From 34ab1844469b8e98ad2a2a30e5fd21fb7375f894 Mon Sep 17 00:00:00 2001 From: Gerald Carter Date: Tue, 19 Feb 2002 16:47:59 +0000 Subject: merge from 2.2 (This used to be commit 8b357e6551c3a91aa7017ae8dcf38558f15f1c0b) --- docs/docbook/manpages/smb.conf.5.sgml | 22 +- docs/htmldocs/smb.conf.5.html | 763 +--------------------------------- docs/manpages/smb.conf.5 | 297 +------------ 3 files changed, 38 insertions(+), 1044 deletions(-) diff --git a/docs/docbook/manpages/smb.conf.5.sgml b/docs/docbook/manpages/smb.conf.5.sgml index ac428f88d1..5779f72da5 100644 --- a/docs/docbook/manpages/smb.conf.5.sgml +++ b/docs/docbook/manpages/smb.conf.5.sgml @@ -1068,8 +1068,8 @@ ON DEMAND when a user accesses the Samba server. In order to use this option, smbd - must be set to security = server or - security = domain and add user script + must NOT be set to security = share + and add user script must be set to a full pathname for a script that will create a UNIX user given one argument of %u, which expands into the UNIX user name to create. @@ -1879,19 +1879,11 @@ Windows NT user no longer exists. In order to use this option, smbd must be - set to security = domain and delete - user script must be set to a full pathname for a script - that will delete a UNIX user given one argument of %u - , which expands into the UNIX user name to delete. - NOTE that this is different to the add user script - which will work with the security = server option - as well as security = domain. The reason for this - is only when Samba is a domain member does it get the information - on an attempted user logon that a user no longer exists. In the - security = server mode a missing user - is treated the same as an invalid password logon attempt. Deleting - the user in this circumstance would not be a good idea. + set to security = domain or security = + user and delete user script + must be set to a full pathname for a script + that will delete a UNIX user given one argument of %u, + which expands into the UNIX user name to delete. When the Windows user attempts to access the Samba server, at login (session setup in the SMB protocol) diff --git a/docs/htmldocs/smb.conf.5.html b/docs/htmldocs/smb.conf.5.html index 1438b1436b..f473afa2d2 100644 --- a/docs/htmldocs/smb.conf.5.html +++ b/docs/htmldocs/smb.conf.5.html @@ -1029,54 +1029,6 @@ CLASS="PARAMETER" >

  • character set

  • client code page

  • code page directory

  • coding system

  • valid chars

  • COMPLETE LIST OF SERVICE PARAMETERS

    EXPLANATION OF EACH PARAMETER

    smbd - must be set to security = server or NOT be set to security = domainsecurity = share and + and add user script
    character set (G)

    This allows smbd to map incoming filenames - from a DOS Code page (see the client - code page parameter) to several built in UNIX character sets. - The built in code page translations are:

    • ISO8859-1 : Western European - UNIX character set. The parameter client code page - MUST be set to code page 850 if the - character set parameter is set to - ISO8859-1 in order for the conversion to the - UNIX character set to be done correctly.

    • ISO8859-2 : Eastern European - UNIX character set. The parameter client code page - MUST be set to code page 852 if - the character set parameter is set - to ISO8859-2 in order for the conversion - to the UNIX character set to be done correctly.

    • ISO8859-5 : Russian Cyrillic - UNIX character set. The parameter client code page - MUST be set to code page - 866 if the character set parameter is - set to ISO8859-5 in order for the conversion - to the UNIX character set to be done correctly.

    • ISO8859-7 : Greek UNIX - character set. The parameter client code page - MUST be set to code page - 737 if the character set parameter is - set to ISO8859-7 in order for the conversion - to the UNIX character set to be done correctly.

    • KOI8-R : Alternate mapping - for Russian Cyrillic UNIX character set. The parameter - client code page MUST - be set to code page 866 if the character set - parameter is set to KOI8-R in order for the - conversion to the UNIX character set to be done correctly.

    BUG. These MSDOS code page to UNIX character - set mappings should be dynamic, like the loading of MS DOS code pages, - not static.

    Normally this parameter is not set, meaning no filename - translation is done.

