Andrew Tridgell [Tue, 10 Jul 2007 03:04:16 +0000 (03:04 +0000)]
r23796: main COPYING file for samba4, plus some formatting varients
Andrew Tridgell [Tue, 10 Jul 2007 02:46:15 +0000 (02:46 +0000)]
r23795: more v2->v3 conversion
Andrew Tridgell [Tue, 10 Jul 2007 02:31:50 +0000 (02:31 +0000)]
r23794: convert more code from LGPLv2+ to LGPLv3+
Andrew Tridgell [Tue, 10 Jul 2007 02:07:03 +0000 (02:07 +0000)]
r23792: convert Samba4 to GPLv3
There are still a few tidyups of old FSF addresses to come (in both s3
and s4). More commits soon.
Andrew Bartlett [Mon, 9 Jul 2007 12:31:35 +0000 (12:31 +0000)]
r23762: Fix DN renames over LDAP, and instrument the partition module. Add a
test to prove the behaviour of LDAP renames etc.
Fix LDB to return correct error code when failing to rename one DN
onto another.
Andrew Bartlett
Andrew Bartlett [Mon, 9 Jul 2007 03:08:20 +0000 (03:08 +0000)]
r23754: Make sure to check the status return before we de-reference the
returned elements.
Andrew Bartlett
Tim Potter [Mon, 9 Jul 2007 01:49:49 +0000 (01:49 +0000)]
r23753: Fix bitrot afflicting the ldb Python swig bindings.
Andrew Bartlett [Sat, 7 Jul 2007 04:34:36 +0000 (04:34 +0000)]
r23737: Validate that we object to duplicate values in an add or replace.
We can't ever allow duplicates, even if the client sends them
Andrew Bartlett
Jeremy Allison [Fri, 6 Jul 2007 19:45:01 +0000 (19:45 +0000)]
r23734: Add a test to prove tridge's theory that the mask, once set
on an open directory handle, is never changed. (He's right btw :-).
Jeremy.
Ronnie Sahlberg [Thu, 5 Jul 2007 07:19:14 +0000 (07:19 +0000)]
r23721: For unions, The TAG and the ARM are aligned independently.
Move emitting ALIGN_TO_x_BYTES to after the tag has been dissected so the
alignment only affects the ARM.
Andrew Bartlett [Thu, 5 Jul 2007 06:15:40 +0000 (06:15 +0000)]
r23720: Allow the member server to work against an LDAP Backend. Another case
where LDB isn't as strict as OpenLDAP, the self join record contains
duplicate servicePrincipalNames once the DNS name and domain name are
made equal. (Easier to just skip the useless self-join).
Andrew Bartlett
Andrew Bartlett [Thu, 5 Jul 2007 03:30:46 +0000 (03:30 +0000)]
r23719: ejs being case sensitive, while LDAP is not is a real pain when
dereferencing attributes.
Fix the case to match between the attributes searched for and the ejs
element. (Fixes LDAP-backend selftest)
Andrew Bartlett
Andrew Bartlett [Thu, 5 Jul 2007 03:06:59 +0000 (03:06 +0000)]
r23718: Make Samba4 work against the LDAP backend again.
When we set up the schema, we don't have a partitions container yet.
The LDAP error differs from that given by LDB, so I think we still
have some conformance work to do.
Andrew Bartlett
Andrew Bartlett [Thu, 5 Jul 2007 02:52:58 +0000 (02:52 +0000)]
r23717: We need to remove the _ in LDAP_MANAGERPASS for the
--ldap-manager-pass= option to work.
Andrew Bartlett
Andrew Bartlett [Thu, 5 Jul 2007 01:45:37 +0000 (01:45 +0000)]
r23716: Clarify LDAP Manager DN and fix slapd startup syntax.
Andrew Bartlett
Andrew Bartlett [Thu, 5 Jul 2007 00:34:11 +0000 (00:34 +0000)]
r23715: Make the provision-backend script print out the exact commands to run,
to set up the LDAP backend.
Andrew Bartlett
Andrew Bartlett [Wed, 4 Jul 2007 11:06:32 +0000 (11:06 +0000)]
r23703: Start to get Samba4 to again work with LDAP backends, after I turned
on metze's schema work.
