option is \\%N\%U\profile, namely \\sambaserver\username\profile. The \\N%\%U service is created automatically by the [homes] service. If you are using a samba server for the profiles, you _must_ make the share specified in the logon path browseable. Please refer to the man page for smb.conf in respect of the different symantics of %L and %N, as well as %U and %u.

MS Windows NT/2K clients at times do not disconnect a connection to a server between logons. It is recommended to NOT use the homes meta-service name as part of the profile share path.

---

17.1.1.2. Windows 9x / Me User Profiles

To support Windows 9x / Me clients, you must use the "logon home" parameter. Samba has now been fixed so that "net use /home" now works as well, and it, too, relies on the "logon home" parameter.

By using the logon home parameter, you are restricted to putting Win9x / Me profiles in the user's home directory. But wait! There is a trick you can use. If you set the following in the [global] section of your smb.conf file:

	logon home = \\%L\%U\.profiles

then your Windows 9x / Me clients will dutifully put their clients in a subdirectory of your home directory called .profiles (thus making them hidden).

Not only that, but 'net use/home' will also work, because of a feature in Windows 9x / Me. It removes any directory stuff off the end of the home directory area and only uses the server and share portion. That is, it looks like you specified \\%L\%U for "logon home".

---

17.1.1.3. Mixed Windows 9x / Me and Windows NT4/200x User Profiles

You can support profiles for both Win9X and WinNT clients by setting both the "logon home" and "logon path" parameters. For example:

	logon home = \\%L\%u\.profiles
	logon path = \\%L\profiles\%u

---

17.1.2. Windows Client Profile Configuration Information

17.1.2.1. Windows 9x / Me Profile Setup

When a user first logs in on Windows 9X, the file user.DAT is created, as are folders "Start Menu", "Desktop", "Programs" and "Nethood". These directories and their contents will be merged with the local versions stored in c:\windows\profiles\username on subsequent logins, taking the most recent from each. You will need to use the [global] options "preserve case = yes", "short preserve case = yes" and "case sensitive = no" in order to maintain capital letters in shortcuts in any of the profile folders.

The user.DAT file contains all the user's preferences. If you wish to enforce a set of preferences, rename their user.DAT file to user.MAN, and deny them write access to this file.

1. On the Windows 9x / Me machine, go to Control Panel -> Passwords and select the User Profiles tab. Select the required level of roaming preferences. Press OK, but do _not_ allow the computer to reboot.

2. On the Windows 9x / Me machine, go to Control Panel -> Network -> Client for Microsoft Networks -> Preferences. Select 'Log on to NT Domain'. Then, ensure that the Primary Logon is 'Client for Microsoft Networks'. Press OK, and this time allow the computer to reboot.

Under Windows 9x / Me Profiles are downloaded from the Primary Logon. If you have the Primary Logon as 'Client for Novell Networks', then the profiles and logon script will be downloaded from your Novell Server. If you have the Primary Logon as 'Windows Logon', then the profiles will be loaded from the local machine - a bit against the concept of roaming profiles, it would seem!

You will now find that the Microsoft Networks Login box contains [user, password, domain] instead of just [user, password]. Type in the samba server's domain name (or any other domain known to exist, but bear in mind that the user will be authenticated against this domain and profiles downloaded from it, if that domain logon server supports it), user name and user's password.

Once the user has been successfully validated, the Windows 9x / Me machine will inform you that 'The user has not logged on before' and asks you if you wish to save the user's preferences? Select 'yes'.

Once the Windows 9x / Me client comes up with the desktop, you should be able to examine the contents of the directory specified in the "logon path" on the samba server and verify that the "Desktop", "Start Menu", "Programs" and "Nethood" folders have been created.

These folders will be cached locally on the client, and updated when the user logs off (if you haven't made them read-only by then). You will find that if the user creates further folders or short-cuts, that the client will merge the profile contents downloaded with the contents of the profile directory already on the local client, taking the newest folders and short-cuts from each set.

If you have made the folders / files read-only on the samba server, then you will get errors from the Windows 9x / Me machine on logon and logout, as it attempts to merge the local and the remote profile. Basically, if you have any errors reported by the Windows 9x / Me machine, check the Unix file permissions and ownership rights on the profile directory contents, on the samba server.

If you have problems creating user profiles, you can reset the user's local desktop cache, as shown below. When this user then next logs in, they will be told that they are logging in "for the first time".

1. instead of logging in under the [user, password, domain] dialog, press escape.

2. run the regedit.exe program, and look in:

   HKEY_LOCAL_MACHINE\Windows\CurrentVersion\ProfileList

   you will find an entry, for each user, of ProfilePath. Note the contents of this key (likely to be c:\windows\profiles\username), then delete the key ProfilePath for the required user. [Exit the registry editor].

3. WARNING - before deleting the contents of the directory listed in the ProfilePath (this is likely to be c:\windows\profiles\username), ask them if they have any important files stored on their desktop or in their start menu. Delete the contents of the directory ProfilePath (making a backup if any of the files are needed).

   This will have the effect of removing the local (read-only hidden system file) user.DAT in their profile directory, as well as the local "desktop", "nethood", "start menu" and "programs" folders.

4. search for the user's .PWL password-caching file in the c:\windows directory, and delete it.

5. log off the windows 9x / Me client.

6. check the contents of the profile path (see "logon path" described above), and delete the user.DAT or user.MAN file for the user, making a backup if required.

If all else fails, increase samba's debug log levels to between 3 and 10, and / or run a packet trace program such as ethereal or netmon.exe, and look for error messages.

If you have access to an Windows NT4/200x server, then first set up roaming profiles and / or netlogons on the Windows NT4/200x server. Make a packet trace, or examine the example packet traces provided with Windows NT4/200x server, and see what the differences are with the equivalent samba trace.

---

17.1.2.2. Windows NT4 Workstation

When a user first logs in to a Windows NT Workstation, the profile NTuser.DAT is created. The profile location can be now specified through the "logon path" parameter.

There is a parameter that is now available for use with NT Profiles: "logon drive". This should be set to H: or any other drive, and should be used in conjunction with the new "logon home" parameter.

The entry for the NT4 profile is a _directory_ not a file. The NT help on profiles mentions that a directory is also created with a .PDS extension. The user, while logging in, must have write permission to create the full profile path (and the folder with the .PDS extension for those situations where it might be created.)

In the profile directory, Windows NT4 creates more folders than Windows 9x / Me. It creates "Application Data" and others, as well as "Desktop", "Nethood", "Start Menu" and "Programs". The profile itself is stored in a file NTuser.DAT. Nothing appears to be stored in the .PDS directory, and its purpose is currently unknown.

You can use the System Control Panel to copy a local profile onto a samba server (see NT Help on profiles: it is also capable of firing up the correct location in the System Control Panel for you). The NT Help file also mentions that renaming NTuser.DAT to NTuser.MAN turns a profile into a mandatory one.

The case of the profile is significant. The file must be called NTuser.DAT or, for a mandatory profile, NTuser.MAN.

---

17.1.2.3. Windows 2000/XP Professional

You must first convert the profile from a local profile to a domain profile on the MS Windows workstation as follows:

• Log on as the LOCAL workstation administrator.

• Right click on the 'My Computer' Icon, select 'Properties'

• Click on the 'User Profiles' tab

• Select the profile you wish to convert (click on it once)

• Click on the button 'Copy To'

• In the "Permitted to use" box, click on the 'Change' button.

• Click on the 'Look in" area that lists the machine name, when you click here it will open up a selection box. Click on the domain to which the profile must be accessible.

  You will need to log on if a logon box opens up. Eg: In the connect as: MIDEARTH\root, password: mypassword.

• To make the profile capable of being used by anyone select 'Everyone'

• Click OK. The Selection box will close.

• Now click on the 'Ok' button to create the profile in the path you nominated.

Done. You now have a profile that can be editted using the samba-3.0.0 profiles tool.

Under NT/2K the use of mandotory profiles forces the use of MS Exchange storage of mail data. That keeps desktop profiles usable.

This is a security check new to Windows XP (or maybe only Windows XP service pack 1). It can be disabled via a group policy in Active Directory. The policy is: "Computer Configuration\Administrative Templates\System\User Profiles\Do not check for user ownership of Roaming Profile Folders" ...and it should be set to "Enabled". Does the new version of samba have an Active Directory analogue? If so, then you may be able to set the policy through this. If you cannot set group policies in samba, then you may be able to set the policy locally on each machine. If you want to try this, then do the following (N.B. I don't know for sure that this will work in the same way as a domain group policy): On the XP workstation log in with an Administrator account. Click: "Start", "Run" Type: "mmc" Click: "OK" A Microsoft Management Console should appear. Click: File, "Add/Remove Snap-in...", "Add" Double-Click: "Group Policy" Click: "Finish", "Close" Click: "OK" In the "Console Root" window: Expand: "Local Computer Policy", "Computer Configuration", "Administrative Templates", "System", "User Profiles" Double-Click: "Do not check for user ownership of Roaming Profile Folders" Select: "Enabled" Click: OK" Close the whole console. You do not need to save the settings (this refers to the console settings rather than the policies you have changed). Reboot

17.2. Mandatory profiles

A Mandatory Profile is a profile that the user does NOT have the ability to overwrite. During the user's session it may be possible to change the desktop environment, but as the user logs out all changes made will be lost. If it is desired to NOT allow the user any ability to change the desktop environment then this must be done through policy settings. See previous chapter.

Under NO circumstances should the profile directory (or it's contents) be made read-only as this may render the profile un-usable.

