20 years agoChanges from APPLIANCE_HEAD:
David O'Neill [Wed, 22 Nov 2000 16:19:07 +0000 (16:19 +0000)]
Changes from APPLIANCE_HEAD:
    - Add code to test equivalence of private data in NT_DEVICEMODE
(This used to be commit 684981851ffa3b51e78a6fd5960e219823eb90d5)

20 years agoADDPRINTERDRIVER info level 6 fix - this is dynamic depending
Jeremy Allison [Wed, 22 Nov 2000 01:26:09 +0000 (01:26 +0000)]
ADDPRINTERDRIVER info level 6 fix - this is dynamic depending
on a field in the struct - that's why it was so hard to find.
(This used to be commit 0ab6eb01b0d15e5eb744333d2bd79c27451f50e3)

20 years agoDon't forget to convert into UNIX character set before calling
Jeremy Allison [Tue, 21 Nov 2000 22:37:03 +0000 (22:37 +0000)]
Don't forget to convert into UNIX character set before calling
(This used to be commit 00cd72c385f1e5d075dbacf834b68769b5ac38f3)

20 years agocombined 2 if statments which used the same condition
Gerald Carter [Tue, 21 Nov 2000 06:05:08 +0000 (06:05 +0000)]
combined 2 if statments which used the same condition

-- jerry
(This used to be commit 445fd1dbd8bb93f56f20b5dd9e9d5b018147b21d)

20 years agomake proto
Gerald Carter [Tue, 21 Nov 2000 06:01:53 +0000 (06:01 +0000)]
make proto
(This used to be commit e0a02249d394a530d911d9b650c6eb56c188a6d8)

20 years agoAnother large patch for the passdb rewrite.
Gerald Carter [Tue, 21 Nov 2000 05:55:16 +0000 (05:55 +0000)]
Another large patch for the passdb rewrite.

  o added BOOL own_memory flag in SAM_ACCOUNT so we could
    use static memory for string pointer assignment or
    allocate a new string

  o added a reference TDB passdb backend.  This is only a reference
    and should not be used in production because
- RID's are generated using the same algorithm as with smbpasswd
  - a TDB can only have one key (w/o getting into problems) and we
  need three.  Therefore the pdb_sam-getpwuid() and
  pdb_getsampwrid() functions are interative searches :-(

    we need transaction support, multiple indexes, and a nice open
    source DBM.  The Berkeley DB (from seems to fit
    this criteria now)

  o added a new parameter "private dir" as many places in the code were
    using lp_smb_passwd_file() and chopping off the filename part.
    This makes more sense to me and I will docuement it in the man pages

  o Ran through Insure-lite and corrected memory leaks.  Need for
    a public flogging this time Jeremy (-:

-- jerry
(This used to be commit 4792029a2991bd84251d152a62b1033dec62cee2)

20 years agoFix for off-by-one StrnCpy.
Jeremy Allison [Tue, 21 Nov 2000 02:38:22 +0000 (02:38 +0000)]
Fix for off-by-one StrnCpy.
(This used to be commit 63cafb34b39443d03d17ae01b47adc0926b05fe2)

20 years agoFix for updating of print queues changed from a local box. Essentially,
Jeremy Allison [Tue, 21 Nov 2000 00:30:15 +0000 (00:30 +0000)]
Fix for updating of print queues changed from a local box. Essentially,
this makes sure that the change messages sent to ourselves are handled
synchronously w.r.t. other smb packets incoming.
(This used to be commit 78a13074455618308d048d1c69f62e660988eb90)

20 years agoFix from John Reilly for equivalence tests.
Jeremy Allison [Fri, 17 Nov 2000 23:10:56 +0000 (23:10 +0000)]
Fix from John Reilly for equivalence tests.
(This used to be commit b456274a7ba22d512bab5ede979cd9363398825b)

20 years agowe don't need the separate lp_status() connection records any more
Andrew Tridgell [Fri, 17 Nov 2000 03:31:03 +0000 (03:31 +0000)]
we don't need the separate lp_status() connection records any more
(This used to be commit 209e20365e562371aafafea301b4ffecc3d4c3ed)

20 years agofixed the problem with messages not getting through
Andrew Tridgell [Fri, 17 Nov 2000 03:15:18 +0000 (03:15 +0000)]
fixed the problem with messages not getting through

the problem had nothing to do with being your own pid, it was instead
a problem with IPC$ connections not being registered in the
connections database and an incorrect test for -1 in the messaging

These changes also mean that IPC$ shares now show up in
smbstatus. That is probably a good thing.
(This used to be commit 3575ad10985a18f897e38179ca69fa9a49a7ea02)

20 years agoFix for memory leak when adding driver.
Jeremy Allison [Fri, 17 Nov 2000 02:22:35 +0000 (02:22 +0000)]
Fix for memory leak when adding driver.
(This used to be commit eeab4e0290f9df84025e91c85d27b21e0c02781f)

