12 years agor21316: if we got an unexpected nbt packet that most times mean
Stefan Metzmacher [Tue, 13 Feb 2007 15:17:29 +0000 (15:17 +0000)]
r21316: if we got an unexpected nbt packet that most times mean
we got a 2nd answer to a broadcast message and have already remove
the packet id from out list while getting the first response

(This used to be commit 8c26e04900da02bdf440f1d48b512e2550e89c34)

12 years agor21315: ldb now supports filters like (&(dn=%s)(&(objectClass=kerberosSecret)(private...
Stefan Metzmacher [Tue, 13 Feb 2007 13:43:23 +0000 (13:43 +0000)]
r21315: ldb now supports filters like (&(dn=%s)(&(objectClass=kerberosSecret)(privateKeytab=*))) again
we can use such a filter:-)

we should only update the keytab for records matching this filter,
that means we need to do a search before calling cli_credentials_set_secrets()

(This used to be commit 23adca4e3426360fe0685548ae2b808578f6ba75)

12 years agor21314: add more usefull debug output
Stefan Metzmacher [Tue, 13 Feb 2007 13:14:14 +0000 (13:14 +0000)]
r21314: add more usefull debug output

(This used to be commit a246e4bbaaab6f98f50a3c28b47d2c541af7b44a)

12 years agor21311: fix very ugly "using free'ed memory" bug
Stefan Metzmacher [Tue, 13 Feb 2007 12:32:48 +0000 (12:32 +0000)]
r21311: fix very ugly "using free'ed memory" bug

This was there since 2005...

(This used to be commit 393e4eeb82df8017eb0afb526f4d723cf8814311)

12 years agor21307: make it possible to pass in NULL for domain or rid,
Stefan Metzmacher [Tue, 13 Feb 2007 09:27:56 +0000 (09:27 +0000)]
r21307: make it possible to pass in NULL for domain or rid,
if someone isn't interessted in one of it

(This used to be commit 1fdc71918a430c35af91fa7788e191d381f76d56)

12 years agor21306: fix the RPC-LSA tests the admin couldn't no longer get the 'currentValue'
Stefan Metzmacher [Tue, 13 Feb 2007 08:08:23 +0000 (08:08 +0000)]
r21306: fix the RPC-LSA tests the admin couldn't no longer get the 'currentValue'

this needs more works, but make it work again for now

(This used to be commit 608d24f0016ff090b7de7fbd0bed85153bcc703d)

12 years agor21305: Change the skel module a little, so make it names clearer.
Andrew Bartlett [Tue, 13 Feb 2007 03:52:57 +0000 (03:52 +0000)]
r21305: Change the skel module a little, so make it names clearer.

Andrew Bartlett
(This used to be commit 26758fc24ca0b28e1ed07cc20801175e94362d57)

12 years agor21304: - Get rid of the extra set of scroll bars in the tree. The tree never grows
Derrell Lipman [Tue, 13 Feb 2007 03:49:42 +0000 (03:49 +0000)]
r21304: - Get rid of the extra set of scroll bars in the tree.  The tree never grows
  or needs scroll bars; rather, it's internal areas within the tree widget
  that provide the scroll bars.  (The next version of TreeVirtual will prevent
  setting overflow to anything other than "hidden".)
(This used to be commit 2dc07486405967dfa74fc6f4a6586fb9da8e5ed7)

12 years agor21303: As discussed on samba-technical: Change the static array for the in-memory
Volker Lendecke [Mon, 12 Feb 2007 23:16:02 +0000 (23:16 +0000)]
r21303: As discussed on samba-technical: Change the static array for the in-memory
mirrors of the hash chain locks to a dynamically allocated one.

Jeremy, I count on you to revert it if the build farm freaks out, it's after
midnight here :-)

(This used to be commit 7b5db2e472c7e27231fa432d3930789e708abd09)

12 years agor21302: Provide simple redirecting headers for standard headers.
Jelmer Vernooij [Mon, 12 Feb 2007 17:36:00 +0000 (17:36 +0000)]
r21302: Provide simple redirecting headers for standard headers.
(This used to be commit 74c47839536c9ccfa1240289d20df65d9d13839d)

12 years agor21301: - the history contains lm and nt history
Stefan Metzmacher [Mon, 12 Feb 2007 15:25:43 +0000 (15:25 +0000)]
r21301: - the history contains lm and nt history
- autoset the size values