For MS Windows NT4/200x/XP the above method can be used to create mandatory profiles also. To convert a group profile into a mandatory profile simply locate the NTUser.DAT file in the copied profile and rename it to NTUser.MAN.

For MS Windows 9x / Me it is the User.DAT file that must be renamed to User.MAN to affect a mandatory profile.

17.3. Creating/Managing Group Profiles

Most organisations are arranged into departments. There is a nice benenfit in this fact since usually most users in a department will require the same desktop applications and the same desktop layout. MS Windows NT4/200x/XP will allow the use of Group Profiles. A Group Profile is a profile that is created firstly using a template (example) user. Then using the profile migration tool (see above) the profile is assigned access rights for the user group that needs to be given access to the group profile.

The next step is rather important. PLEASE NOTE: Instead of assigning a group profile to users (ie: Using User Manager) on a "per user" basis, the group itself is assigned the now modified profile.

Be careful with group profiles, if the user who is a member of a group also has a personal profile, then the result will be a fusion (merge) of the two.

17.4. Default Profile for Windows Users

MS Windows 9x / Me and NT4/200x/XP will use a default profile for any user for whom a profile does not already exist. Armed with a knowledge of where the default profile is located on the Windows workstation, and knowing which registry keys affect the path from which the default profile is created, it is possible to modify the default profile to one that has been optimised for the site. This has significant administrative advantages.

	HKEY_LOCAL_MACHINE\SYSTEM\Software\Microsoft\Windows NT\CurrentVersion\winlogon\
	"DeleteRoamingCache"=dword:00000001

        HKEY_CURRENT_USER
                \Software
                        \Microsoft
                                \Windows
                                        \CurrentVersion
                                                \Explorer
                                                        \User Shell Folders\

        Name            Default Value
        --------------  -----------------------------------------
        AppData         %USERPROFILE%\Application Data
        Desktop         %USERPROFILE%\Desktop
        Favorites       %USERPROFILE%\Favorites
        NetHood         %USERPROFILE%\NetHood
        PrintHood       %USERPROFILE%\PrintHood
        Programs        %USERPROFILE%\Start Menu\Programs
        Recent          %USERPROFILE%\Recent
        SendTo          %USERPROFILE%\SendTo
        Start Menu      %USERPROFILE%\Start Menu
        Startup         %USERPROFILE%\Start Menu\Programs\Startup
        

	HKEY_LOCAL_MACHINE
		\SOFTWARE
			\Microsoft
				\Windows
					\CurrentVersion
						\Explorer
							\User Shell Folders

	Common Desktop		%SystemRoot%\Profiles\All Users\Desktop
	Common Programs		%SystemRoot%\Profiles\All Users\Programs
	Common Start Menu	%SystemRoot%\Profiles\All Users\Start Menu
	Common Startu	p	%SystemRoot%\Profiles\All Users\Start Menu\Progams\Startup

---

17.4.3. MS Windows 200x/XP

MS Windows XP Home Edition does use default per user profiles, but can not participate in domain security, can not log onto an NT/ADS style domain, and thus can obtain the profile only from itself. While there are benefits in doing this the beauty of those MS Windows clients that CAN participate in domain logon processes allows the administrator to create a global default profile and to enforce it through the use of Group Policy Objects (GPOs).

When a new user first logs onto MS Windows 200x/XP machine the default profile is obtained from C:\Documents and Settings\Default User. The administrator can modify (or change the contents of this location and MS Windows 200x/XP will gladly user it. This is far from the optimum arrangement since it will involve copying a new default profile to every MS Windows 200x/XP client workstation.

When MS Windows 200x/XP participate in a domain security context, and if the default user profile is not found, then the client will search for a default profile in the NETLOGON share of the authenticating server. ie: In MS Windows parlance: %LOGONSERVER%\NETLOGON\Default User and if one exits there it will copy this to the workstation to the C:\Documents and Settings\ under the Windows login name of the user.

This path translates, in Samba parlance, to the smb.conf [NETLOGON] share. The directory should be created at the root of this share and msut be called Default Profile.

If a default profile does not exist in this location then MS Windows 200x/XP will use the local default profile.

On loging out, the users' desktop profile will be stored to the location specified in the registry settings that pertain to the user. If no specific policies have been created, or passed to the client during the login process (as Samba does automatically), then the user's profile will be written to the local machine only under the path C:\Documents and Settings\%USERNAME%.

Those wishing to modify the default behaviour can do so through up to three methods:

• Modify the registry keys on the local machine manually and place the new default profile in the NETLOGON share root - NOT recommended as it is maintenance intensive.

• Create an NT4 style NTConfig.POL file that specified this behaviour and locate this file in the root of the NETLOGON share along with the new default profile.

• Create a GPO that enforces this through Active Directory, and place the new default profile in the NETLOGON share.

The Registry Hive key that affects the behaviour of folders that are part of the default user profile are controlled by entries on Windows 200x/XP is:

	HKEY_CURRENT_USER
		\Software
			\Microsoft
				\Windows
					\CurrentVersion
						\Explorer
							\User Shell Folders\

The above hive key contains a list of automatically managed folders. The default entries are:

	Name		Default Value
	--------------	-----------------------------------------
	AppData		%USERPROFILE%\Application Data
	Cache		%USERPROFILE%\Local Settings\Temporary Internet Files
	Cookies		%USERPROFILE%\Cookies
	Desktop		%USERPROFILE%\Desktop
	Favorites	%USERPROFILE%\Favorites
	History		%USERPROFILE%\Local Settings\History
	Local AppData	%USERPROFILE%\Local Settings\Application Data
	Local Settings	%USERPROFILE%\Local Settings
	My Pictures	%USERPROFILE%\My Documents\My Pictures
	NetHood		%USERPROFILE%\NetHood
	Personal	%USERPROFILE%\My Documents
	PrintHood	%USERPROFILE%\PrintHood
	Programs	%USERPROFILE%\Start Menu\Programs
	Recent		%USERPROFILE%\Recent
	SendTo		%USERPROFILE%\SendTo
	Start Menu	%USERPROFILE%\Start Menu
	Startup		%USERPROFILE%\Start Menu\Programs\Startup
	Templates	%USERPROFILE%\Templates
	

There is also an entry called "Default" that has no value set. The default entry is of type REG_SZ, all the others are of type REG_EXPAND_SZ.

It makes a huge difference to the speed of handling roaming user profiles if all the folders are stored on a dedicated location on a network server. This means that it will NOT be necessary to write Outlook PST file over the network for every login and logout.

To set this to a network location you could use the following examples:

	%LOGONSERVER%\%USERNAME%\Default Folders

This would store the folders in the user's home directory under a directory called "Default Folders" You could also use:

	\\SambaServer\FolderShare\%USERNAME%

in which case the default folders will be stored in the server named SambaServer in the share called FolderShare under a directory that has the name of the MS Windows user as seen by the Linux/Unix file system.

Please note that once you have created a default profile share, you MUST migrate a user's profile (default or custom) to it.

MS Windows 200x/XP profiles may be Local or Roaming. A roaming profile will be cached locally unless the following registry key is created:

	HKEY_LOCAL_MACHINE\SYSTEM\Software\Microsoft\Windows NT\CurrentVersion\winlogon\
	"DeleteRoamingCache"=dword:00000001

In which case, the local cache copy will be deleted on logout.

18.1. Samba and PAM

A number of Unix systems (eg: Sun Solaris), as well as the xxxxBSD family and Linux, now utilize the Pluggable Authentication Modules (PAM) facility to provide all authentication, authorization and resource control services. Prior to the introduction of PAM, a decision to use an alternative to the system password database (/etc/passwd) would require the provision of alternatives for all programs that provide security services. Such a choice would involve provision of alternatives to such programs as: login, passwd, chown, etc.

PAM provides a mechanism that disconnects these security programs from the underlying authentication/authorization infrastructure. PAM is configured either through one file /etc/pam.conf (Solaris), or by editing individual files that are located in /etc/pam.d.

If the PAM authentication module (loadable link library file) is located in the default location then it is not necessary to specify the path. In the case of Linux, the default location is /lib/security. If the module is located other than default then the path may be specified as: auth required /other_path/pam_strange_module.so

The following is an example /etc/pam.d/login configuration file. This example had all options been uncommented is probably not usable as it stacks many conditions before allowing successful completion of the login process. Essentially all conditions can be disabled by commenting them out except the calls to pam_pwdb.so.

	#%PAM-1.0
	# The PAM configuration file for the `login' service
	#
	auth 		required	pam_securetty.so
	auth 		required	pam_nologin.so
	# auth 		required	pam_dialup.so
	# auth 		optional	pam_mail.so
	auth		required	pam_pwdb.so shadow md5
	# account    	requisite  	pam_time.so
	account		required	pam_pwdb.so
	session		required	pam_pwdb.so
	# session 	optional	pam_lastlog.so
	# password   	required   	pam_cracklib.so retry=3
	password	required	pam_pwdb.so shadow md5

PAM allows use of replacable modules. Those available on a sample system include:

$/bin/ls /lib/security

	pam_access.so    pam_ftp.so          pam_limits.so     
	pam_ncp_auth.so  pam_rhosts_auth.so  pam_stress.so     
	pam_cracklib.so  pam_group.so        pam_listfile.so   
	pam_nologin.so   pam_rootok.so       pam_tally.so      
	pam_deny.so      pam_issue.so        pam_mail.so       
	pam_permit.so    pam_securetty.so    pam_time.so       
	pam_dialup.so    pam_lastlog.so      pam_mkhomedir.so  
	pam_pwdb.so      pam_shells.so       pam_unix.so       
	pam_env.so       pam_ldap.so         pam_motd.so       
	pam_radius.so    pam_smbpass.so      pam_unix_acct.so  
	pam_wheel.so     pam_unix_auth.so    pam_unix_passwd.so
	pam_userdb.so    pam_warn.so         pam_unix_session.so

The following example for the login program replaces the use of the pam_pwdb.so module which uses the system password database (/etc/passwd, /etc/shadow, /etc/group) with the module pam_smbpass.so which uses the Samba database which contains the Microsoft MD4 encrypted password hashes. This database is stored in either /usr/local/samba/private/smbpasswd, /etc/samba/smbpasswd, or in /etc/samba.d/smbpasswd, depending on the Samba implementation for your Unix/Linux system. The pam_smbpass.so module is provided by Samba version 2.2.1 or later. It can be compiled by specifying the --with-pam_smbpass options when running Samba's configure script. For more information on the pam_smbpass module, see the documentation in the source/pam_smbpass directory of the Samba source distribution.