20 years agosupport smbcontrol sending messages to itself (for testing purposes)
Andrew Tridgell [Fri, 17 Nov 2000 01:20:03 +0000 (01:20 +0000)]
support smbcontrol sending messages to itself (for testing purposes)
(This used to be commit 5dfae86dab50fba1890d0ce33fa9d68f12bc0bfa)

20 years agosave and restore errno in select
Andrew Tridgell [Fri, 17 Nov 2000 01:05:50 +0000 (01:05 +0000)]
save and restore errno in select
(This used to be commit 34f0379096d0701c74a51c51649ffe4cb1a24291)

20 years agoDelete queue on empty.
Jeremy Allison [Fri, 17 Nov 2000 00:48:02 +0000 (00:48 +0000)]
Delete queue on empty.
(This used to be commit 6e18a2aa58bc485e3c803ff357acc1b7fe6d95e1)

20 years agoJohn Reillys fix for de-registering broadcast names (NT doesn't do this).
Jeremy Allison [Fri, 17 Nov 2000 00:33:06 +0000 (00:33 +0000)]
John Reillys fix for de-registering broadcast names (NT doesn't do this).
(This used to be commit 245907f2affb530237809b81b8748f7f0a1e4502)

20 years agoFix for incorrect break using wrong offset when updating a form.
Jeremy Allison [Fri, 17 Nov 2000 00:31:29 +0000 (00:31 +0000)]
Fix for incorrect break using wrong offset when updating a form.
(This used to be commit 365c76e39d2397a12ac6785a51c4d1529527a356)

20 years agothe duplicate checking code will cause unaligned accesses on non-intel
Andrew Tridgell [Fri, 17 Nov 2000 00:08:42 +0000 (00:08 +0000)]
the duplicate checking code will cause unaligned accesses on non-intel
processors. Fixed.
(This used to be commit 64d38c24100cb3409b38b9923734f2a6202bdc2a)

20 years agouse process_exists() not kill(pid, 0)
Andrew Tridgell [Fri, 17 Nov 2000 00:02:16 +0000 (00:02 +0000)]
use process_exists() not kill(pid, 0)
(This used to be commit 30048cff12e03c95ef43ba4ee16af1df2de9dbc8)

20 years agoFix for a problem with the new messaging system. If a sender is using the
Jeremy Allison [Thu, 16 Nov 2000 21:38:24 +0000 (21:38 +0000)]
Fix for a problem with the new messaging system. If a sender is using the
messaging system as a notification mechanism, and the speed of notification
greatly exceeds the speed of message recovery, then you get a massively (>75Mb)
growing tdb. If the message is a simple notification, then the message is
static, and you only need one of them in transit to a target process at
any one time.
This patch adds a BOOL "allow_duplicates" to the message_send_XX primitives.
If set to False, then before sending a message the sender checks the existing
message queue for a target pid for a duplicate of this message, and doesn't
add to it if one already exists.
Also added code into msgtest.c to test this.
(This used to be commit 3aa7995660395ecb85c8e35b638fa9fbbb952558)

20 years agoFix for plaintext passwords from Pat Sandfort @ HP.
Jeremy Allison [Thu, 16 Nov 2000 21:33:21 +0000 (21:33 +0000)]
Fix for plaintext passwords from Pat Sandfort @ HP.
(This used to be commit b8753b92fbeb1d6768d0559e12ff2aa1d0148419)

20 years agoFix from Jim McDonough @ IBM for OS/2 clients.
Jeremy Allison [Thu, 16 Nov 2000 19:08:55 +0000 (19:08 +0000)]
Fix from Jim McDonough @ IBM for OS/2 clients.
(This used to be commit f571e1efd01c7b1b500a833df3bd074a8c4c65ec)

20 years agoChanges from APPLIANCE_HEAD:
David O'Neill [Thu, 16 Nov 2000 18:54:42 +0000 (18:54 +0000)]
Changes from APPLIANCE_HEAD:
- add empty README file for vlp.  Blame Tim.  =)
(This used to be commit 0ee469f566dbfdac1769fefbb5e61ce39d267872)

20 years agoOk - fixed a bug in our levelII oplock code. We need to break a level II on
Jeremy Allison [Thu, 16 Nov 2000 00:59:18 +0000 (00:59 +0000)]
Ok - fixed a bug in our levelII oplock code. We need to break a level II on
a byte range lock (write lock only, but Win2k breaks on read lock also so I
do the same) - if you think about why, this is obvious. Also fixed our client
code to do level II oplocks, if requested, and fixed the code where we would
assume the client wanted level II if it advertised itself as being level II
capable - it may not want that.
(This used to be commit 213cd0b5192307cd4b0026cae94b2f52fb1b0c02)

20 years agoUninitialized memory read fixes.
Jeremy Allison [Wed, 15 Nov 2000 02:39:11 +0000 (02:39 +0000)]
Uninitialized memory read fixes.
open_file_shared takes a DOS pathname, not a UNIX one.
(This used to be commit b2b59b93ff1c5bdc43b099e2db1ee86fe8514152)