Andrew Bartlett
Andrew Tridgell [Wed, 4 Jul 2007 07:45:41 +0000 (07:45 +0000)]
r23702: fixed a "falling back to select" error in the standard process
model. It's much cleaner to get the event system to close the fd,
especially with the complexity of forks
Andrew Tridgell [Wed, 4 Jul 2007 07:43:26 +0000 (07:43 +0000)]
r23701: when we create a new socket with socket_accept(), clear any flags that
were set in the old one. Otherwise SOCKET_FLAG_NOCLOSE causes a major
fd leak
Andrew Tridgell [Wed, 4 Jul 2007 05:16:19 +0000 (05:16 +0000)]
r23700: pre-open the sam in the parent smbd. This has the effect of loading
the schema. That stops us loading the schema for each new connection.
In future I would prefer to share a lot more of our ldb contexts with
children. That will require a larger piece of surgery.
Andrew Tridgell [Wed, 4 Jul 2007 05:15:06 +0000 (05:15 +0000)]
r23699: when we create the global schema, remember to set the ldb opaque so we
can find it again
Andrew Tridgell [Wed, 4 Jul 2007 04:18:44 +0000 (04:18 +0000)]
r23698: fixed notify:backend so it actually works again
Andrew Tridgell [Wed, 4 Jul 2007 04:16:16 +0000 (04:16 +0000)]
r23697: use the file perm options in the posix backend
Andrew Tridgell [Wed, 4 Jul 2007 04:15:07 +0000 (04:15 +0000)]
r23696: added the create mask and related share permissions options to Samba4,
using the new share_int_option() code from Simo
speaking of which, this is the first time I've looked closely at the
share_classic.c code. It is absolutely and completely braindead and
broken. Whatever drugs Simo was on at the time, he better not try to
cross a border with them on him!
Problems with it:
- if you actually set a value, it gets ignored, and the defvalue gets
used instead ('ret' is never returned). If you don't set a value,
then defvalue gets returned too. Sound useful?
- it means we now have to list parameters in source/param/ in lots
and lots of places, all of which have to match exactly. code like
this is supposed to reduce the likelyhood of errors, not increase
it!
- code which has a long line of if() statements with strcmp() should
cause your fingers to burn on the keyboard when you type it
in. That's what structure lists are for. Strangely enough, we have
all the info in loadparm.c in a structure list, but instead it gets
replicated in share_classic.c in this strange if() strcmp() form
expect some changes to this code shortly. I'll need a calming cup of
tea first though :-)
Andrew Bartlett [Wed, 4 Jul 2007 03:25:44 +0000 (03:25 +0000)]
r23695: By not using the NULL context for these large structures, we don't
falsely report them as leaked.
Andrew Bartlett
Andrew Bartlett [Wed, 4 Jul 2007 00:34:16 +0000 (00:34 +0000)]
r23693: Give the process a chance to write out it's coverage data, before we
blast a kill -9 at it.
Andrew Bartlett
Stefan Metzmacher [Tue, 3 Jul 2007 16:27:35 +0000 (16:27 +0000)]
r23687: not every compiler supports __typeof__()
metze
Stefan Metzmacher [Tue, 3 Jul 2007 14:44:27 +0000 (14:44 +0000)]
r23684: Only use the bzr and git information when there's
a .bzr or .git in the top branch directory
I use bzr for my home directory and
bzr version-info in ~/devel/samba/4.0/samba4-git
gives the info about the ~/.bzr branch
metze
Stefan Metzmacher [Tue, 3 Jul 2007 08:22:24 +0000 (08:22 +0000)]
r23681: merge 23670 from SAMBA_3_0:
metze
Andrew Bartlett [Tue, 3 Jul 2007 08:05:55 +0000 (08:05 +0000)]
r23680: Make it easier to setup a domain member server - the 'server role'
will now control the auth methods, but an override is still available,
ex:
auth methods:domain controller = <methods>
Andrew Bartlett
Andrew Bartlett [Tue, 3 Jul 2007 08:01:34 +0000 (08:01 +0000)]
r23679: invocationID is a GUID too.