	#%PAM-1.0
	# The PAM configuration file for the `login' service
	#
	auth		required	pam_smbpass.so nodelay
	account		required	pam_smbpass.so nodelay
	session		required	pam_smbpass.so nodelay
	password	required	pam_smbpass.so nodelay

The following is the PAM configuration file for a particular Linux system. The default condition uses pam_pwdb.so.

	#%PAM-1.0
	# The PAM configuration file for the `samba' service
	#
	auth       required     pam_pwdb.so nullok nodelay shadow audit
	account    required     pam_pwdb.so audit nodelay
	session    required     pam_pwdb.so nodelay
	password   required     pam_pwdb.so shadow md5

In the following example the decision has been made to use the smbpasswd database even for basic samba authentication. Such a decision could also be made for the passwd program and would thus allow the smbpasswd passwords to be changed using the passwd program.

	#%PAM-1.0
	# The PAM configuration file for the `samba' service
	#
	auth       required     pam_smbpass.so nodelay
	account    required     pam_pwdb.so audit nodelay
	session    required     pam_pwdb.so nodelay
	password   required     pam_smbpass.so nodelay smbconf=/etc/samba.d/smb.conf

PAM allows stacking of authentication mechanisms. It is also possible to pass information obtained within one PAM module through to the next module in the PAM stack. Please refer to the documentation for your particular system implementation for details regarding the specific capabilities of PAM in this environment. Some Linux implmentations also provide the pam_stack.so module that allows all authentication to be configured in a single central file. The pam_stack.so method has some very devoted followers on the basis that it allows for easier administration. As with all issues in life though, every decision makes trade-offs, so you may want examine the PAM documentation for further helpful information.

18.2. Distributed Authentication

The astute administrator will realize from this that the combination of pam_smbpass.so, winbindd, and a distributed passdb backend, such as ldap, will allow the establishment of a centrally managed, distributed user/password database that can also be used by all PAM (eg: Linux) aware programs and applications. This arrangement can have particularly potent advantages compared with the use of Microsoft Active Directory Service (ADS) in so far as reduction of wide area network authentication traffic.

18.3. PAM Configuration in smb.conf

There is an option in smb.conf called obey pam restrictions. The following is from the on-line help for this option in SWAT;

When Samba is configured to enable PAM support (i.e. --with-pam), this parameter will control whether or not Samba should obey PAM's account and session management directives. The default behavior is to use PAM for clear text authentication only and to ignore any account or session management. Note that Samba always ignores PAM for authentication in the case of encrypt passwords = yes. The reason is that PAM modules cannot support the challenge/response authentication mechanism needed in the presence of SMB password encryption.

Default: obey pam restrictions = no

19.1. Introduction and configuration

Since samba 3.0, samba supports stackable VFS(Virtual File System) modules. Samba passes each request to access the unix file system thru the loaded VFS modules. This chapter covers all the modules that come with the samba source and references to some external modules.

You may have problems to compile these modules, as shared libraries are compiled and linked in different ways on different systems. They currently have been tested against GNU/linux and IRIX.

To use the VFS modules, create a share similar to the one below. The important parameter is the vfs object parameter which must point to the exact pathname of the shared library objects. For example, to log all access to files and use a recycle bin:

       [audit]
                comment = Audited /data directory
                path = /data
                vfs object = /path/to/audit.so /path/to/recycle.so
                writeable = yes
                browseable = yes

The modules are used in the order they are specified.

Further documentation on writing VFS modules for Samba can be found in the Samba Developers Guide.

19.2. Included modules

19.3. VFS modules available elsewhere

This section contains a listing of various other VFS modules that have been posted but don't currently reside in the Samba CVS tree for one reason ot another (e.g. it is easy for the maintainer to have his or her own CVS tree).

No statemets about the stability or functionality any module should be implied due to its presence here.

20.1. Instructions

The Distributed File System (or Dfs) provides a means of separating the logical view of files and directories that users see from the actual physical locations of these resources on the network. It allows for higher availability, smoother storage expansion, load balancing etc. For more information about Dfs, refer to Microsoft documentation.

This document explains how to host a Dfs tree on a Unix machine (for Dfs-aware clients to browse) using Samba.

To enable SMB-based DFS for Samba, configure it with the --with-msdfs option. Once built, a Samba server can be made a Dfs server by setting the global boolean host msdfs parameter in the smb.conf file. You designate a share as a Dfs root using the share level boolean msdfs root parameter. A Dfs root directory on Samba hosts Dfs links in the form of symbolic links that point to other servers. For example, a symbolic link junction->msdfs:storage1\share1 in the share directory acts as the Dfs junction. When Dfs-aware clients attempt to access the junction link, they are redirected to the storage location (in this case, \\storage1\share1).

Dfs trees on Samba work with all Dfs-aware clients ranging from Windows 95 to 2000.

Here's an example of setting up a Dfs tree on a Samba server.

# The smb.conf file:
[global]
	netbios name = SAMBA
	host msdfs   = yes

[dfs]
	path = /export/dfsroot
	msdfs root = yes
	

In the /export/dfsroot directory we set up our dfs links to other servers on the network.

root# cd /export/dfsroot

root# chown root /export/dfsroot

root# chmod 755 /export/dfsroot

root# ln -s msdfs:storageA\\shareA linka

root# ln -s msdfs:serverB\\share,serverC\\share linkb

You should set up the permissions and ownership of the directory acting as the Dfs root such that only designated users can create, delete or modify the msdfs links. Also note that symlink names should be all lowercase. This limitation exists to have Samba avoid trying all the case combinations to get at the link name. Finally set up the symbolic links to point to the network shares you want, and start Samba.

Users on Dfs-aware clients can now browse the Dfs tree on the Samba server at \\samba\dfs. Accessing links linka or linkb (which appear as directories to the client) takes users directly to the appropriate shares on the network.

21.1. Name Resolution in a pure Unix/Linux world

The key configuration files covered in this section are:

• /etc/hosts

• /etc/resolv.conf

• /etc/host.conf

• /etc/nsswitch.conf

	127.0.0.1	localhost localhost.localdomain
	192.168.1.1	bigbox.caldera.com	bigbox	alias4box

	order hosts,bind
	multi on

	# /etc/nsswitch.conf
	#
	# Name Service Switch configuration file.
	#

	passwd:		compat
	# Alternative entries for password authentication are:
	# passwd:	compat files nis ldap winbind
	shadow:		compat
	group:		compat

	hosts:		files nis dns
	# Alternative entries for host name resolution are:
	# hosts:	files dns nis nis+ hesoid db compat ldap wins
	networks:	nis files dns

	ethers:		nis files
	protocols:	nis files
	rpc:		nis files
	services:	nis files

21.2. Name resolution as used within MS Windows networking

MS Windows networking is predicated about the name each machine is given. This name is known variously (and inconsistently) as the "computer name", "machine name", "networking name", "netbios name", "SMB name". All terms mean the same thing with the exception of "netbios name" which can apply also to the name of the workgroup or the domain name. The terms "workgroup" and "domain" are really just a simply name with which the machine is associated. All NetBIOS names are exactly 16 characters in length. The 16th character is reserved. It is used to store a one byte value that indicates service level information for the NetBIOS name that is registered. A NetBIOS machine name is therefore registered for each service type that is provided by the client/server.

The following are typical NetBIOS name/service type registrations:

	Unique NetBIOS Names:
		MACHINENAME<00>	= Server Service is running on MACHINENAME
		MACHINENAME<03> = Generic Machine Name (NetBIOS name)
		MACHINENAME<20> = LanMan Server service is running on MACHINENAME
		WORKGROUP<1b> = Domain Master Browser

	Group Names:
		WORKGROUP<03> = Generic Name registered by all members of WORKGROUP
		WORKGROUP<1c> = Domain Controllers / Netlogon Servers
		WORKGROUP<1d> = Local Master Browsers
		WORKGROUP<1e> = Internet Name Resolvers

It should be noted that all NetBIOS machines register their own names as per the above. This is in vast contrast to TCP/IP installations where traditionally the system administrator will determine in the /etc/hosts or in the DNS database what names are associated with each IP address.

One further point of clarification should be noted, the /etc/hosts file and the DNS records do not provide the NetBIOS name type information that MS Windows clients depend on to locate the type of service that may be needed. An example of this is what happens when an MS Windows client wants to locate a domain logon server. It find this service and the IP address of a server that provides it by performing a lookup (via a NetBIOS broadcast) for enumeration of all machines that have registered the name type *<1c>. A logon request is then sent to each IP address that is returned in the enumerated list of IP addresses. Which ever machine first replies then ends up providing the logon services.

