Table of Contents
In this section the function and purpose of Samba's security modes are described.
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.
I'll describe user level security first, as its simpler. In user level security the client will send a "session setup" command directly after the protocol negotiation. This contains a username and password. The server can either accept or reject that username/password combination. Note that at this stage the server has no idea what share the client will eventually try to connect to, so it can't base the "accept/reject" on anything other than:
the username/password
the machine that the client is coming from
If the server accepts the username/password then the client expects to be able to mount any share (using a "tree connection") without specifying a password. It expects that all access rights will be as the username/password specified in the "session setup".
It is also possible for a client to send multiple "session setup" requests. When the server responds it gives the client a "uid" to use as an authentication tag for that username/password. The client can maintain multiple authentication contexts in this way (WinDD is an example of an application that does this)
Ok, now for share level security. In share level security the client authenticates itself separately for each share. It will send a password along with each "tree connection" (share mount). It does not explicitly send a username with this operation. The client is expecting a password to be associated with each share, independent of the user. This means that samba has to work out what username the client probably wants to use. It is never explicitly sent the username. Some commercial SMB servers such as NT actually associate passwords directly with shares in share level security, but samba always uses the unix authentication scheme where it is a username/password that is authenticated, not a "share/password".
Many clients send a "session setup" even if the server is in share level security. They normally send a valid username but no password. Samba records this username in a list of "possible usernames". When the client then does a "tree connection" it also adds to this list the name of the share they try to connect to (useful for home directories) and any users listed in the user = smb.conf line. The password is then checked in turn against these "possible usernames". If a match is found then the client is authenticated as that user.
Finally "server level" security. In server level security the samba server reports to the client that it is in user level security. The client then does a "session setup" as described earlier. The samba server takes the username/password that the client sends and attempts to login to the "password server" by sending exactly the same username/password that it got from the client. If that server is in user level security and accepts the password then samba accepts the clients connection. This allows the samba server to use another SMB server as the "password server".
You should also note that at the very start of all this, where the server tells the client what security level it is in, it also tells the client if it supports encryption. If it does then it supplies the client with a random "cryptkey". The client will then send all passwords in encrypted form. You have to compile samba with encryption enabled to support this feature, and you have to maintain a separate smbpasswd file with SMB style encrypted passwords. It is cryptographically impossible to translate from unix style encryption to SMB style encryption, although there are some fairly simple management schemes by which the two could be kept in sync.
"security = server" means that Samba reports to clients that it is running in "user mode" but actually passes off all authentication requests to another "user mode" server. This requires an additional parameter "password server =" that points to the real authentication server. That real authentication server can be another Samba server or can be a Windows NT server, the later natively capable of encrypted password support.
Server level security is incompatible with what is known as schannel or "sign and seal" protocols. This means that if you want to use server level security you must disable the use of "sign and seal" on all machines on your network.
MS Windows clients may use encrypted passwords as part of a challenege/response authentication model (a.k.a. NTLMv1) or alone, or clear text strings for simple password based authentication. It should be realized that with the SMB protocol the password is passed over the network either in plain text or encrypted, but not both in the same authentication request.
When encrypted passwords are used a password that has been entered by the user is encrypted in two ways:
An MD4 hash of the UNICODE of the password string. This is known as the NT hash.
The password is converted to upper case, and then padded or trucated to 14 bytes. This string is then appended with 5 bytes of NULL characters and split to form two 56 bit DES keys to encrypt a "magic" 8 byte value. The resulting 16 bytes for the LanMan hash.
MS Windows 95 pre-service pack 1, MS Windows NT versions 3.x and version 4.0 pre-service pack 3 will use either mode of password authentication. All versions of MS Windows that follow these versions no longer support plain text passwords by default.
MS Windows clients have a habit of dropping network mappings that have been idle for 10 minutes or longer. When the user attempts to use the mapped drive connection that has been dropped, the client re-establishes the connection using a cached copy of the password.
When Microsoft changed the default password mode, support was dropped for caching of the plain text password. This means that when the registry parameter is changed to re-enable use of plain text passwords it appears to work, but when a dropped service connection mapping attempts to revalidate it will fail if the remote authentication server does not support encrypted passwords. This means that it is definitely not a good idea to re-enable plain text password support in such clients.
The following parameters can be used to work around the issue of Windows 9x client upper casing usernames and password before transmitting them to the SMB server when using clear text authentication.
passsword level = integer username level = integer
By default Samba will lower case the username before attempting to lookup the user in the database of local system accounts. Because UNIX usernames conventionally only contain lower case character, the username level parameter is rarely needed.
However, passwords on UNIX systems often make use of mixed case characters. This means that in order for a user on a Windows 9x client to connect to a Samba server using clear text authentication, the password level must be set to the maximum number of upper case letter which could appear is a password. Note that the server OS uses the traditional DES version of crypt(), a password level of 8 will result in case insensitive passwords as seen from Windows users. This will also result in longer login times as Samba has to compute the permutations of the password string and try them one by one until a match is located (or all combinations fail).
The best option to adopt is to enable support for encrypted passwords where ever Samba is used. There are three configuration possibilities for support of encrypted passwords:
This method involves the additions of the following parameters in the smb.conf file:
encrypt passwords = Yes security = server password server = "NetBIOS_name_of_PDC"
There are two ways of identifying whether or not a username and password pair was valid or not. One uses the reply information provided as part of the authentication messaging process, the other uses just an error code.
The down-side of this mode of configuration is the fact that for security reasons Samba will send the password server a bogus username and a bogus password and if the remote server fails to reject the username and password pair then an alternative mode of identification of validation is used. Where a site uses password lock out after a certain number of failed authentication attempts this will result in user lockouts.
Use of this mode of authentication does require there to be a standard Unix account for the user, this account can be blocked to prevent logons by other than MS Windows clients.
When samba is operating in security = domain mode this means that the Samba server has a domain security trust account (a machine account) and will cause all authentication requests to be passed through to the domain controllers.
This method involves addition of the following parameters in the smb.conf file:
encrypt passwords = Yes security = domain workgroup = "name of NT domain" password server = *
The use of the "*" argument to password server will cause samba to locate the domain controller in a way analogous to the way this is done within MS Windows NT. This is the default behaviour.
In order for this method to work the Samba server needs to join the MS Windows NT security domain. This is done as follows:
On the MS Windows NT domain controller using the Server Manager add a machine account for the Samba server.
Next, on the Linux system execute: smbpasswd -r PDC_NAME -j DOMAIN_NAME (samba 2.x) net join -U administrator%password (samba-3)
Use of this mode of authentication does require there to be a standard Unix account for the user in order to assign a uid once the account has been authenticated by the remote Windows DC. This account can be blocked to prevent logons by clients other than MS Windows through things such as setting an invalid shell in the /etc/passwd entry.
An alternative to assigning UIDs to Windows users on a Samba member server is presented in the Winbind Overview chapter in this HOWTO collection.