20 years agoTuyrn debug timestamps on by default.
Jeremy Allison [Wed, 15 Nov 2000 01:11:38 +0000 (01:11 +0000)]
Tuyrn debug timestamps on by default.
Add Tim's lpq race fix.
(This used to be commit d43405bc47c95bf8c906035cba23bf95d252d13b)

20 years agoMerge from appliance head of JR's changes for driver versioning.
Jeremy Allison [Tue, 14 Nov 2000 21:56:32 +0000 (21:56 +0000)]
Merge from appliance head of JR's changes for driver versioning.
(This used to be commit cdbd2e99775642dc2e92004be9014bf38a92d80f)

20 years ago- fix "declaration of 'time' shadows global declaration" warning.
David O'Neill [Tue, 14 Nov 2000 15:50:30 +0000 (15:50 +0000)]
- fix "declaration of 'time' shadows global declaration" warning.
(This used to be commit 92ff07132b3834b469ad7bb73d6e714b175a12af)

20 years agoChanges from APPLIANCE_HEAD:
David O'Neill [Tue, 14 Nov 2000 15:26:53 +0000 (15:26 +0000)]
Changes from APPLIANCE_HEAD:
- merged Tim's vlp (virtual lp) test program.  Enable it with
      -DDEVELOPER or by using ./configure.developer
  (source/include/smb.h source/configure.developer
   source/printing/lpq_parse.c source/param/loadparm.c
   testsuite/printing/.cvsignore testsuite/printing/Makefile.vlp
(This used to be commit fbcf83140da1823e74f63227f0a95d07c6e76764)

20 years agoprinting/nt_printing.c: After long soul searching and making both Andrew and my
Jeremy Allison [Tue, 14 Nov 2000 02:14:58 +0000 (02:14 +0000)]
printing/nt_printing.c: After long soul searching and making both Andrew and my
life a misery, here is the only possible null driver fix we have found.
This *SUCKS*.
rpc_server/srv_spoolss_nt.c: Correct printername search. Correct portname reply
Correct attributes reply. Removal of unused temp variable.
(This used to be commit 06e71c9f8be20b84e33e143c3d0b7904225efb45)

20 years agoFixed the cut-n-paste bugs in the new passdb backend code that leaked memory.
Jeremy Allison [Tue, 14 Nov 2000 01:40:40 +0000 (01:40 +0000)]
Fixed the cut-n-paste bugs in the new passdb backend code that leaked memory.
(This used to be commit 55c6acce26a26af68149865dafb42e5a03b497e0)

20 years agoTims fixes for non-local lookup.
Jeremy Allison [Mon, 13 Nov 2000 23:37:53 +0000 (23:37 +0000)]
Tims fixes for non-local lookup.
(This used to be commit 38e04c43cc2ce8479b8cc3ff0f87d3d150d9af83)

20 years agoLarge commit which restructures the local password storage API.
Gerald Carter [Mon, 13 Nov 2000 23:03:34 +0000 (23:03 +0000)]
Large commit which restructures the local password storage API.
Currently the only backend which works is smbpasswd (tdb, LDAP, and NIS+)
are broken, but they were somewhat broken before. :)

The following functions implement the storage manipulation interface

/*The following definitions come from  passdb/pdb_smbpasswd.c  */

BOOL pdb_setsampwent (BOOL update);
void pdb_endsampwent (void);
SAM_ACCOUNT* pdb_getsampwent (void);
SAM_ACCOUNT* pdb_getsampwnam (char *username);
SAM_ACCOUNT* pdb_getsampwuid (uid_t uid);
SAM_ACCOUNT* pdb_getsampwrid (uint32 rid);
BOOL pdb_add_sam_account (SAM_ACCOUNT *sampass);
BOOL pdb_update_sam_account (SAM_ACCOUNT *sampass, BOOL override);
BOOL pdb_delete_sam_account (char* username);

There is also a host of pdb_set..() and pdb_get..() functions for
manipulating SAM_ACCOUNT struct members.  Note that the struct
passdb_ops {} has gone away.  Also notice that struct smb_passwd
(formally in smb.h) has been moved to passdb/pdb_smbpasswd.c
and is not accessed outisde of static internal functions in this
file.  All local password searches should make use of the the SAM_ACCOUNT
struct and the previously mentioned functions.

I'll write some documentation for this later.  The next step is to fix
the TDB passdb backend, then work on spliting the backends out into
share libraries, and finally get the LDAP backend going.

What works and may not:

o domain logons from Win9x  works
o domain logons from WinNT 4 works
o user and group enumeration
as implemented by Tim works
o file and print access works
o changing password from
Win9x & NT ummm...i'll fix this tonight :)

If I broke anything else, just yell and I'll fix it.  I think it
should be fairly quite.