The name "workgroup" or "domain" really can be confusing since these have the added significance of indicating what is the security architecture of the MS Windows network. The term "workgroup" indicates that the primary nature of the network environment is that of a peer-to-peer design. In a WORKGROUP all machines are responsible for their own security, and generally such security is limited to use of just a password (known as SHARE MODE security). In most situations with peer-to-peer networking the users who control their own machines will simply opt to have no security at all. It is possible to have USER MODE security in a WORKGROUP environment, thus requiring use of a user name and a matching password.

MS Windows networking is thus predetermined to use machine names for all local and remote machine message passing. The protocol used is called Server Message Block (SMB) and this is implemented using the NetBIOS protocol (Network Basic Input Output System). NetBIOS can be encapsulated using LLC (Logical Link Control) protocol - in which case the resulting protocol is called NetBEUI (Network Basic Extended User Interface). NetBIOS can also be run over IPX (Internetworking Packet Exchange) protocol as used by Novell NetWare, and it can be run over TCP/IP protocols - in which case the resulting protocol is called NBT or NetBT, the NetBIOS over TCP/IP.

MS Windows machines use a complex array of name resolution mechanisms. Since we are primarily concerned with TCP/IP this demonstration is limited to this area.

	# Copyright (c) 1998 Microsoft Corp.
	#
	# This is a sample LMHOSTS file used by the Microsoft Wins Client (NetBIOS
	# over TCP/IP) stack for Windows98
	#
	# This file contains the mappings of IP addresses to NT computernames
	# (NetBIOS) names.  Each entry should be kept on an individual line.
	# The IP address should be placed in the first column followed by the
	# corresponding computername. The address and the comptername
	# should be separated by at least one space or tab. The "#" character
	# is generally used to denote the start of a comment (see the exceptions
	# below).
	#
	# This file is compatible with Microsoft LAN Manager 2.x TCP/IP lmhosts
	# files and offers the following extensions:
	#
	#      #PRE
	#      #DOM:<domain>
	#      #INCLUDE <filename>
	#      #BEGIN_ALTERNATE
	#      #END_ALTERNATE
	#      \0xnn (non-printing character support)
	#
	# Following any entry in the file with the characters "#PRE" will cause
	# the entry to be preloaded into the name cache. By default, entries are
	# not preloaded, but are parsed only after dynamic name resolution fails.
	#
	# Following an entry with the "#DOM:<domain>" tag will associate the
	# entry with the domain specified by <domain>. This affects how the
	# browser and logon services behave in TCP/IP environments. To preload
	# the host name associated with #DOM entry, it is necessary to also add a
	# #PRE to the line. The <domain> is always preloaded although it will not
	# be shown when the name cache is viewed.
	#
	# Specifying "#INCLUDE <filename>" will force the RFC NetBIOS (NBT)
	# software to seek the specified <filename> and parse it as if it were
	# local. <filename> is generally a UNC-based name, allowing a
	# centralized lmhosts file to be maintained on a server.
	# It is ALWAYS necessary to provide a mapping for the IP address of the
	# server prior to the #INCLUDE. This mapping must use the #PRE directive.
	# In addtion the share "public" in the example below must be in the
	# LanManServer list of "NullSessionShares" in order for client machines to
	# be able to read the lmhosts file successfully. This key is under
	# \machine\system\currentcontrolset\services\lanmanserver\parameters\nullsessionshares
	# in the registry. Simply add "public" to the list found there.
	#
	# The #BEGIN_ and #END_ALTERNATE keywords allow multiple #INCLUDE
	# statements to be grouped together. Any single successful include
	# will cause the group to succeed.
	#
	# Finally, non-printing characters can be embedded in mappings by
	# first surrounding the NetBIOS name in quotations, then using the
	# \0xnn notation to specify a hex value for a non-printing character.
	#
	# The following example illustrates all of these extensions:
	#
	# 102.54.94.97     rhino         #PRE #DOM:networking  #net group's DC
	# 102.54.94.102    "appname  \0x14"                    #special app server
	# 102.54.94.123    popular            #PRE             #source server
	# 102.54.94.117    localsrv           #PRE             #needed for the include
	#
	# #BEGIN_ALTERNATE
	# #INCLUDE \\localsrv\public\lmhosts
	# #INCLUDE \\rhino\public\lmhosts
	# #END_ALTERNATE
	#
	# In the above example, the "appname" server contains a special
	# character in its name, the "popular" and "localsrv" server names are
	# preloaded, and the "rhino" server name is specified so it can be used
	# to later #INCLUDE a centrally maintained lmhosts file if the "localsrv"
	# system is unavailable.
	#
	# Note that the whole file is parsed including comments on each lookup,
	# so keeping the number of comments to a minimum will improve performance.
	# Therefore it is not advisable to simply add lmhosts file entries onto the
	# end of this file.

	wins support = Yes

	wins support = No
	wins server = xxx.xxx.xxx.xxx

22.1. Overview of browsing

SMB networking provides a mechanism by which clients can access a list of machines in a network, a so-called browse list. This list contains machines that are ready to offer file and/or print services to other machines within the network. Thus it does not include machines which aren't currently able to do server tasks. The browse list is heavily used by all SMB clients. Configuration of SMB browsing has been problematic for some Samba users, hence this document.

MS Windows 2000 and later, as with Samba 3 and later, can be configured to not use NetBIOS over TCP/IP. When configured this way it is imperative that name resolution (using DNS/LDAP/ADS) be correctly configured and operative. Browsing will NOT work if name resolution from SMB machine names to IP addresses does not function correctly.

Where NetBIOS over TCP/IP is enabled use of a WINS server is highly recommended to aid the resolution of NetBIOS (SMB) names to IP addresses. WINS allows remote segment clients to obtain NetBIOS name_type information that can NOT be provided by any other means of name resolution.

22.2. Browsing support in samba

Samba facilitates browsing. The browsing is supported by nmbd and is also controlled by options in the smb.conf file (see smb.conf(5)). Samba can act as a local browse master for a workgroup and the ability for samba to support domain logons and scripts is now available.

Samba can also act as a domain master browser for a workgroup. This means that it will collate lists from local browse masters into a wide area network server list. In order for browse clients to resolve the names they may find in this list, it is recommended that both samba and your clients use a WINS server.

Note that you should NOT set Samba to be the domain master for a workgroup that has the same name as an NT Domain: on each wide area network, you must only ever have one domain master browser per workgroup, regardless of whether it is NT, Samba or any other type of domain master that is providing this service.

Nmbd can be configured as a WINS server, but it is not necessary to specifically use samba as your WINS server. MS Windows NT4, Server or Advanced Server 2000 or 2003 can be configured as your WINS server. In a mixed NT/2000/2003 server and samba environment on a Wide Area Network, it is recommended that you use the Microsoft WINS server capabilities. In a samba-only environment, it is recommended that you use one and only one Samba server as your WINS server.

To get browsing to work you need to run nmbd as usual, but will need to use the workgroup option in smb.conf to control what workgroup Samba becomes a part of.

Samba also has a useful option for a Samba server to offer itself for browsing on another subnet. It is recommended that this option is only used for 'unusual' purposes: announcements over the internet, for example. See remote announce in the smb.conf man page.

22.3. Problem resolution

If something doesn't work then hopefully the log.nmb file will help you track down the problem. Try a debug level of 2 or 3 for finding problems. Also note that the current browse list usually gets stored in text form in a file called browse.dat.

Note that if it doesn't work for you, then you should still be able to type the server name as \\SERVER in filemanager then hit enter and filemanager should display the list of available shares.

Some people find browsing fails because they don't have the global guest account set to a valid account. Remember that the IPC$ connection that lists the shares is done as guest, and thus you must have a valid guest account.

MS Windows 2000 and upwards (as with Samba) can be configured to disallow anonymous (ie: Guest account) access to the IPC$ share. In that case, the MS Windows 2000/XP/2003 machine acting as an SMB/CIFS client will use the name of the currently logged in user to query the IPC$ share. MS Windows 9X clients are not able to do this and thus will NOT be able to browse server resources.

The other big problem people have is that their broadcast address, netmask or IP address is wrong (specified with the "interfaces" option in smb.conf)

22.4. Browsing across subnets

Since the release of Samba 1.9.17(alpha1) Samba has been updated to enable it to support the replication of browse lists across subnet boundaries. New code and options have been added to achieve this. This section describes how to set this feature up in different settings.

To see browse lists that span TCP/IP subnets (ie. networks separated by routers that don't pass broadcast traffic) you must set up at least one WINS server. The WINS server acts as a DNS for NetBIOS names, allowing NetBIOS name to IP address translation to be done by doing a direct query of the WINS server. This is done via a directed UDP packet on port 137 to the WINS server machine. The reason for a WINS server is that by default, all NetBIOS name to IP address translation is done by broadcasts from the querying machine. This means that machines on one subnet will not be able to resolve the names of machines on another subnet without using a WINS server.

Remember, for browsing across subnets to work correctly, all machines, be they Windows 95, Windows NT, or Samba servers must have the IP address of a WINS server given to them by a DHCP server, or by manual configuration (for Win95 and WinNT, this is in the TCP/IP Properties, under Network settings) for Samba this is in the smb.conf file.