(This used to be commit 5e51845137a05b6c4ca6804feaced3b5251b455f)

12 years agor21300: let the caller decide if it wants rid decrypted hashes or not
Stefan Metzmacher [Mon, 12 Feb 2007 13:34:04 +0000 (13:34 +0000)]
r21300: let the caller decide if it wants rid decrypted hashes or not

(This used to be commit 8711d01ffd080c43512b88b995daf2d6b7c06ba1)

12 years agor21299: Fix the build for those that don't have talloc.h installed.
Jelmer Vernooij [Mon, 12 Feb 2007 13:04:09 +0000 (13:04 +0000)]
r21299: Fix the build for those that don't have talloc.h installed.
(This used to be commit e782035251fd3d51a7a4221d107519fb1ba70ba7)

12 years agor21298: protect windows password attributes too
Stefan Metzmacher [Mon, 12 Feb 2007 12:13:00 +0000 (12:13 +0000)]
r21298: protect windows password attributes too

(This used to be commit 5c779b3767b47c140fc658fb9aed0ebfd5d956f0)

12 years agor21297: Remove the GTK+ tools and library from the main repository. They are now...
Jelmer Vernooij [Mon, 12 Feb 2007 12:12:12 +0000 (12:12 +0000)]
r21297: Remove the GTK+ tools and library from the main repository. They are now maintained separately in bzr at

This also adds some more headers to the list that is installed and a couple of extra #include lines so these
headers can be used externally without problems.
(This used to be commit 07652f65ce7a5b19130f1a27cbf0e1e5fae13454)

12 years agor21296: remove the session specific encryption from the attributes
Stefan Metzmacher [Mon, 12 Feb 2007 11:46:35 +0000 (11:46 +0000)]
r21296: remove the session specific encryption from the attributes
before storing them.

(This used to be commit 7146e265a441eaa46c20361178be371eb6985371)

12 years agor21295: pass down the gensec session key to the store_chunk hook
Stefan Metzmacher [Mon, 12 Feb 2007 11:20:57 +0000 (11:20 +0000)]
r21295: pass down the gensec session key to the store_chunk hook

(This used to be commit 91de9591732aab822f2a44c4a6698edce48f723b)

12 years agor21294: this is useless, now that there's no operation in between...
Stefan Metzmacher [Mon, 12 Feb 2007 11:16:06 +0000 (11:16 +0000)]
r21294: this is useless, now that there's no operation in between...

(This used to be commit 838d307e6ca0740bc330a5ebc46b95f3181a5c14)

12 years agor21293: store the gensec session key for each drsuapi connection
Stefan Metzmacher [Mon, 12 Feb 2007 10:58:06 +0000 (10:58 +0000)]
r21293: store the gensec session key for each drsuapi connection

(This used to be commit d399739820650bcbe486956374af84c436d3ec16)

12 years agor21292: add become_dc:print = yes option to dump out the drsuapi traffic
Stefan Metzmacher [Mon, 12 Feb 2007 09:53:52 +0000 (09:53 +0000)]
r21292: add become_dc:print = yes option to dump out the drsuapi traffic

(This used to be commit 5065a9b5ffe4dc9a77f4e5955dc0d8d86feb8c5e)

12 years agor21291: add two more error codes
Stefan Metzmacher [Mon, 12 Feb 2007 09:36:53 +0000 (09:36 +0000)]
r21291: add two more error codes

(This used to be commit dd04c5dec58b18048ca5029f7bfe513242cbe4e9)

12 years agor21287: Improve user feedback. Cursor now changes to stopwatch during RPC and module...
Derrell Lipman [Sun, 11 Feb 2007 22:59:02 +0000 (22:59 +0000)]
r21287: Improve user feedback.  Cursor now changes to stopwatch during RPC and module loading.
(This used to be commit a6fc47eff3aa87b1dfe4eaf25719b51b24c88425)

12 years agor21286: Tree column now resizes wiith splitter adjustment, but scrollbars don't yet...
Derrell Lipman [Sun, 11 Feb 2007 21:54:14 +0000 (21:54 +0000)]
r21286: Tree column now resizes wiith splitter adjustment, but scrollbars don't yet update properly
(This used to be commit a66b64026a5d80a1e16b2fc9665143512975a53e)

12 years agor21285: - Add the new ResizeTableColumnModel and make use of it in TreeVirtual. ...
Derrell Lipman [Sun, 11 Feb 2007 20:22:45 +0000 (20:22 +0000)]
r21285: - Add the new ResizeTableColumnModel and make use of it in TreeVirtual.  This
  allows the Ldb Browser tree to properly size itself upon initially appearing
  and upon window resizes.  There are still a few problems with it that I need
  to resolve, including an occasional set of double scrollbars, and making it
  resize the tree column when the splitter is resized.
(This used to be commit c3c93ad36a9e850865aa8b09e319a77441243b01)

12 years agor21282: we only need one for loop...
Stefan Metzmacher [Sun, 11 Feb 2007 17:51:38 +0000 (17:51 +0000)]
r21282: we only need one for loop...