    Default: character set = <empty string>

    Example: character set = ISO8859-1

    client code page (G)

    This parameter specifies the DOS code page - that the clients accessing Samba are using. To determine what code - page a Windows or DOS client is using, open a DOS command prompt - and type the command chcp. This will output - the code page. The default for USA MS-DOS, Windows 95, and - Windows NT releases is code page 437. The default for western - European releases of the above operating systems is code page 850.

    This parameter tells smbd(8) - which of the codepage.XXX - files to dynamically load on startup. These files, - described more fully in the manual page make_smbcodepage(1), tell smbd how to map lower to upper case characters to provide - the case insensitivity of filenames that Windows clients expect.

    Samba currently ships with the following code page files :

    • Code Page 437 - MS-DOS Latin US

    • Code Page 737 - Windows '95 Greek

    • Code Page 850 - MS-DOS Latin 1

    • Code Page 852 - MS-DOS Latin 2

    • Code Page 861 - MS-DOS Icelandic

    • Code Page 866 - MS-DOS Cyrillic

    • Code Page 932 - MS-DOS Japanese SJIS

    • Code Page 936 - MS-DOS Simplified Chinese

    • Code Page 949 - MS-DOS Korean Hangul

    • Code Page 950 - MS-DOS Traditional Chinese

    Thus this parameter may have any of the values 437, 737, 850, 852, - 861, 932, 936, 949, or 950. If you don't find the codepage you need, - read the comments in one of the other codepage files and the - make_smbcodepage(1) man page and write one. Please - remember to donate it back to the Samba user community.

    This parameter co-operates with the valid - chars parameter in determining what characters are - valid in filenames and how capitalization is done. If you set both - this parameter and the valid chars parameter - the client code page parameter - MUST be set before the valid - chars parameter in the smb.conf - file. The valid chars string will then - augment the character settings in the client code page - parameter.

    If not set, client code page defaults - to 850.

    See also : valid - chars, code page directory

    Default: client code page = 850

    Example: client code page = 936

    code page directory (G)

    Define the location of the various client code page - files.

    See also client - code page

    Default: code page directory = ${prefix}/lib/codepages -

    Example: code page directory = /usr/share/samba/codepages -

    coding system (G)

    This parameter is used to determine how incoming - Shift-JIS Japanese characters are mapped from the incoming client code page - used by the client, into file names in the UNIX filesystem. - Only useful if client code page is set to - 932 (Japanese Shift-JIS). The options are :

    • SJIS - Shift-JIS. Does no - conversion of the incoming filename.

    • JIS8, J8BB, J8BH, J8@B, - J8@J, J8@H - Convert from incoming Shift-JIS to eight - bit JIS code with different shift-in, shift out codes.

    • JIS7, J7BB, J7BH, J7@B, J7@J, - J7@H - Convert from incoming Shift-JIS to seven bit - JIS code with different shift-in, shift out codes.

    • JUNET, JUBB, JUBH, JU@B, JU@J, JU@H - - Convert from incoming Shift-JIS to JUNET code with different shift-in, - shift out codes.

    • EUC - Convert an incoming - Shift-JIS character to EUC code.

    • HEX - Convert an incoming - Shift-JIS character to a 3 byte hex representation, i.e. - :AB.

    • CAP - Convert an incoming - Shift-JIS character to the 3 byte hex representation used by - the Columbia AppleTalk Program (CAP), i.e. :AB. - This is used for compatibility between Samba and CAP.

    Default: coding system = <empty value> -

    comment (S)
    security = domain     and delete - user script must be set to a full pathname for a script - that will delete a UNIX user given one argument of %u - , which expands into the UNIX user name to delete. - NOTE that this is different to the add user script - which will work with the or security = serversecurity = + user option - as well as and security = domaindelete user script. The reason for this - is only when Samba is a domain member does it get the information - on an attempted user logon that a user no longer exists. In the - + must be set to a full pathname for a script + that will delete a UNIX user given one argument of security = server%u mode a missing user - is treated the same as an invalid password logon attempt. Deleting - the user in this circumstance would not be a good idea.

    , + which expands into the UNIX user name to delete.

    When the Windows user attempts to access the Samba server, at

    valid chars (G)

    The option allows you to specify additional - characters that should be considered valid by the server in - filenames. This is particularly useful for national character - sets, such as adding u-umlaut or a-ring.

    The option takes a list of characters in either integer - or character form with spaces between them. If you give two - characters with a colon between them then it will be taken as - an lowercase:uppercase pair.