                                   (DMB)
             N1_A      N1_B        N1_C       N1_D        N1_E
              |          |           |          |           |
          -------------------------------------------------------
            |          subnet 1                       |
          +---+                                      +---+
          |R1 | Router 1                  Router 2   |R2 |
          +---+                                      +---+
            |                                          |
            |  subnet 2              subnet 3          |
  --------------------------       ------------------------------------
  |     |     |      |               |        |         |           |
 N2_A  N2_B  N2_C   N2_D           N3_A     N3_B      N3_C        N3_D 
                    (WINS)

Subnet           Browse Master   List
------           -------------   ----
Subnet1          N1_C            N1_A, N1_B, N1_C, N1_D, N1_E

Subnet2          N2_B            N2_A, N2_B, N2_C, N2_D

Subnet3          N3_D            N3_A, N3_B, N3_C, N3_D

Subnet           Browse Master   List
------           -------------   ----
Subnet1          N1_C            N1_A, N1_B, N1_C, N1_D, N1_E, 
                                 N2_A(*), N2_B(*), N2_C(*), N2_D(*)

Subnet2          N2_B            N2_A, N2_B, N2_C, N2_D
                                 N1_A(*), N1_B(*), N1_C(*), N1_D(*), N1_E(*)

Subnet3          N3_D            N3_A, N3_B, N3_C, N3_D

Servers with a (*) after them are non-authoritative names.

Subnet           Browse Master   List
------           -------------   ----
Subnet1          N1_C            N1_A, N1_B, N1_C, N1_D, N1_E, 
                                 N2_A(*), N2_B(*), N2_C(*), N2_D(*),
                                 N3_A(*), N3_B(*), N3_C(*), N3_D(*)

Subnet2          N2_B            N2_A, N2_B, N2_C, N2_D
                                 N1_A(*), N1_B(*), N1_C(*), N1_D(*), N1_E(*)

Subnet3          N3_D            N3_A, N3_B, N3_C, N3_D
                                 N1_A(*), N1_B(*), N1_C(*), N1_D(*), N1_E(*),
                                 N2_A(*), N2_B(*), N2_C(*), N2_D(*)

Servers with a (*) after them are non-authoritative names.

Subnet           Browse Master   List
------           -------------   ----
Subnet1          N1_C            N1_A, N1_B, N1_C, N1_D, N1_E, 
                                 N2_A(*), N2_B(*), N2_C(*), N2_D(*),
                                 N3_A(*), N3_B(*), N3_C(*), N3_D(*)

Subnet2          N2_B            N2_A, N2_B, N2_C, N2_D
                                 N1_A(*), N1_B(*), N1_C(*), N1_D(*), N1_E(*)
                                 N3_A(*), N3_B(*), N3_C(*), N3_D(*)

Subnet3          N3_D            N3_A, N3_B, N3_C, N3_D
                                 N1_A(*), N1_B(*), N1_C(*), N1_D(*), N1_E(*),
                                 N2_A(*), N2_B(*), N2_C(*), N2_D(*)
	
Servers with a (*) after them are non-authoritative names.

22.5. Setting up a WINS server

Either a Samba machine or a Windows NT Server machine may be set up as a WINS server. To set a Samba machine to be a WINS server you must add the following option to the smb.conf file on the selected machine : in the [globals] section add the line

wins support = yes

Versions of Samba prior to 1.9.17 had this parameter default to yes. If you have any older versions of Samba on your network it is strongly suggested you upgrade to a recent version, or at the very least set the parameter to 'no' on all these machines.

Machines with wins support = yes will keep a list of all NetBIOS names registered with them, acting as a DNS for NetBIOS names.

You should set up only ONE wins server. Do NOT set the wins support = yes option on more than one Samba server.

To set up a Windows NT Server as a WINS server you need to set up the WINS service - see your NT documentation for details. Note that Windows NT WINS Servers can replicate to each other, allowing more than one to be set up in a complex subnet environment. As Microsoft refuse to document these replication protocols Samba cannot currently participate in these replications. It is possible in the future that a Samba->Samba WINS replication protocol may be defined, in which case more than one Samba machine could be set up as a WINS server but currently only one Samba server should have the wins support = yes parameter set.

After the WINS server has been configured you must ensure that all machines participating on the network are configured with the address of this WINS server. If your WINS server is a Samba machine, fill in the Samba machine IP address in the "Primary WINS Server" field of the "Control Panel->Network->Protocols->TCP->WINS Server" dialogs in Windows 95 or Windows NT. To tell a Samba server the IP address of the WINS server add the following line to the [global] section of all smb.conf files :

wins server = >name or IP address<

where >name or IP address< is either the DNS name of the WINS server machine or its IP address.

Note that this line MUST NOT BE SET in the smb.conf file of the Samba server acting as the WINS server itself. If you set both the wins support = yes option and the wins server = <name> option then nmbd will fail to start.

There are two possible scenarios for setting up cross subnet browsing. The first details setting up cross subnet browsing on a network containing Windows 95, Samba and Windows NT machines that are not configured as part of a Windows NT Domain. The second details setting up cross subnet browsing on networks that contain NT Domains.

22.6. Setting up Browsing in a WORKGROUP

To set up cross subnet browsing on a network containing machines in up to be in a WORKGROUP, not an NT Domain you need to set up one Samba server to be the Domain Master Browser (note that this is *NOT* the same as a Primary Domain Controller, although in an NT Domain the same machine plays both roles). The role of a Domain master browser is to collate the browse lists from local master browsers on all the subnets that have a machine participating in the workgroup. Without one machine configured as a domain master browser each subnet would be an isolated workgroup, unable to see any machines on any other subnet. It is the presense of a domain master browser that makes cross subnet browsing possible for a workgroup.

In an WORKGROUP environment the domain master browser must be a Samba server, and there must only be one domain master browser per workgroup name. To set up a Samba server as a domain master browser, set the following option in the [global] section of the smb.conf file :

domain master = yes

The domain master browser should also preferrably be the local master browser for its own subnet. In order to achieve this set the following options in the [global] section of the smb.conf file :

domain master = yes
local master = yes
preferred master = yes
os level = 65

The domain master browser may be the same machine as the WINS server, if you require.

Next, you should ensure that each of the subnets contains a machine that can act as a local master browser for the workgroup. Any MS Windows NT/2K/XP/2003 machine should be able to do this, as will Windows 9x machines (although these tend to get rebooted more often, so it's not such a good idea to use these). To make a Samba server a local master browser set the following options in the [global] section of the smb.conf file :

domain master = no
local master = yes
preferred master = yes
os level = 65

Do not do this for more than one Samba server on each subnet, or they will war with each other over which is to be the local master browser.

The local master parameter allows Samba to act as a local master browser. The preferred master causes nmbd to force a browser election on startup and the os level parameter sets Samba high enough so that it should win any browser elections.

If you have an NT machine on the subnet that you wish to be the local master browser then you can disable Samba from becoming a local master browser by setting the following options in the [global] section of the smb.conf file :

domain master = no
local master = no
preferred master = no
os level = 0

22.7. Setting up Browsing in a DOMAIN

If you are adding Samba servers to a Windows NT Domain then you must not set up a Samba server as a domain master browser. By default, a Windows NT Primary Domain Controller for a Domain name is also the Domain master browser for that name, and many things will break if a Samba server registers the Domain master browser NetBIOS name (DOMAIN<1B>) with WINS instead of the PDC.

For subnets other than the one containing the Windows NT PDC you may set up Samba servers as local master browsers as described. To make a Samba server a local master browser set the following options in the [global] section of the smb.conf file :

domain master = no
local master = yes
preferred master = yes
os level = 65

If you wish to have a Samba server fight the election with machines on the same subnet you may set the os level parameter to lower levels. By doing this you can tune the order of machines that will become local master browsers if they are running. For more details on this see the section Forcing samba to be the master browser below.

If you have Windows NT machines that are members of the domain on all subnets, and you are sure they will always be running then you can disable Samba from taking part in browser elections and ever becoming a local master browser by setting following options in the [global] section of the smb.conf file :

domain master = no local master = no preferred master = no os level = 0

22.8. Forcing samba to be the master

Who becomes the master browser is determined by an election process using broadcasts. Each election packet contains a number of parameters which determine what precedence (bias) a host should have in the election. By default Samba uses a very low precedence and thus loses elections to just about anyone else.

If you want Samba to win elections then just set the os level global option in smb.conf to a higher number. It defaults to 0. Using 34 would make it win all elections over every other system (except other samba systems!)

A os level of 2 would make it beat WfWg and Win95, but not MS Windows NT/2K Server. A MS Windows NT/2K Server domain controller uses level 32.

The maximum os level is 255

If you want samba to force an election on startup, then set the preferred master global option in smb.conf to "yes". Samba will then have a slight advantage over other potential master browsers that are not preferred master browsers. Use this parameter with care, as if you have two hosts (whether they are windows 95 or NT or samba) on the same local subnet both set with preferred master to "yes", then periodically and continually they will force an election in order to become the local master browser.

If you want samba to be a domain master browser, then it is recommended that you also set preferred master to "yes", because samba will not become a domain master browser for the whole of your LAN or WAN if it is not also a local master browser on its own broadcast isolated subnet.

It is possible to configure two samba servers to attempt to become the domain master browser for a domain. The first server that comes up will be the domain master browser. All other samba servers will attempt to become the domain master browser every 5 minutes. They will find that another samba server is already the domain master browser and will fail. This provides automatic redundancy, should the current domain master browser fail.

22.9. Making samba the domain master

The domain master is responsible for collating the browse lists of multiple subnets so that browsing can occur between subnets. You can make samba act as the domain master by setting domain master = yes in smb.conf. By default it will not be a domain master.