-- jerry
(This used to be commit 0b92d0838ebdbe24f34f17e313ecbf61a0301389)

20 years agoFetch authentication info before actually using it.
Tim Potter [Sat, 11 Nov 2000 01:21:31 +0000 (01:21 +0000)]
Fetch authentication info before actually using it.
(This used to be commit 600eb0eb0036631b7e6d8aa7cddbc378368c002c)

20 years agoMerge of Herb's profiling code.
Jeremy Allison [Sat, 11 Nov 2000 00:33:33 +0000 (00:33 +0000)]
Merge of Herb's profiling code.
(This used to be commit 3be056c71aa8e0a4ba70d397107199004bdb7d3f)

20 years agoMerge in Herb's changes from 2.2.
Jeremy Allison [Fri, 10 Nov 2000 22:07:57 +0000 (22:07 +0000)]
Merge in Herb's changes from 2.2.
(This used to be commit 24d76c5fbda29d89c96d7c22193ec2eb93ad3887)

20 years agoAdded Tim's changes to lock DB during update.
Jeremy Allison [Fri, 10 Nov 2000 22:05:08 +0000 (22:05 +0000)]
Added Tim's changes to lock DB during update.
(This used to be commit f25b5798b8610de9748ae90ea1bf813b125f517d)

20 years agorpc_parse/parse_spoolss.c: Updated comment for old version of W2K.
Jeremy Allison [Fri, 10 Nov 2000 21:24:09 +0000 (21:24 +0000)]
rpc_parse/parse_spoolss.c: Updated comment for old version of W2K.
tdb/tdbutil.c: With varargs uint16 is cast to (int).
(This used to be commit 98764c5df9f363780427a423ead0913f7b2a0905)

20 years agoprinting/nt_printing.c: use getpwuid not smbgetpwuid. Canonicalize printernames.
Jeremy Allison [Fri, 10 Nov 2000 19:36:34 +0000 (19:36 +0000)]
printing/nt_printing.c: use getpwuid not smbgetpwuid. Canonicalize printernames.
printing/printing.c: Insure fix for malloc of zero.
rpc_parse/parse_misc.c: Enusre UNISTR's are zero filled.
rpc_parse/parse_spoolss.c: Correct INFO_6 - differs between pre-releases of W2K and shipping build.
rpc_server/srv_spoolss_nt.c: Canonicalize printernames.
(This used to be commit b17e23a8ff2b44540726968355a4b7e26f244f3b)

20 years agoFix from John E. Malmberg <> for -1 return in interfaces scan.
Jeremy Allison [Fri, 10 Nov 2000 19:02:32 +0000 (19:02 +0000)]
Fix from John E. Malmberg <> for -1 return in interfaces scan.
(This used to be commit 4d25a53c36ad2c33cc2ecaf1486e18f1536eff95)

20 years agoadded more explanation about the required permissions (and authorization)
Gerald Carter [Fri, 10 Nov 2000 15:33:15 +0000 (15:33 +0000)]
added more explanation about the required permissions (and authorization)
in order to add printer drivers to a samba server.

(This used to be commit b4e2fc0bd223993e61f215c89167e8549254c214)

20 years agominor update to help file
Herb Lewis [Thu, 9 Nov 2000 21:23:51 +0000 (21:23 +0000)]
minor update to help file
(This used to be commit 99b764e1c7d508e0caf2ffa82f279dff4fbb5aac)

20 years agorestructure PCP metrics (merge from 2.2 branch)
Herb Lewis [Thu, 9 Nov 2000 20:45:09 +0000 (20:45 +0000)]
restructure PCP metrics (merge from 2.2 branch)
(This used to be commit dc725c2256ff5f3d8ac37fe33cfa1685194f7f46)

20 years agoan attempt to get the handling of fields in printer info structures
Andrew Tridgell [Wed, 8 Nov 2000 03:12:16 +0000 (03:12 +0000)]
an attempt to get the handling of fields in printer info structures
consistent. Still working with Jeremy on this, there is probably more
to be done
(This used to be commit c4bb9c598cf9781d48bc123a8cbbed9c2049bf89)

20 years agoMerge fest!!!
Tim Potter [Wed, 8 Nov 2000 00:20:26 +0000 (00:20 +0000)]
Merge fest!!!
(This used to be commit 87775074b22bb969c3585556e2a86ac4c7d4b0d4)

20 years agomake proto
Tim Potter [Wed, 8 Nov 2000 00:19:30 +0000 (00:19 +0000)]
make proto
(This used to be commit 88a2d224f1c3ca09abbad7f33b0e93bdf503da4e)

20 years agoMore merging of printing stuff from appliance.
Tim Potter [Wed, 8 Nov 2000 00:16:38 +0000 (00:16 +0000)]
More merging of printing stuff from appliance.
(This used to be commit 18fa724a7969666dd5aa176af187054abc94bfd3)