Andrew Bartlett
Andrew Bartlett [Tue, 3 Jul 2007 08:00:08 +0000 (08:00 +0000)]
r23678: Update to current lorikeet-heimdal (-r 767), which should fix the
panics on hosts without /dev/random.
Andrew Bartlett
Andrew Bartlett [Tue, 3 Jul 2007 07:28:46 +0000 (07:28 +0000)]
r23677: When I removed data_blob_equal, I clearly didn't test the PIDL code.
Fix these to use the new data_blob_cmp()
Andrew Bartlett
Kai Blin [Tue, 3 Jul 2007 06:20:00 +0000 (06:20 +0000)]
r23676: Added a .gitignore file to hide generated files from git status.
It's identical to .bzrignore
Kai Blin [Tue, 3 Jul 2007 06:17:39 +0000 (06:17 +0000)]
r23675: Build libnss_winbind on linux.
Kai Blin [Tue, 3 Jul 2007 06:16:44 +0000 (06:16 +0000)]
r23674: Merge changes from Samba 3 SVN r23539:
Avoid code duplication in winbind_nss_linux.h
Stefan Metzmacher [Fri, 29 Jun 2007 11:07:19 +0000 (11:07 +0000)]
r23653: add dummy implementation for all nss calls to let nsstest
pass against a samba4 winbind
metze
Jim McDonough [Thu, 28 Jun 2007 18:08:04 +0000 (18:08 +0000)]
r23643: Fix the build farm tests. We were incorrectly passing the
RPC-SAMBA3-GETUSERNAME tests before the previous password expiration
fixes, because if you create a user and only set the password
administratrively, the "last set time" should not get updated. Needed
to add some more of the fields_present flags to do this.
Stefan Metzmacher [Wed, 27 Jun 2007 20:32:35 +0000 (20:32 +0000)]
r23638: add SMB_LIBRARY() macro to create a LIBRARY::foo section in
source/config.mk from within *.m4 files
metze
Stefan Metzmacher [Wed, 27 Jun 2007 19:46:53 +0000 (19:46 +0000)]
r23637: fix detection of visibility attribute support
James: AC_TRY_LINK() has a different syntax as AC_TRY_RUN() :-)
so this was wrong:
http://websvn.samba.org/cgi-bin/viewcvs.cgi?rev=20697&view=rev
metze
Stefan Metzmacher [Wed, 27 Jun 2007 17:48:30 +0000 (17:48 +0000)]
r23635: fix the samba4 build...
we need a bit more work to get linking of
shared libraries completely right...
metze
Stefan Metzmacher [Wed, 27 Jun 2007 15:37:03 +0000 (15:37 +0000)]
r23631: when linking shared libraries we need to make shure the linker doesn't
skip "unneeded" object files from within .a files.
If the build-farm doesn't like -Wl,--whole-archive, I'll add some more
configure logic fix it tomorrow.
Kai: with this your libnss_winbind.so patch works for me
metze
Günther Deschner [Wed, 27 Jun 2007 12:35:12 +0000 (12:35 +0000)]
r23630: Found out what LSA_LOOKUP_NAMES level 5 means:
only query transitive forest trusts.
Guenther
Günther Deschner [Wed, 27 Jun 2007 12:30:21 +0000 (12:30 +0000)]
r23629: Document lsa lookupname lookuplevels Volker discovered a while ago.
Guenther
Kai Blin [Tue, 26 Jun 2007 06:04:14 +0000 (06:04 +0000)]
r23605: Update mkversion.sh to support git and git-svn checkouts.
Jerry, did I get you right that you'll merge this to 3_0 yourself?
Kai
Jeremy Allison [Fri, 22 Jun 2007 17:36:10 +0000 (17:36 +0000)]
r23590: Fix realloc leak on failure case from Jim Meyering <jim@meyering.net>.
Jeremy.
Michael Adam [Thu, 21 Jun 2007 11:14:12 +0000 (11:14 +0000)]
r23561: Fix an output message.
Andrew Bartlett [Thu, 21 Jun 2007 10:18:20 +0000 (10:18 +0000)]
r23560: - Activate metze's schema modules (from metze's schema-loading-13 patch).