Note that you should NOT set Samba to be the domain master for a workgroup that has the same name as an NT Domain.

When samba is the domain master and the master browser it will listen for master announcements (made roughly every twelve minutes) from local master browsers on other subnets and then contact them to synchronise browse lists.

If you want samba to be the domain master then I suggest you also set the os level high enough to make sure it wins elections, and set preferred master to "yes", to get samba to force an election on startup.

Note that all your servers (including samba) and clients should be using a WINS server to resolve NetBIOS names. If your clients are only using broadcasting to resolve NetBIOS names, then two things will occur:

1. your local master browsers will be unable to find a domain master browser, as it will only be looking on the local subnet.

2. if a client happens to get hold of a domain-wide browse list, and a user attempts to access a host in that list, it will be unable to resolve the NetBIOS name of that host.

If, however, both samba and your clients are using a WINS server, then:

1. your local master browsers will contact the WINS server and, as long as samba has registered that it is a domain master browser with the WINS server, your local master browser will receive samba's ip address as its domain master browser.

2. when a client receives a domain-wide browse list, and a user attempts to access a host in that list, it will contact the WINS server to resolve the NetBIOS name of that host. as long as that host has registered its NetBIOS name with the same WINS server, the user will be able to see that host.

22.10. Note about broadcast addresses

If your network uses a "0" based broadcast address (for example if it ends in a 0) then you will strike problems. Windows for Workgroups does not seem to support a 0's broadcast and you will probably find that browsing and name lookups won't work.

22.11. Multiple interfaces

Samba now supports machines with multiple network interfaces. If you have multiple interfaces then you will need to use the interfaces option in smb.conf to configure them. See smb.conf(5) for details.

23.1. Introduction

This note was attached to the Samba 2.2.8 release notes as it contained an important security fix. The information contained here applies to Samba installations in general.

23.2. Using host based protection

In many installations of Samba the greatest threat comes for outside your immediate network. By default Samba will accept connections from any host, which means that if you run an insecure version of Samba on a host that is directly connected to the Internet you can be especially vulnerable.

One of the simplest fixes in this case is to use the 'hosts allow' and 'hosts deny' options in the Samba smb.conf configuration file to only allow access to your server from a specific range of hosts. An example might be:

  hosts allow = 127.0.0.1 192.168.2.0/24 192.168.3.0/24
  hosts deny = 0.0.0.0/0

The above will only allow SMB connections from 'localhost' (your own computer) and from the two private networks 192.168.2 and 192.168.3. All other connections will be refused connections as soon as the client sends its first packet. The refusal will be marked as a 'not listening on called name' error.

23.3. Using interface protection

By default Samba will accept connections on any network interface that it finds on your system. That means if you have a ISDN line or a PPP connection to the Internet then Samba will accept connections on those links. This may not be what you want.

You can change this behaviour using options like the following:

  interfaces = eth* lo
  bind interfaces only = yes

This tells Samba to only listen for connections on interfaces with a name starting with 'eth' such as eth0, eth1, plus on the loopback interface called 'lo'. The name you will need to use depends on what OS you are using, in the above I used the common name for Ethernet adapters on Linux.

If you use the above and someone tries to make a SMB connection to your host over a PPP interface called 'ppp0' then they will get a TCP connection refused reply. In that case no Samba code is run at all as the operating system has been told not to pass connections from that interface to any process.

23.4. Using a firewall

Many people use a firewall to deny access to services that they don't want exposed outside their network. This can be a very good idea, although I would recommend using it in conjunction with the above methods so that you are protected even if your firewall is not active for some reason.

If you are setting up a firewall then you need to know what TCP and UDP ports to allow and block. Samba uses the following:

UDP/137    - used by nmbd
UDP/138    - used by nmbd
TCP/139    - used by smbd
TCP/445    - used by smbd

The last one is important as many older firewall setups may not be aware of it, given that this port was only added to the protocol in recent years.

23.5. Using a IPC$ share deny

If the above methods are not suitable, then you could also place a more specific deny on the IPC$ share that is used in the recently discovered security hole. This allows you to offer access to other shares while denying access to IPC$ from potentially untrustworthy hosts.

To do that you could use:

  [ipc$]
     hosts allow = 192.168.115.0/24 127.0.0.1
     hosts deny = 0.0.0.0/0

this would tell Samba that IPC$ connections are not allowed from anywhere but the two listed places (localhost and a local subnet). Connections to other shares would still be allowed. As the IPC$ share is the only share that is always accessible anonymously this provides some level of protection against attackers that do not know a username/password for your host.

If you use this method then clients will be given a 'access denied' reply when they try to access the IPC$ share. That means that those clients will not be able to browse shares, and may also be unable to access some other resources.

This is not recommended unless you cannot use one of the other methods listed above for some reason.

23.6. Upgrading Samba

Please check regularly on http://www.samba.org/ for updates and important announcements. Occasionally security releases are made and it is highly recommended to upgrade Samba when a security vulnerability is discovered.

24.1. What are charsets and unicode?

Computers communicate in numbers. In texts, each number will be translated to a corresponding letter. The meaning that will be assigned to a certain number depends on the character set(charset) that is used. A charset can be seen as a table that is used to translate numbers to letters. Not all computers use the same charset (there are charsets with German umlauts, Japanese characters, etc). Usually a charset contains 256 characters, which means that storing a character with it takes exactly one byte.

There are also charsets that support even more characters, but those need twice(or even more) as much storage space. These charsets can contain 256 * 256 = 65536 characters, which is more then all possible characters one could think of. They are called multibyte charsets (because they use more then one byte to store one character).

A standardised multibyte charset is unicode, info available at www.unicode.org. Big advantage of using a multibyte charset is that you only need one; no need to make sure two computers use the same charset when they are communicating.

Old windows clients used to use single-byte charsets, named 'codepages' by microsoft. However, there is no support for negotiating the charset to be used in the smb protocol. Thus, you have to make sure you are using the same charset when talking to an old client. Newer clients (Windows NT, 2K, XP) talk unicode over the wire.

24.2. Samba and charsets

As of samba 3.0, samba can (and will) talk unicode over the wire. Internally, samba knows of three kinds of character sets:

25.1. SWAT Features and Benefits

You must use at least the following ...

26.1. Planning and Getting Started

You must use at least the following ...

26.2. Managing Samba-3 Domain Control

Lots of blah blah here.

27.1. Comparisons

The Samba server uses TCP to talk to the client. Thus if you are trying to see if it performs well you should really compare it to programs that use the same protocol. The most readily available programs for file transfer that use TCP are ftp or another TCP based SMB server.

If you want to test against something like a NT or WfWg server then you will have to disable all but TCP on either the client or server. Otherwise you may well be using a totally different protocol (such as Netbeui) and comparisons may not be valid.

Generally you should find that Samba performs similarly to ftp at raw transfer speed. It should perform quite a bit faster than NFS, although this very much depends on your system.

Several people have done comparisons between Samba and Novell, NFS or WinNT. In some cases Samba performed the best, in others the worst. I suspect the biggest factor is not Samba vs some other system but the hardware and drivers used on the various systems. Given similar hardware Samba should certainly be competitive in speed with other systems.

27.2. Socket options

There are a number of socket options that can greatly affect the performance of a TCP based server like Samba.

The socket options that Samba uses are settable both on the command line with the -O option, or in the smb.conf file.

The "socket options" section of the smb.conf manual page describes how to set these and gives recommendations.

Getting the socket options right can make a big difference to your performance, but getting them wrong can degrade it by just as much. The correct settings are very dependent on your local network.

The socket option TCP_NODELAY is the one that seems to make the biggest single difference for most networks. Many people report that adding "socket options = TCP_NODELAY" doubles the read performance of a Samba drive. The best explanation I have seen for this is that the Microsoft TCP/IP stack is slow in sending tcp ACKs.

27.3. Read size

The option "read size" affects the overlap of disk reads/writes with network reads/writes. If the amount of data being transferred in several of the SMB commands (currently SMBwrite, SMBwriteX and SMBreadbraw) is larger than this value then the server begins writing the data before it has received the whole packet from the network, or in the case of SMBreadbraw, it begins writing to the network before all the data has been read from disk.

This overlapping works best when the speeds of disk and network access are similar, having very little effect when the speed of one is much greater than the other.

The default value is 16384, but very little experimentation has been done yet to determine the optimal value, and it is likely that the best value will vary greatly between systems anyway. A value over 65536 is pointless and will cause you to allocate memory unnecessarily.

27.4. Max xmit

At startup the client and server negotiate a "maximum transmit" size, which limits the size of nearly all SMB commands. You can set the maximum size that Samba will negotiate using the "max xmit = " option in smb.conf. Note that this is the maximum size of SMB request that Samba will accept, but not the maximum size that the *client* will accept. The client maximum receive size is sent to Samba by the client and Samba honours this limit.

It defaults to 65536 bytes (the maximum), but it is possible that some clients may perform better with a smaller transmit unit. Trying values of less than 2048 is likely to cause severe problems.

In most cases the default is the best option.

27.5. Log level

If you set the log level (also known as "debug level") higher than 2 then you may suffer a large drop in performance. This is because the server flushes the log file after each operation, which can be very expensive.

27.6. Read raw

The "read raw" operation is designed to be an optimised, low-latency file read operation. A server may choose to not support it, however. and Samba makes support for "read raw" optional, with it being enabled by default.

In some cases clients don't handle "read raw" very well and actually get lower performance using it than they get using the conventional read operations.