20 years agoMerge of latest round of printing fixes from appliance branch.
Tim Potter [Tue, 7 Nov 2000 23:05:53 +0000 (23:05 +0000)]
Merge ... latest round of printing fixes from appliance branch.
(This used to be commit 8fe17fd59488814cdbe9abcfe53ab2627901e421)

20 years agosync with 2.2 changes
Herb Lewis [Tue, 7 Nov 2000 19:46:43 +0000 (19:46 +0000)]
sync with 2.2 changes
(This used to be commit 0edc78937f4c92a0d18bba9a433a1bc2b00ce9b8)

20 years agomake proto
Tim Potter [Tue, 7 Nov 2000 17:48:54 +0000 (17:48 +0000)]
make proto
(This used to be commit 6ced1bb8b92101e11c53467d0380a97b840de7c7)

20 years agoMerge of printer security descriptor, info level and printerdata
Tim Potter [Tue, 7 Nov 2000 02:54:50 +0000 (02:54 +0000)]
Merge ... printer security descriptor, info level and printerdata
comparison changes from appliance branch.
(This used to be commit ae087bdf312806e08848695cad70a943bb3d71b9)

20 years ago64 bit fix from Uros Prestor <>.
Jeremy Allison [Mon, 6 Nov 2000 23:14:59 +0000 (23:14 +0000)]
64 bit fix from Uros Prestor <>.
(This used to be commit 9a5471b3e861aa864ffff5fc87ac2681de7b0068)

20 years agoFix to the "known/unknown user" difference patch from "Ron Alexander" <rcalex@home...
Jeremy Allison [Mon, 6 Nov 2000 23:08:00 +0000 (23:08 +0000)]
Fix to the "known/unknown user" difference patch from "Ron Alexander" <>.
(This used to be commit 7a698c1f23ea8740755c544f631789848a360e42)

20 years agoAdded a VFS version return to init call. Allows smbd to fail an init if
Jeremy Allison [Mon, 6 Nov 2000 21:44:33 +0000 (21:44 +0000)]
Added a VFS version return to init call. Allows smbd to fail an init if
versions don't match.
(This used to be commit d0fbb4f5d999abade8930cc6fff231a2af6cccfb)

20 years agoAdded a VFS version return to init call. Allows smbd to fail an init if
Jeremy Allison [Mon, 6 Nov 2000 21:33:49 +0000 (21:33 +0000)]
Added a VFS version return to init call. Allows smbd to fail an init if
versions don't match.
(This used to be commit b29ff816734c0424f69124feb316da13f2e094f7)

20 years agoFixes for sample VFS audit code from "Brad Sahr" <>.
Jeremy Allison [Mon, 6 Nov 2000 20:01:03 +0000 (20:01 +0000)]
Fixes for sample VFS audit code from "Brad Sahr" <>.
(This used to be commit 5d6dec3a94f496651a5f0643ab228c975911e6cd)

20 years agoAdded documentation for printer-notify smbcontrol command.
Tim Potter [Sun, 5 Nov 2000 23:06:13 +0000 (23:06 +0000)]
Added documentation for printer-notify smbcontrol command.
(This used to be commit fdf6a5c76abc38e16cb7ee3f31512bfc1f639651)

20 years agoMerge from Tim - don't reinsert UNIX job if already exists.
Jeremy Allison [Sat, 4 Nov 2000 19:48:53 +0000 (19:48 +0000)]
Merge from Tim - don't reinsert UNIX job if already exists.
(This used to be commit a33c395c98d80f620257fed480d5824fc609d9c9)

20 years agoFix some compiler warnings.
Tim Potter [Sat, 4 Nov 2000 18:44:34 +0000 (18:44 +0000)]
Fix some compiler warnings.
(This used to be commit 852de9226d50ccac71ec1691052a6e395283ca56)

20 years agoMerge of printing improvements/fixes from appliance branch.
Tim Potter [Sat, 4 Nov 2000 18:24:15 +0000 (18:24 +0000)]
Merge ... printing improvements/fixes from appliance branch.
(This used to be commit 8e163eac33e8f407de00c318d3de29fcfbf2b8a7)

20 years agoFix for admin user being ignored.
Jeremy Allison [Sat, 4 Nov 2000 06:45:26 +0000 (06:45 +0000)]
Fix for admin user being ignored.
(This used to be commit f0dcc39d34202ed67d778ff40166856d4c2ad87b)

20 years agoSync up with Tim's changes in applience-head for SACLs.
Jeremy Allison [Sat, 4 Nov 2000 06:36:18 +0000 (06:36 +0000)]
Sync up with Tim's changes in applience-head for SACLs.
(This used to be commit 6b94273fc61b26e83c8d791bc626fd52055ef293)

20 years agoAdded Using Samba book reference back into welcome page.
John Terpstra [Thu, 2 Nov 2000 19:42:02 +0000 (19:42 +0000)]
Added Using Samba book reference back into welcome page.
(This used to be commit bb552f7eaeb87d1d09704122da756d3d3e21697a)