- samba3sam.js: rework the samba3sam test to not use objectCategory,
as it's has special rules (dnsName a simple match)
- ldap.js: Test the ordering of the objectClass attributes for the baseDN
- schema_init.c: Load the mayContain and mustContain (and system...) attributes when
reading the schema from ldb
- To make the schema load not suck in terms of performance, write the
schema into a static global variable
- ldif_handlers.c: Match objectCategory for equality and canonicolisation
based on the loaded schema, not simple tring manipuation
- ldb_msg.c: don't duplicate attributes when adding attributes to a list
- kludge_acl.c: return allowedAttributesEffective based on schema results
and privilages
Andrew Bartlett
Günther Deschner [Thu, 21 Jun 2007 08:52:15 +0000 (08:52 +0000)]
r23559: After running testprogs/ejs/minschema.js update schema.ldif (on abartlet's
request).
Guenther
Andrew Bartlett [Thu, 21 Jun 2007 05:14:13 +0000 (05:14 +0000)]
r23558: MMC seems to ask for this, so I think we need to include it in our schema.
Andrew Bartlett
Andrew Bartlett [Thu, 21 Jun 2007 01:28:33 +0000 (01:28 +0000)]
r23557: Ensure that we don't reorder the objectClass list, if we don't have
any subclasses loaded yet, or none are applicable.
This fixes MMC so that it at least displays the Samba domain as a
domain, but there is still work to be done.
Andrew Bartlett
Andrew Bartlett [Wed, 20 Jun 2007 04:15:39 +0000 (04:15 +0000)]
r23551: Change data_blob_equal to data_blob_cmp, suitable for sorting with qsort().
Andrew Bartlett
Jelmer Vernooij [Mon, 18 Jun 2007 21:48:57 +0000 (21:48 +0000)]
r23542: Include function documentation in prototype headers.
Alexander Bokovoy [Mon, 18 Jun 2007 14:06:04 +0000 (14:06 +0000)]
r23540: Continue fixing bench-readwrite. We need to initialize state->readcnt not only when file is written but at initial write as well.
Andrew Tridgell [Sun, 17 Jun 2007 21:54:11 +0000 (21:54 +0000)]
r23535: accept numeric ntstatus codes in nbench load files (ronnies wireshark
script generates numeric codes)
Andrew Tridgell [Sun, 17 Jun 2007 20:04:57 +0000 (20:04 +0000)]
r23534: honour the targetrate even for loadfiles with timestamps
Andrew Tridgell [Sun, 17 Jun 2007 20:03:31 +0000 (20:03 +0000)]
r23533: added --option torture:targetrate=rate to target a particular
throughput in MBytes/sec
Andrew Tridgell [Sun, 17 Jun 2007 20:02:56 +0000 (20:02 +0000)]
r23532: added lp_parm_double()
Andrew Tridgell [Sun, 17 Jun 2007 19:35:00 +0000 (19:35 +0000)]
r23531: added optional latency reporting in BENCH-NBENCH. To use it, you need
to have a load file that puts a timestamp in the first column. That
tells NBENCH to target the same throughput as the capture, and to
report the maximum amount of latency miss in the test
This allows you to quickly see how many clients you can run with while
still meeting a given target
Alexander Bokovoy [Sun, 17 Jun 2007 16:46:44 +0000 (16:46 +0000)]
r23529: When using READX, fill in readx fields. Thanks to Metze for spotting it.
Andrew Tridgell [Sat, 16 Jun 2007 17:54:36 +0000 (17:54 +0000)]
r23521: NT_STATUS_OBJECT_NAME_NOT_FOUND is not an error either
Andrew Tridgell [Sat, 16 Jun 2007 17:14:02 +0000 (17:14 +0000)]
r23520: display shadow copy information in 'allinfo'
Andrew Tridgell [Sat, 16 Jun 2007 17:13:42 +0000 (17:13 +0000)]
r23519: added libcli code for fetching shadow copy information
Alexander Bokovoy [Fri, 15 Jun 2007 12:23:14 +0000 (12:23 +0000)]
r23506: Turn back to RAW_READ_READ because buildfarm fails. Investigating.