So you might like to try "read raw = no" and see what happens on your network. It might lower, raise or not affect your performance. Only testing can really tell.

27.7. Write raw

The "write raw" operation is designed to be an optimised, low-latency file write operation. A server may choose to not support it, however. and Samba makes support for "write raw" optional, with it being enabled by default.

Some machines may find "write raw" slower than normal write, in which case you may wish to change this option.

27.8. Slow Clients

One person has reported that setting the protocol to COREPLUS rather than LANMAN2 gave a dramatic speed improvement (from 10k/s to 150k/s).

I suspect that his PC's (386sx16 based) were asking for more data than they could chew. I suspect a similar speed could be had by setting "read raw = no" and "max xmit = 2048", instead of changing the protocol. Lowering the "read size" might also help.

27.9. Slow Logins

Slow logins are almost always due to the password checking time. Using the lowest practical "password level" will improve things a lot. You could also enable the "UFC crypt" option in the Makefile.

27.10. Client tuning

Often a speed problem can be traced to the client. The client (for example Windows for Workgroups) can often be tuned for better TCP performance.

See your client docs for details. In particular, I have heard rumours that the WfWg options TCPWINDOWSIZE and TCPSEGMENTSIZE can have a large impact on performance.

Also note that some people have found that setting DefaultRcvWindow in the [MSTCP] section of the SYSTEM.INI file under WfWg to 3072 gives a big improvement. I don't know why.

My own experience wth DefaultRcvWindow is that I get much better performance with a large value (16384 or larger). Other people have reported that anything over 3072 slows things down enourmously. One person even reported a speed drop of a factor of 30 when he went from 3072 to 8192. I don't know why.

It probably depends a lot on your hardware, and the type of unix box you have at the other end of the link.

Paul Cochrane has done some testing on client side tuning and come to the following conclusions:

Install the W2setup.exe file from www.microsoft.com. This is an update for the winsock stack and utilities which improve performance.

Configure the win95 TCPIP registry settings to give better perfomance. I use a program called MTUSPEED.exe which I got off the net. There are various other utilities of this type freely available. The setting which give the best performance for me are:

1. MaxMTU Remove

2. RWIN Remove

3. MTUAutoDiscover Disable

4. MTUBlackHoleDetect Disable

5. Time To Live Enabled

6. Time To Live - HOPS 32

7. NDI Cache Size 0

I tried virtually all of the items mentioned in the document and the only one which made a difference to me was the socket options. It turned out I was better off without any!!!!!

In terms of overall speed of transfer, between various win95 clients and a DX2-66 20MB server with a crappy NE2000 compatible and old IDE drive (Kernel 2.0.30). The transfer rate was reasonable for 10 baseT.

The figures are:          Put              Get 
P166 client 3Com card:    420-440kB/s      500-520kB/s
P100 client 3Com card:    390-410kB/s      490-510kB/s
DX4-75 client NE2000:     370-380kB/s      330-350kB/s

I based these test on transfer two files a 4.5MB text file and a 15MB textfile. The results arn't bad considering the hardware Samba is running on. It's a crap machine!!!!

The updates mentioned in 1 and 2 brought up the transfer rates from just over 100kB/s in some clients.

A new client is a P333 connected via a 100MB/s card and hub. The transfer rates from this were good: 450-500kB/s on put and 600+kB/s on get.

Looking at standard FTP throughput, Samba is a bit slower (100kB/s upwards). I suppose there is more going on in the samba protocol, but if it could get up to the rate of FTP the perfomance would be quite staggering.

28.1. HPUX

HP's implementation of supplementary groups is, er, non-standard (for hysterical reasons). There are two group files, /etc/group and /etc/logingroup; the system maps UIDs to numbers using the former, but initgroups() reads the latter. Most system admins who know the ropes symlink /etc/group to /etc/logingroup (hard link doesn't work for reasons too stupid to go into here). initgroups() will complain if one of the groups you're in in /etc/logingroup has what it considers to be an invalid ID, which means outside the range [0..UID_MAX], where UID_MAX is (I think) 60000 currently on HP-UX. This precludes -2 and 65534, the usual 'nobody' GIDs.

If you encounter this problem, make sure that the programs that are failing to initgroups() be run as users not in any groups with GIDs outside the allowed range.

This is documented in the HP manual pages under setgroups(2) and passwd(4).

On HPUX you must use gcc or the HP Ansi compiler. The free compiler that comes with HP-UX is not Ansi compliant and cannot compile Samba.

28.2. SCO Unix

If you run an old version of SCO Unix then you may need to get important TCP/IP patches for Samba to work correctly. Without the patch, you may encounter corrupt data transfers using samba.

The patch you need is UOD385 Connection Drivers SLS. It is available from SCO (ftp.sco.com, directory SLS, files uod385a.Z and uod385a.ltr.Z).

28.3. DNIX

DNIX has a problem with seteuid() and setegid(). These routines are needed for Samba to work correctly, but they were left out of the DNIX C library for some reason.

For this reason Samba by default defines the macro NO_EID in the DNIX section of includes.h. This works around the problem in a limited way, but it is far from ideal, some things still won't work right.

To fix the problem properly you need to assemble the following two functions and then either add them to your C library or link them into Samba.

put this in the file setegid.s:

        .globl  _setegid
_setegid:
        moveq   #47,d0
        movl    #100,a0
        moveq   #1,d1
        movl    4(sp),a1
        trap    #9
        bccs    1$
        jmp     cerror
1$:
        clrl    d0
        rts

put this in the file seteuid.s:

        .globl  _seteuid
_seteuid:
        moveq   #47,d0
        movl    #100,a0
        moveq   #0,d1
        movl    4(sp),a1
        trap    #9
        bccs    1$
        jmp     cerror
1$:
        clrl    d0
        rts

after creating the above files you then assemble them using

as seteuid.s

as setegid.s

that should produce the files seteuid.o and setegid.o

then you need to add these to the LIBSM line in the DNIX section of the Samba Makefile. Your LIBSM line will then look something like this:

LIBSM = setegid.o seteuid.o -ln

You should then remove the line:

#define NO_EID

from the DNIX section of includes.h

28.4. RedHat Linux Rembrandt-II

By default RedHat Rembrandt-II during installation adds an entry to /etc/hosts as follows:

	127.0.0.1 loopback "hostname"."domainname"

This causes Samba to loop back onto the loopback interface. The result is that Samba fails to communicate correctly with the world and therefor may fail to correctly negotiate who is the master browse list holder and who is the master browser.

Corrective Action: Delete the entry after the word loopback in the line starting 127.0.0.1

28.5. AIX

29.1. Macintosh clients?

Yes. Thursby now have a CIFS Client / Server called DAVE - see

They test it against Windows 95, Windows NT and samba for compatibility issues. At the time of writing, DAVE was at version 1.0.1. The 1.0.0 to 1.0.1 update is available as a free download from the Thursby web site (the speed of finder copies has been greatly enhanced, and there are bug-fixes included).

Alternatives - There are two free implementations of AppleTalk for several kinds of UNIX machnes, and several more commercial ones. These products allow you to run file services and print services natively to Macintosh users, with no additional support required on the Macintosh. The two free omplementations are Netatalk, and CAP. What Samba offers MS Windows users, these packages offer to Macs. For more info on these packages, Samba, and Linux (and other UNIX-based systems) see http://www.eats.com/linux_mac_win.html

29.2. OS2 Client

		20=setup.exe
		20=netwksta.sys
		20=netvdd.sys
		

29.3. Windows for Workgroups

29.4. Windows '95/'98

When using Windows 95 OEM SR2 the following updates are recommended where Samba is being used. Please NOTE that the above change will affect you once these updates have been installed.

There are more updates than the ones mentioned here. You are referred to the Microsoft Web site for all currently available updates to your specific version of Windows 95.

1. Kernel Update: KRNLUPD.EXE

2. Ping Fix: PINGUPD.EXE

3. RPC Update: RPCRTUPD.EXE

4. TCP/IP Update: VIPUPD.EXE

5. Redirector Update: VRDRUPD.EXE

Also, if using MS OutLook it is desirable to install the OLEUPD.EXE fix. This fix may stop your machine from hanging for an extended period when exiting OutLook and you may also notice a significant speedup when accessing network neighborhood services.

29.5. Windows 2000 Service Pack 2

There are several annoyances with Windows 2000 SP2. One of which only appears when using a Samba server to host user profiles to Windows 2000 SP2 clients in a Windows domain. This assumes that Samba is a member of the domain, but the problem will likely occur if it is not.

In order to server profiles successfully to Windows 2000 SP2 clients (when not operating as a PDC), Samba must have nt acl support = no added to the file share which houses the roaming profiles. If this is not done, then the Windows 2000 SP2 client will complain about not being able to access the profile (Access Denied) and create multiple copies of it on disk (DOMAIN.user.001, DOMAIN.user.002, etc...). See the smb.conf(5) man page for more details on this option. Also note that the nt acl support parameter was formally a global parameter in releases prior to Samba 2.2.2.

The following is a minimal profile share:

	[profile]
		path = /export/profile
		create mask = 0600
		directory mask = 0700
		nt acl support = no
		read only = no

The reason for this bug is that the Win2k SP2 client copies the security descriptor for the profile which contains the Samba server's SID, and not the domain SID. The client compares the SID for SAMBA\user and realizes it is different that the one assigned to DOMAIN\user. Hence the reason for the "access denied" message.

By disabling the nt acl support parameter, Samba will send the Win2k client a response to the QuerySecurityDescriptor trans2 call which causes the client to set a default ACL for the profile. This default ACL includes

DOMAIN\user "Full Control"

This bug does not occur when using winbind to create accounts on the Samba host for Domain users.