20 years agoRemove CGI logging code. Make username/password lookups take
Jeremy Allison [Wed, 1 Nov 2000 19:43:53 +0000 (19:43 +0000)]
Remove CGI logging code. Make username/password lookups take
the same time.
(This used to be commit 9698f746858f0f2b37ec6494fca990e958d97cb7)

20 years agodocumentation update and general description
Gerald Carter [Wed, 1 Nov 2000 18:28:22 +0000 (18:28 +0000)]
documentation update and general description

(This used to be commit f2cc34519504efb12526a5ddad9341edb02ac762)

20 years agoAdded a authentication hook to smbwrapper which allows a (username,
Tim Potter [Wed, 1 Nov 2000 04:31:19 +0000 (04:31 +0000)]
Added a authentication hook to smbwrapper which allows a (username,
workgroup, password) tuple to be provided by another function.
(This used to be commit 644c78d64a6d7b7e44414c579de660321a0a6b34)

20 years agoAdded smbsh to list of ignored files.
Tim Potter [Wed, 1 Nov 2000 03:22:38 +0000 (03:22 +0000)]
Added smbsh to list of ignored files.
(This used to be commit 3ff75e92ad2f663d1306048d132daa486ae02d96)

20 years agoMerge of comment repair.
Tim Potter [Mon, 30 Oct 2000 21:55:30 +0000 (21:55 +0000)]
Merge of comment repair.
(This used to be commit 7712421910b12fd64cf90100e4495669bec67c2f)

20 years agoOnly use prs_uint8s on marshalling as we know we're little endian. On
Jeremy Allison [Mon, 30 Oct 2000 21:47:20 +0000 (21:47 +0000)]
Only use prs_uint8s on marshalling as we know we're little endian. On
unmarshalling use prs_uint16s. Yes this sucks :-).
(This used to be commit 3d68257850be66c26074a43d92eb9c3c8ffaf04e)

20 years agoFixed double btyte-swap problems with bigendian architectures.
Jeremy Allison [Mon, 30 Oct 2000 19:45:22 +0000 (19:45 +0000)]
Fixed double btyte-swap problems with bigendian architectures.
(This used to be commit c96a77957be49418b62e401db49235a11d711217)

20 years agoCall standard_sub_advanced() after lp_logon_XX() calls.
Jeremy Allison [Mon, 30 Oct 2000 19:35:43 +0000 (19:35 +0000)]
Call standard_sub_advanced() after lp_logon_XX() calls.
(This used to be commit d5be2334d4fb007e39565b771cbe68e338a7fe68)

20 years agoproto.h was out of date.
Tim Potter [Sun, 29 Oct 2000 23:52:21 +0000 (23:52 +0000)]
proto.h was out of date.
(This used to be commit 63e4f629163babc1dcba8a419ae432057e46f6f8)

20 years agoPatches from John Reilly @ HP for print job time bugs.
Jeremy Allison [Sun, 29 Oct 2000 17:27:41 +0000 (17:27 +0000)]
Patches from John Reilly @ HP for print job time bugs.
(This used to be commit 0eb7f2d514317882684e9c0c51d46b1ac1eb2568)

20 years agoAnother patch to fix cli_reestablish_connection from Kenichi Okuyama@Tokyo Research...
Jeremy Allison [Sat, 28 Oct 2000 20:54:45 +0000 (20:54 +0000)]
Another patch to fix cli_reestablish_connection from Kenichi Okuyama@Tokyo Research Lab. IBM-Japan. Co. Jp.
(This used to be commit 06f5da5d4bf044969364afe0298347811fb4ae91)

20 years agoDavid Lee's utmp patch (finally). Thanks David !
Jeremy Allison [Sat, 28 Oct 2000 19:38:39 +0000 (19:38 +0000)]
David Lee's utmp patch (finally). Thanks David !
(This used to be commit b809a2d0c81c54e917ccc0c99b3e70ea8d7ceab1)

20 years agoFixed silly bug in dup2 code found by Kenichi Okuyama@Tokyo Research Lab. IBM-Japan...
Jeremy Allison [Sat, 28 Oct 2000 19:30:21 +0000 (19:30 +0000)]
Fixed silly bug in dup2 code found by Kenichi Okuyama@Tokyo Research Lab. IBM-Japan. Co. Jp.
(This used to be commit 9c2272e056aef741c4b86f9a247c3534944d9eff)

20 years agoRolled back find NetBIOS name of PDC/BDC code as a temp fix. This
Jeremy Allison [Fri, 27 Oct 2000 03:02:22 +0000 (03:02 +0000)]
Rolled back find NetBIOS name of PDC/BDC code as a temp fix. This
code works :-).
(This used to be commit 4f66eda11e0dc15ff04893da7b7d6e578a30c4dc)