Alexander Bokovoy [Fri, 15 Jun 2007 11:16:19 +0000 (11:16 +0000)]
r23505: Use RAW_READ_READX, not RAW_READ_READ in BENCH-READWRITE test (we already write with RAW_WRITE_WRITEX). Noticed by Gomati Mohanan
Andrew Bartlett [Fri, 15 Jun 2007 00:14:11 +0000 (00:14 +0000)]
r23503: use hdb_dbc not hdb_openp.
Andrew Bartlett
Stefan Metzmacher [Thu, 14 Jun 2007 13:02:55 +0000 (13:02 +0000)]
r23493: regenerate lex.c files with flex 2.5.33
this makes sure we include config.h as first header
hopefully fixes the build on SerNet-aix
metze
Volker Lendecke [Thu, 14 Jun 2007 13:02:31 +0000 (13:02 +0000)]
r23492: Also activate the split out test...
Volker Lendecke [Thu, 14 Jun 2007 13:01:58 +0000 (13:01 +0000)]
r23491: Split out the RENAME trans2 call into a separate test
Stefan Metzmacher [Thu, 14 Jun 2007 12:33:40 +0000 (12:33 +0000)]
r23490: try to fix the build without automatic dependencies
metze
Stefan Metzmacher [Thu, 14 Jun 2007 12:19:53 +0000 (12:19 +0000)]
r23488: hdb_openp has changed from void * to int...
lha: what is the reason for this? it's really bad to use
an int for storing a pointer value...
metze
Stefan Metzmacher [Thu, 14 Jun 2007 12:05:08 +0000 (12:05 +0000)]
r23487: fix the build with automatic dependencies
metze
Kai Blin [Thu, 14 Jun 2007 06:05:27 +0000 (06:05 +0000)]
r23481: As per metze's suggestion, move the nsstest files from torture/ to nsswitch/
Andrew Bartlett [Wed, 13 Jun 2007 05:44:24 +0000 (05:44 +0000)]
r23456: Update Samba4 to current lorikeet-heimdal.
Andrew Bartlett
Andrew Bartlett [Wed, 13 Jun 2007 05:14:00 +0000 (05:14 +0000)]
r23455: These buffers may not be null terminated. Ensure we don't run past the
end of teh buffer printing the error strings.
Andrew Bartlett
Stefan Metzmacher [Mon, 11 Jun 2007 14:41:14 +0000 (14:41 +0000)]
r23420: try to make the IBM checker happy
metze
Stefan Metzmacher [Mon, 11 Jun 2007 14:40:23 +0000 (14:40 +0000)]
r23419: initialize variable...
metze
Stefan Metzmacher [Mon, 11 Jun 2007 13:02:30 +0000 (13:02 +0000)]
r23418: 'S' is 0x53...
metze
Stefan Metzmacher [Mon, 11 Jun 2007 09:50:18 +0000 (09:50 +0000)]
r23413: TP5 is releast...
metze
Andrew Bartlett [Mon, 11 Jun 2007 03:45:56 +0000 (03:45 +0000)]
r23412: We don't need hdb.h here any more
Kai Blin [Sat, 9 Jun 2007 07:17:24 +0000 (07:17 +0000)]
r23403: Make nsstest build on Samba4.
For now, only build on Linux systems. If the build farm is happy with this,
I will gradually turn on this feature for other platforms, too.
Günther Deschner [Fri, 8 Jun 2007 10:56:46 +0000 (10:56 +0000)]
r23386: Correctly fill in server name in DsrGetDcSiteCoverageW() test.
Guenther
Günther Deschner [Fri, 8 Jun 2007 10:53:47 +0000 (10:53 +0000)]
r23385: Adding netr_DsRGetForestTrustInformation() test to query transitive forest
trusts as well as our primary domain.
Guenther
Günther Deschner [Fri, 8 Jun 2007 10:49:46 +0000 (10:49 +0000)]
r23384: Fill in NETLOGON netr_DsRGetForestTrustInformation().
Guenther
Günther Deschner [Fri, 8 Jun 2007 10:43:42 +0000 (10:43 +0000)]
r23383: Query transitive forest trust info in LSA torture test once we see such trusts.
Guenther
Günther Deschner [Fri, 8 Jun 2007 10:42:33 +0000 (10:42 +0000)]
r23382: Fill in lsa_lsaRQueryForestTrustInformation.