    If you have an editor capable of entering the characters - into the config file then it is probably easiest to use this - method. Otherwise you can specify the characters in octal, - decimal or hexadecimal form using the usual C notation.

    For example to add the single character 'Z' to the charset - (which is a pointless thing to do as it's already there) you could - do one of the following

    		valid chars = Z
-		valid chars = z:Z
-		valid chars = 0132:0172
-

    The last two examples above actually add two characters, - and alter the uppercase and lowercase mappings appropriately.

    Note that you MUST specify this parameter - after the client code page parameter if you - have both set. If client code page is set after - the valid chars parameter the valid - chars settings will be overwritten.

    See also the client - code page parameter.

    Default: Samba defaults to using a reasonable set - of valid characters for English systems

    Example: valid chars = 0345:0305 0366:0326 0344:0304 -

    The above example allows filenames to have the Swedish - characters in them.

    NOTE: It is actually quite difficult to - correctly produce a valid chars line for - a particular system. To automate the process tino@augsburg.net has written - a package called validchars which will automatically - produce a complete valid chars line for - a given client system. Look in the examples/validchars/ - subdirectory of your Samba source code distribution - for this package.

    valid users (S)

    WARNINGS

    VERSION

    SEE ALSO

    AUTHOR

    .\" Please send any bug reports, improvements, comments, patches, .\" etc. to Steve Cheng . -.TH "SMB.CONF" "5" "28 January 2002" "" "" +.TH "SMB.CONF" "5" "19 February 2002" "" "" .SH NAME smb.conf \- The configuration file for the Samba suite .SH "SYNOPSIS" @@ -545,18 +545,6 @@ each parameter for details. Note that some are synonyms. \fIchange share command\fR .TP 0.2i \(bu -\fIcharacter set\fR -.TP 0.2i -\(bu -\fIclient code page\fR -.TP 0.2i -\(bu -\fIcode page directory\fR -.TP 0.2i -\(bu -\fIcoding system\fR -.TP 0.2i -\(bu \fIconfig file\fR .TP 0.2i \(bu @@ -1007,9 +995,6 @@ each parameter for details. Note that some are synonyms. \fIutmp directory\fR .TP 0.2i \(bu -\fIvalid chars\fR -.TP 0.2i -\(bu \fIwinbind cache time\fR .TP 0.2i \(bu @@ -1560,7 +1545,8 @@ Windows NT PDC is an onerous task. This option allows smbdto create the required \fBON DEMAND\fR when a user accesses the Samba server. In order to use this option, smbd -must be set to \fIsecurity = server\fR or \fI security = domain\fR and \fIadd user script\fR +must \fBNOT\fR be set to \fIsecurity = share\fR +and \fIadd user script\fR must be set to a full pathname for a script that will create a UNIX user given one argument of \fI%u\fR, which expands into the UNIX user name to create. @@ -1834,208 +1820,6 @@ Default: \fBnone\fR Example: \fBchange share command = /usr/local/bin/addshare\fR .PP .TP -\fBcharacter set (G)\fR -This allows smbdto map incoming filenames -from a DOS Code page (see the client -code page parameter) to several built in UNIX character sets. -The built in code page translations are: -.RS -.TP 0.2i -\(bu -ISO8859-1 : Western European -UNIX character set. The parameter \fIclient code page\fR -\fBMUST\fR be set to code page 850 if the -\fIcharacter set\fR parameter is set to -ISO8859-1 in order for the conversion to the -UNIX character set to be done correctly. -.TP 0.2i -\(bu -ISO8859-2 : Eastern European -UNIX character set. The parameter \fIclient code page -\fR\fBMUST\fR be set to code page 852 if -the \fI character set\fR parameter is set -to ISO8859-2 in order for the conversion -to the UNIX character set to be done correctly. -.TP 0.2i -\(bu -ISO8859-5 : Russian Cyrillic -UNIX character set. The parameter \fIclient code page -\fR\fBMUST\fR be set to code page -866 if the \fIcharacter set \fR parameter is -set to ISO8859-5 in order for the conversion -to the UNIX character set to be done correctly. -.TP 0.2i -\(bu -ISO8859-7 : Greek UNIX -character set. The parameter \fIclient code page -\fR\fBMUST\fR be set to code page -737 if the \fIcharacter set\fR parameter is -set to ISO8859-7 in order for the conversion -to the UNIX character set to be done correctly. -.TP 0.2i -\(bu -KOI8-R : Alternate mapping -for Russian Cyrillic UNIX character set. The parameter -\fIclient code page\fR \fBMUST\fR -be set to code page 866 if the \fIcharacter set\fR -parameter is set to KOI8-R in order for the -conversion to the UNIX character set to be done correctly. -.RE -.PP -\fBBUG\fR. These MSDOS code page to UNIX character -set mappings should be dynamic, like the loading of MS DOS code pages, -not static. -.PP -.PP -Normally this parameter is not set, meaning no filename -translation is done. -.PP -.PP -Default: \fBcharacter set = \fR -.PP -.PP -Example: \fBcharacter set = ISO8859-1\fR -.PP -.TP -\fBclient code page (G)\fR -This parameter specifies the DOS code page -that the clients accessing Samba are using. To determine what code -page a Windows or DOS client is using, open a DOS command prompt -and type the command \fBchcp\fR. This will output -the code page. The default for USA MS-DOS, Windows 95, and -Windows NT releases is code page 437. The default for western -European releases of the above operating systems is code page 850. - -This parameter tells smbd(8) -which of the \fIcodepage.XXX -\fRfiles to dynamically load on startup. These files, -described more fully in the manual page \fBmake_smbcodepage(1)\fR, tell \fB smbd\fR how to map lower to upper case characters to provide -the case insensitivity of filenames that Windows clients expect. - -Samba currently ships with the following code page files : -.RS -.TP 0.2i -\(bu -Code Page 437 - MS-DOS Latin US -.TP 0.2i -\(bu -Code Page 737 - Windows '95 Greek -.TP 0.2i -\(bu -Code Page 850 - MS-DOS Latin 1 -.TP 0.2i -\(bu -Code Page 852 - MS-DOS Latin 2 -.TP 0.2i -\(bu -Code Page 861 - MS-DOS Icelandic -.TP 0.2i -\(bu -Code Page 866 - MS-DOS Cyrillic -.TP 0.2i -\(bu -Code Page 932 - MS-DOS Japanese SJIS -.TP 0.2i -\(bu -Code Page 936 - MS-DOS Simplified Chinese -.TP 0.2i -\(bu -Code Page 949 - MS-DOS Korean Hangul -.TP 0.2i -\(bu -Code Page 950 - MS-DOS Traditional Chinese -.RE -.PP -Thus this parameter may have any of the values 437, 737, 850, 852, -861, 932, 936, 949, or 950. If you don't find the codepage you need, -read the comments in one of the other codepage files and the -\fBmake_smbcodepage(1)\fR man page and write one. Please -remember to donate it back to the Samba user community. -.PP -.PP -This parameter co-operates with the \fIvalid -chars\fR parameter in determining what characters are -valid in filenames and how capitalization is done. If you set both -this parameter and the \fIvalid chars\fR parameter -the \fIclient code page\fR parameter -\fBMUST\fR be set before the \fIvalid -chars\fR parameter in the \fIsmb.conf\fR -file. The \fIvalid chars\fR string will then -augment the character settings in the \fIclient code page\fR -parameter. -.PP -.PP -If not set, \fIclient code page\fR defaults -to 850. -.PP -.PP -See also : \fIvalid -chars\fR, \fIcode page directory\fR -.PP -.PP -Default: \fBclient code page = 850\fR -.PP -.PP -Example: \fBclient code page = 936\fR -.PP -.TP -\fBcode page directory (G)\fR -Define the location of the various client code page -files. - -See also \fIclient -code page\fR - -Default: \fBcode page directory = ${prefix}/lib/codepages -\fR -Example: \fBcode page directory = /usr/share/samba/codepages -\fR.TP -\fBcoding system (G)\fR -This parameter is used to determine how incoming -Shift-JIS Japanese characters are mapped from the incoming \fIclient code page\fR -used by the client, into file names in the UNIX filesystem. -Only useful if \fIclient code page\fR is set to -932 (Japanese Shift-JIS). The options are : -.RS -.TP 0.2i -\(bu -SJIS - Shift-JIS. Does no -conversion of the incoming filename. -.TP 0.2i -\(bu -JIS8, J8BB, J8BH, J8@B, -J8@J, J8@H - Convert from incoming Shift-JIS to eight -bit JIS code with different shift-in, shift out codes. -.TP 0.2i -\(bu -JIS7, J7BB, J7BH, J7@B, J7@J, -J7@H - Convert from incoming Shift-JIS to seven bit -JIS code with different shift-in, shift out codes. -.TP 0.2i -\(bu -JUNET, JUBB, JUBH, JU@B, JU@J, JU@H -- Convert from incoming Shift-JIS to JUNET code with different shift-in, -shift out codes. -.TP 0.2i -\(bu -EUC - Convert an incoming -Shift-JIS character to EUC code. -.TP 0.2i -\(bu -HEX - Convert an incoming -Shift-JIS character to a 3 byte hex representation, i.e. -:AB. -.TP 0.2i -\(bu -CAP - Convert an incoming -Shift-JIS character to the 3 byte hex representation used by -the Columbia AppleTalk Program (CAP), i.e. :AB. -This is used for compatibility between Samba and CAP. -.RE -.PP -Default: \fBcoding system = \fR -.PP -.TP \fBcomment (S)\fR This is a text field that is seen next to a share when a client does a queries the server, either via the network @@ -2350,18 +2134,11 @@ DEMAND\fR when a user accesses the Samba server and the Windows NT user no longer exists. In order to use this option, \fBsmbd\fR must be -set to \fIsecurity = domain\fR and \fIdelete -user script\fR must be set to a full pathname for a script -that will delete a UNIX user given one argument of \fI%u -\fR, which expands into the UNIX user name to delete. -\fBNOTE\fR that this is different to the \fIadd user script\fR -which will work with the \fIsecurity = server\fR option -as well as \fIsecurity = domain\fR. The reason for this -is only when Samba is a domain member does it get the information -on an attempted user logon that a user no longer exists. In the -\fIsecurity = server\fR mode a missing user -is treated the same as an invalid password logon attempt. Deleting -the user in this circumstance would not be a good idea. +set to \fIsecurity = domain\fR or \fIsecurity = +user\fR and \fIdelete user script\fR +must be set to a full pathname for a script +that will delete a UNIX user given one argument of \fI%u\fR, +which expands into the UNIX user name to delete. When the Windows user attempts to access the Samba server, at \fBlogin\fR (session setup in the SMB protocol) @@ -6989,64 +6766,6 @@ native system is set to use (usually Default: \fBno utmp directory\fR .TP -\fBvalid chars (G)\fR -The option allows you to specify additional -characters that should be considered valid by the server in -filenames. This is particularly useful for national character -sets, such as adding u-umlaut or a-ring. - -The option takes a list of characters in either integer -or character form with spaces between them. If you give two -characters with a colon between them then it will be taken as -an lowercase:uppercase pair. - -If you have an editor capable of entering the characters -into the config file then it is probably easiest to use this -method. Otherwise you can specify the characters in octal, -decimal or hexadecimal form using the usual C notation. - -For example to add the single character 'Z' to the charset -(which is a pointless thing to do as it's already there) you could -do one of the following - -.sp -.nf - valid chars = Z - valid chars = z:Z - valid chars = 0132:0172 - -.sp -.fi - -The last two examples above actually add two characters, -and alter the uppercase and lowercase mappings appropriately. - -Note that you \fBMUST\fR specify this parameter -after the \fIclient code page\fR parameter if you -have both set. If \fIclient code page\fR is set after -the \fIvalid chars\fR parameter the \fIvalid -chars\fR settings will be overwritten. - -See also the \fIclient -code page\fR parameter. - -Default: \fBSamba defaults to using a reasonable set -of valid characters for English systems\fR - -Example: \fBvalid chars = 0345:0305 0366:0326 0344:0304 -\fR -The above example allows filenames to have the Swedish -characters in them. - -\fBNOTE:\fR It is actually quite difficult to -correctly produce a \fIvalid chars\fR line for -a particular system. To automate the process tino@augsburg.net has written -a package called \fBvalidchars\fR which will automatically -produce a complete \fIvalid chars\fR line for -a given client system. Look in the \fIexamples/validchars/ -\fRsubdirectory of your Samba source code distribution -for this package. -.TP \fBvalid users (S)\fR This is a list of users that should be allowed to login to this service. Names starting with '@', '+' and '&' -- cgit