20 years agoAdded John Reillys patch for error code returns from clean_up_printer_driver_strunct()
Jeremy Allison [Thu, 26 Oct 2000 22:09:22 +0000 (22:09 +0000)]
Added John Reillys patch for error code returns from clean_up_printer_driver_strunct()
(This used to be commit f81a2a03bf435e65e7484ab021f86a8a4f62b656)

20 years agoSorry JF - no billable hours :-). I fixed the "stream of events" problem
Jeremy Allison [Thu, 26 Oct 2000 21:43:13 +0000 (21:43 +0000)]
Sorry JF - no billable hours :-). I fixed the "stream of events" problem
with PCL drivers. The problem was we were updating the changeid on every
SETPRINTERDATA/DELETEPRINTERDATA call. We should not do this, we should
just update the 'setprinter' called count. We update the changeid on calls
to SETPRINTER/ADDPRINTER/ADDPRINTEREX etc. Also fixed the correct returning
of the create time on printers.
(This used to be commit 521f09829fd329f87b3d19e8871e2b989c98a58e)

20 years agoTDB password backend support written by Simo Sorce <>
Gerald Carter [Thu, 26 Oct 2000 03:31:41 +0000 (03:31 +0000)]
TDB password backend support written by Simo Sorce <>
Marked as an experimental compile time option (defaults to off) for now.

(This used to be commit 0435af4417b876c2ea1dd4591ae7647784c28e30)

20 years agoEven when looking up a users groups via winbindd even if the lookup for
Jeremy Allison [Wed, 25 Oct 2000 23:32:12 +0000 (23:32 +0000)]
Even when looking up a users groups via winbindd even if the lookup for
the list of groups a user is in succeeds via winbind, we must allow the
lookup of the group name -> gid we are checking if the user is a member
of to go via winbind or /etc/group - as it may be a group on the local
box we are checking against. This is a subtle one.....
(This used to be commit 4ffda462b97e4f35c6d050c579dfe3e3b64e2c9f)

20 years agoWe already have a perfectly good next_token() function we should be using
Jeremy Allison [Wed, 25 Oct 2000 21:28:15 +0000 (21:28 +0000)]
We already have a perfectly good next_token() function we should be using
instead of strtok - this fixes a bug with NT users with spaces in their
names when using winbindd. Needs to be added to the other parse_lpXX functions
(currently only added to lprng parsing code).
(This used to be commit c3e4ac9a2db32c40ce330de0eab4bc82ef4fd579)

20 years agoFix for uidtoname not returning DOMAIN\name but just name. This caused
Jeremy Allison [Wed, 25 Oct 2000 20:02:25 +0000 (20:02 +0000)]
Fix for uidtoname not returning DOMAIN\name but just name. This caused
printer_access_check to break in a domain environment.
(This used to be commit 0fc1a461504f87c145f5f91189bd767989c488f2)

20 years agoJohn Reilly @ HP (who is a wonderful human being and *definately* needs
Jeremy Allison [Wed, 25 Oct 2000 17:56:38 +0000 (17:56 +0000)]
John Reilly @ HP (who is a wonderful human being and *definately* needs
CVS commit access :-) has written a simple routine that peeks inside the
MS PE printer driver file format and can tell if a driver is W2K or NT4.x.
So we can now correctly return the driver version number. Hurrah !
JF - this is the code you always wanted ..... :-) :-).
(This used to be commit fd17374e6d888813f4ed7142480cf93b8a16bfef)

20 years agoadded some clarifications.
Gerald Carter [Wed, 25 Oct 2000 17:39:42 +0000 (17:39 +0000)]
added some clarifications.

(This used to be commit 4f76f17a2c10e7b4a5d3cd6fa11930a97878763e)

20 years agoRemoved erroneous +1's on length of unicode domain names for various
Tim Potter [Wed, 25 Oct 2000 04:29:35 +0000 (04:29 +0000)]
Removed erroneous +1's on length of unicode domain names for various
functions in the \\lsarpc pipe.
(This used to be commit 30bb3f2df97662c6547c7994ce50efe9abf37968)

20 years agoYeah - not setting getwd_cache to True as default, that'd do it for a 5x
Jeremy Allison [Tue, 24 Oct 2000 19:44:02 +0000 (19:44 +0000)]
Yeah - not setting getwd_cache to True as default, that'd do it for a 5x
performance hit..... :-) :-).
PS. Yes this was my fault for everyone who wants to throw rotten fruit...
(This used to be commit 4f5bbc38ce492baa6e2b4d94758e9247756ae9ba)

20 years agoSmall change to process of joining a domain with samedit.
Tim Potter [Sun, 22 Oct 2000 22:38:33 +0000 (22:38 +0000)]
Small change to process of joining a domain with samedit.
(This used to be commit 7c3518c55eeff0a962c367438cb5d67044698825)