29.6. Windows NT 3.1

If you have problems communicating across routers with Windows NT 3.1 workstations, read this Microsoft Knowledge Base article.

30.1. Access Samba source code via CVS

30.2. Accessing the samba sources via rsync and ftp

pserver.samba.org also exports unpacked copies of most parts of the CVS tree at ftp://pserver.samba.org/pub/unpacked and also via anonymous rsync at rsync://pserver.samba.org/ftp/unpacked/. I recommend using rsync rather than ftp. See the rsync homepage for more info on rsync.

The disadvantage of the unpacked trees is that they do not support automatic merging of local changes like CVS does. rsync access is most convenient for an initial install.

30.3. Building the Binaries

To do this, first run the program ./configure in the source directory. This should automatically configure Samba for your operating system. If you have unusual needs then you may wish to run

root# ./configure --help

first to see what special options you can enable. Then executing

root# make

will create the binaries. Once it's successfully compiled you can use

root# make install

to install the binaries and manual pages. You can separately install the binaries and/or man pages using

root# make installbin

and

root# make installman

Note that if you are upgrading for a previous version of Samba you might like to know that the old versions of the binaries will be renamed with a ".old" extension. You can go back to the previous version with

root# make revert

if you find this version a disaster!

#define HAVE_KRB5 1
#define HAVE_LDAP 1
		  

30.4. Starting the smbd and nmbd

You must choose to start smbd and nmbd either as daemons or from inetdDon't try to do both! Either you can put them in inetd.conf and have them started on demand by inetd, or you can start them as daemons either from the command line or in /etc/rc.local. See the man pages for details on the command line options. Take particular care to read the bit about what user you need to be in order to start Samba. In many cases you must be root.

The main advantage of starting smbd and nmbd using the recommended daemon method is that they will respond slightly more quickly to an initial connection request.

---

30.4.1. Starting from inetd.conf

NOTE; The following will be different if you use NIS, NIS+ or LDAP to distribute services maps.

Look at your /etc/services. What is defined at port 139/tcp. If nothing is defined then add a line like this:

netbios-ssn 139/tcp

similarly for 137/udp you should have an entry like:

netbios-ns 137/udp

Next edit your /etc/inetd.conf and add two lines something like this:

		netbios-ssn stream tcp nowait root /usr/local/samba/bin/smbd smbd 
		netbios-ns dgram udp wait root /usr/local/samba/bin/nmbd nmbd 
		

The exact syntax of /etc/inetd.conf varies between unixes. Look at the other entries in inetd.conf for a guide.

NOTE: Some unixes already have entries like netbios_ns (note the underscore) in /etc/services. You must either edit /etc/services or /etc/inetd.conf to make them consistent.

NOTE: On many systems you may need to use the "interfaces" option in smb.conf to specify the IP address and netmask of your interfaces. Run ifconfig as root if you don't know what the broadcast is for your net. nmbd tries to determine it at run time, but fails on some unixes.

Many unixes only accept around 5 parameters on the command line in inetd.conf. This means you shouldn't use spaces between the options and arguments, or you should use a script, and start the script from inetd.

Restart inetd, perhaps just send it a HUP. If you have installed an earlier version of nmbd then you may need to kill nmbd as well.

---

30.4.2. Alternative: starting it as a daemon

To start the server as a daemon you should create a script something like this one, perhaps calling it startsmb.

		#!/bin/sh
		/usr/local/samba/bin/smbd -D 
		/usr/local/samba/bin/nmbd -D 
		

then make it executable with chmod +x startsmb

You can then run startsmb by hand or execute it from /etc/rc.local

To kill it send a kill signal to the processes nmbd and smbd.

If you use the SVR4 style init system then you may like to look at the examples/svr4-startup script to make Samba fit into that system.

31.1. Introduction

The email address for bug reports for stable releases is samba@samba.org. Bug reports for alpha releases should go to samba-technical@samba.org.

Please take the time to read this file before you submit a bug report. Also, please see if it has changed between releases, as we may be changing the bug reporting mechanism at some time.

Please also do as much as you can yourself to help track down the bug. Samba is maintained by a dedicated group of people who volunteer their time, skills and efforts. We receive far more mail about it than we can possibly answer, so you have a much higher chance of an answer and a fix if you send us a "developer friendly" bug report that lets us fix it fast.

Do not assume that if you post the bug to the comp.protocols.smb newsgroup or the mailing list that we will read it. If you suspect that your problem is not a bug but a configuration problem then it is better to send it to the Samba mailing list, as there are (at last count) 5000 other users on that list that may be able to help you.

You may also like to look though the recent mailing list archives, which are conveniently accessible on the Samba web pages at http://samba.org/samba/.

31.2. General info

Before submitting a bug report check your config for silly errors. Look in your log files for obvious messages that tell you that you've misconfigured something and run testparm to test your config file for correct syntax.

Have you run through the diagnosis? This is very important.

If you include part of a log file with your bug report then be sure to annotate it with exactly what you were doing on the client at the time, and exactly what the results were.

31.3. Debug levels

If the bug has anything to do with Samba behaving incorrectly as a server (like refusing to open a file) then the log files will probably be very useful. Depending on the problem a log level of between 3 and 10 showing the problem may be appropriate. A higher level givesmore detail, but may use too much disk space.

To set the debug level use log level = in your smb.conf. You may also find it useful to set the log level higher for just one machine and keep separate logs for each machine. To do this use:

log level = 10
log file = /usr/local/samba/lib/log.%m
include = /usr/local/samba/lib/smb.conf.%m

then create a file /usr/local/samba/lib/smb.conf.machine where machine is the name of the client you wish to debug. In that file put any smb.conf commands you want, for example log level= may be useful. This also allows you to experiment with different security systems, protocol levels etc on just one machine.

The smb.conf entry log level = is synonymous with the entry debuglevel = that has been used in older versions of Samba and is being retained for backwards compatibility of smb.conf files.

As the log level = value is increased you will record a significantly increasing level of debugging information. For most debugging operations you may not need a setting higher than 3. Nearly all bugs can be tracked at a setting of 10, but be prepared for a VERY large volume of log data.

31.4. Internal errors

If you get a "INTERNAL ERROR" message in your log files it means that Samba got an unexpected signal while running. It is probably a segmentation fault and almost certainly means a bug in Samba (unless you have faulty hardware or system software).

If the message came from smbd then it will probably be accompanied by a message which details the last SMB message received by smbd. This info is often very useful in tracking down the problem so please include it in your bug report.

You should also detail how to reproduce the problem, if possible. Please make this reasonably detailed.

You may also find that a core file appeared in a corefiles subdirectory of the directory where you keep your samba log files. This file is the most useful tool for tracking down the bug. To use it you do this:

gdb smbd core

adding appropriate paths to smbd and core so gdb can find them. If you don't have gdb then try dbx. Then within the debugger use the command where to give a stack trace of where the problem occurred. Include this in your mail.

If you known any assembly language then do a disass of the routine where the problem occurred (if its in a library routine then disassemble the routine that called it) and try to work out exactly where the problem is by looking at the surrounding code. Even if you don't know assembly then incuding this info in the bug report can be useful.

31.5. Attaching to a running process

Unfortunately some unixes (in particular some recent linux kernels) refuse to dump a core file if the task has changed uid (which smbd does often). To debug with this sort of system you could try to attach to the running process using gdb smbd PID where you get PID from smbstatus. Then use c to continue and try to cause the core dump using the client. The debugger should catch the fault and tell you where it occurred.

31.6. Patches

The best sort of bug report is one that includes a fix! If you send us patches please use diff -u format if your version of diff supports it, otherwise use diff -c4. Make sure your do the diff against a clean version of the source and let me know exactly what version you used.

32.1. Introduction

This file contains a list of tests you can perform to validate your Samba server. It also tells you what the likely cause of the problem is if it fails any one of these steps. If it passes all these tests then it is probably working fine.

You should do ALL the tests, in the order shown. We have tried to carefully choose them so later tests only use capabilities verified in the earlier tests.

If you send one of the samba mailing lists an email saying "it doesn't work" and you have not followed this test procedure then you should not be surprised your email is ignored.

32.2. Assumptions

In all of the tests it is assumed you have a Samba server called BIGSERVER and a PC called ACLIENT both in workgroup TESTGROUP.

The procedure is similar for other types of clients.

It is also assumed you know the name of an available share in your smb.conf. I will assume this share is called "tmp". You can add a "tmp" share like by adding the following to smb.conf:

[tmp]
 comment = temporary files 
 path = /tmp
 read only = yes

THESE TESTS ASSUME VERSION 3.0.0 OR LATER OF THE SAMBA SUITE. SOME COMMANDS SHOWN DID NOT EXIST IN EARLIER VERSIONS

Please pay attention to the error messages you receive. If any error message reports that your server is being unfriendly you should first check that you IP name resolution is correctly set up. eg: Make sure your /etc/resolv.conf file points to name servers that really do exist.

Also, if you do not have DNS server access for name resolution please check that the settings for your smb.conf file results in "dns proxy = no". The best way to check this is with "testparm smb.conf"

32.3. The tests

32.4. Still having troubles?

Try the mailing list or newsgroup, or use the ethereal utility to sniff the problem. The official samba mailing list can be reached at samba@samba.org. To find out more about samba and how to subscribe to the mailing list check out the samba web page at http://samba.org/samba/

Also look at the other docs in the Samba package!