Guenther
Günther Deschner [Fri, 8 Jun 2007 10:32:29 +0000 (10:32 +0000)]
r23381: Merge netr_GetDcName WERROR return and WERROR_DOMAIN_CONTROLLER_NOT_FOUND from
SAMBA_3_0.
Guenther
Andrew Bartlett [Thu, 7 Jun 2007 05:56:18 +0000 (05:56 +0000)]
r23373: Fix spelling...
Jeremy Allison [Thu, 7 Jun 2007 00:14:06 +0000 (00:14 +0000)]
r23371: Fix the misleading comment I added - it really *should*
say "locks chain and returned record", not
"and returns record"
Jeremy.
Jeremy Allison [Thu, 7 Jun 2007 00:00:45 +0000 (00:00 +0000)]
r23370: Traverse in tdb wasn't consistently using the
travlocks.lock_rw for lock read/write types, it
was sometimes using it (tdb_next_lock) and
sometimes explicitly using F_WRLCK instead.
Change this to consistently use travlocks.lock_rw
only.
I'm pretty sure about this fix (else I woudn't
be checking this in :-) but tridge and Volker
please review.
Jeremy.
Andrew Bartlett [Wed, 6 Jun 2007 12:51:45 +0000 (12:51 +0000)]
r23365: Try to make Windows Vista join again. On my new test environment, it
wants to check for an existing domain join account, and fails. This
test shows that we need to return NT_STATUS_NONE_MAPPED when nothing
matches. (not yet tested if this helps vista).
Andrew Bartlett
Andrew Tridgell [Wed, 6 Jun 2007 12:44:04 +0000 (12:44 +0000)]
r23364: add LDB_FLG_NOMMAP flag
Michael Adam [Wed, 6 Jun 2007 09:27:43 +0000 (09:27 +0000)]
r23363: Activate rep_pread and rep_pwrite in lib/replace/replace.h.
This fixes the build on a SuSE 6.1. :-)
I guess this had been merely forgotten.
But beware: The implementations of rep_pread and rep_pwrite are
not thread safe.
Michael
Andrew Bartlett [Wed, 6 Jun 2007 01:36:15 +0000 (01:36 +0000)]
r23361: Merge from SAMBA_4_0_RELEASE:
Print the smb.conf path being created in provision.
Andrew Bartlett
Andrew Bartlett [Tue, 5 Jun 2007 05:35:39 +0000 (05:35 +0000)]
r23351: Merge from SAMBA_4_0_RELEASE:
Fix a nasty issue we had with SWAT. We could not provision into a
different domain, as we didn't re-calcuate the DOMAINDN after the user
changed it in the form.
Andrew Bartlett
Stefan Metzmacher [Mon, 4 Jun 2007 14:26:37 +0000 (14:26 +0000)]
r23339: merge from SAMBA_3_0:
fix a crash bug...I wonder why only HP-UX 11.00 ans 11.11 noticed
it via a SIGBUS...
I missed to remove the samba3 specifc code path to tdb_open_ex()
when I synced lib/tdb/ with samba4. The explicit cast in on tdb_open_ex()
dropped the compiler warning :-(
metze
Andrew Bartlett [Mon, 4 Jun 2007 03:55:36 +0000 (03:55 +0000)]
r23325: Remove items from menu that have been removed from SVN long ago, with
the start of the transition to the new SWAT.
Andrew Bartlett
Kai Blin [Sat, 2 Jun 2007 11:38:27 +0000 (11:38 +0000)]
r23311: Updating the samba4 winbind protocol to version 18.
nsswitch/winbindd_nss.h is just copied from SAMBA_3_0.
nsswitch/winbind_nss_config.h is copied from SAMBA_3_0, too, but I had to
drop some of the defines to make things build again.
Kai
Stefan Metzmacher [Sat, 2 Jun 2007 08:16:42 +0000 (08:16 +0000)]
r23306: pass down LDFLAGS to the link command in the libreplace standalone build
metze
Stefan Metzmacher [Sat, 2 Jun 2007 08:00:45 +0000 (08:00 +0000)]
r23303: so HP-UX 11.11 also likes the 11.00 workarround for broken
pread/pwrite when using large file support.
metze