20 years agoFixes from David Lee for utmp handling.
Jeremy Allison [Sat, 21 Oct 2000 01:47:25 +0000 (01:47 +0000)]
Fixes from David Lee for utmp handling.
(This used to be commit 70cecfefc8e876871b29ff8d5fa9226db0e98153)

20 years agoFix for vfs_fstat() crash with new stat code.
Jeremy Allison [Fri, 20 Oct 2000 23:59:47 +0000 (23:59 +0000)]
Fix for vfs_fstat() crash with new stat code.
(This used to be commit 979aa01e011d433328af0ead7644f47d0cfadf13)

20 years agoMerge from appliance-head
Tim Potter [Fri, 20 Oct 2000 03:23:36 +0000 (03:23 +0000)]
Merge from appliance-head
(This used to be commit 6d39df7cf84d391bb4dd55d9a26f9f5a6368f46f)

20 years agoFixed missing bracket on debug statement.
Tim Potter [Thu, 19 Oct 2000 21:49:48 +0000 (21:49 +0000)]
Fixed missing bracket on debug statement.
(This used to be commit 9328bc7b8a7d43ef0da47413f314f9943ea98847)

20 years agoOk - this is a big patch - and it may break smbd a bit (although
Jeremy Allison [Thu, 19 Oct 2000 02:58:24 +0000 (02:58 +0000)]
Ok - this is a big patch - and it may break smbd a bit (although
I hope not). If you encounter strange file-serving behavior after this
patch then back it out. I analysed our stat() usage and realised we
were doing approx. 3 stat calls per open, and 2 per getattr/setattr.
This patch should fix all that. It causes the stat struct returned
from unix_convert() (which now *must* be passed a valid SMB_STRUCT_STAT
pointer) to be passed through into the open code. This should prevent
the multiple stats that were being done so as not to violate layer
encapsulation in the API's.

Herb - if you could run a NetBench test with this code and do a
padc/par syscall test and also run with the current 2.2.0 code
and test the padc/par syscalls I'd appreciate it - you should
find the number of stat calls reduced - not sure by how much.

The patch depends on unix_convert() actually finding the file
and returning a stat struct, or returning a zero'd out stat
struct if the file didn't exist. I believe we can guarentee this
to be the case - I just wasn't confident enough to make this
an assertion before.

Ok ok - I did write this whilst at the Miami conference.....
sometimes you get a little free time at these things :-).

(This used to be commit 66a5c05ec46b641224fbe01b30bd7e83571a2a1b)

20 years agoRemoved null user checks to enable lanman print queue pausing and print job
Tim Potter [Wed, 18 Oct 2000 06:31:14 +0000 (06:31 +0000)]
Removed null user checks to enable lanman print queue pausing and print job
(This used to be commit e40a0e2f52326199264d622c4f11877f20e4ebce)

20 years agoRenamed WPrintQueuePurge to WPrintQueueCtrl as this function can purge, pause
Tim Potter [Wed, 18 Oct 2000 06:22:20 +0000 (06:22 +0000)]
Renamed WPrintQueuePurge to WPrintQueueCtrl as this function can purge, pause
or resume a print queue.
(This used to be commit 9998518a69c67c1540266549c69fca977a89acde)

20 years agoBug report that on some systems extended characters are being returned as
Christopher R. Hertel [Wed, 18 Oct 2000 01:36:26 +0000 (01:36 +0000)]
Bug report that on some systems extended characters are being returned as
negative values from the mygetc() function.  I've modified the return
line so that it should return values in the 0..255 range for legitimate

This change should probably be copied into SAMBA_2_2 but I haven't checked
that tree out yet.

Chris -)-----
(This used to be commit e2ce5ce0fdaca0e38d953baa2da4c3542b0503ee)

20 years agoMerge of rpc parse buffer underflow debug code.
Tim Potter [Wed, 18 Oct 2000 01:15:05 +0000 (01:15 +0000)]
Merge of rpc parse buffer underflow debug code.
(This used to be commit c8d88713d9f7a646eb3b8e76bdd0250a3b89b722)

20 years agoFixes for io_q_setprinter rpc parsing.
Tim Potter [Wed, 18 Oct 2000 00:55:02 +0000 (00:55 +0000)]
Fixes for io_q_setprinter rpc parsing.
(This used to be commit 1b86b9139e87cc4c4a6675ceeda307b91eaa1482)

20 years agoAdded documentation for winbind enum users and winbind enum groups
Tim Potter [Tue, 17 Oct 2000 02:28:24 +0000 (02:28 +0000)]
Added documentation for winbind enum users and winbind enum groups
configuration parameters.
(This used to be commit c76dfab559003c9194f519cb89869d191cc0d6c0)

20 years agoget rid of $Revision string - sync with 2.2 branch
Herb Lewis [Mon, 16 Oct 2000 23:48:37 +0000 (23:48 +0000)]
get rid of $Revision string - sync with 2.2 branch
(This used to be commit 192b088a4649c2b6b27bd30ae47d1228f9c86092)