(This used to be commit 181b3a031f9683ea5e9aa2c96f121639561c6830)

12 years agor21281: move constinancy checks to the beginning of the function
Stefan Metzmacher [Sun, 11 Feb 2007 17:36:33 +0000 (17:36 +0000)]
r21281: move constinancy checks to the beginning of the function

(This used to be commit f2af44d20484e57495ab0ebd5aab993e4af43fd4)

12 years agor21276: Actually fire off multiple parallel read/write requests.
Volker Lendecke [Sun, 11 Feb 2007 12:03:25 +0000 (12:03 +0000)]
r21276: Actually fire off multiple parallel read/write requests.

(This used to be commit 1d7bcdb33ef19be7bf64356b9cb8eee9658afe09)

12 years agor21275: Reformat the bench-readwrite test a bit
Volker Lendecke [Sun, 11 Feb 2007 10:42:48 +0000 (10:42 +0000)]
r21275: Reformat the bench-readwrite test a bit
(This used to be commit d2d44bf6779c6ac46b011d7e3b372c28c6f8a4f9)

12 years agor21272: in the history fields each hash is rid crypted on its own
Stefan Metzmacher [Sat, 10 Feb 2007 19:44:16 +0000 (19:44 +0000)]
r21272: in the history fields each hash is rid crypted on its own

(This used to be commit 922cfacde4dcf27829040e12c7fae083bd5b44c2)

12 years agor21270: Fix a memleak
Volker Lendecke [Sat, 10 Feb 2007 17:46:33 +0000 (17:46 +0000)]
r21270: Fix a memleak
(This used to be commit 918031a99c583b45c854c6d066f56ca6887011dd)

12 years agor21267: the first 4 bytes in each encrypted attribute is the crc32
Stefan Metzmacher [Sat, 10 Feb 2007 09:46:28 +0000 (09:46 +0000)]
r21267: the first 4 bytes in each encrypted attribute is the crc32
checksum over the remaining bytes

(This used to be commit c9fe6a867c7e23a60e0f9200b04c31e48ccc11a2)

12 years agor21266: Add/Modify/Delete operations seem to work correctly now
Simo Sorce [Fri, 9 Feb 2007 23:43:42 +0000 (23:43 +0000)]
r21266: Add/Modify/Delete operations seem to work correctly now
Still no refresh, so the results may seem confusing
(This used to be commit 0cee51ca2e81c3119a9f428b22909baf0c4535e8)

12 years agor21261: add support for getting to the plain nthash (and I assume others too)
Stefan Metzmacher [Fri, 9 Feb 2007 17:35:32 +0000 (17:35 +0000)]
r21261: add support for getting to the plain nthash (and I assume others too)

I wonder what the first 4 bytes are, but the last 16 bytes of the 20 bytes
are the rid crypted hash. The lmhash and the history fields are not verified
yet. But I get the administrators nthash in plain.

(This used to be commit 7726d4a0272b29a4eade26c9ae3d929df50897da)

12 years agor21260: Fix regex.
Jelmer Vernooij [Fri, 9 Feb 2007 09:44:11 +0000 (09:44 +0000)]
r21260: Fix regex.
(This used to be commit 025597a3593d9dede93180d284beda256caf9483)

12 years agor21256: - msg_type is not needed in the cluster messaging API
Andrew Tridgell [Fri, 9 Feb 2007 01:52:13 +0000 (01:52 +0000)]
r21256: - msg_type is not needed in the cluster messaging API

- merge ctdb_get_num_nodes() from bzr tree
(This used to be commit 3df7527aedeba7ce2f4a6ca2d3b7167f58c6b68a)

12 years agor21255: Add a debugging option to avoid rid decryption in the samsync output.
Andrew Bartlett [Fri, 9 Feb 2007 01:49:26 +0000 (01:49 +0000)]
r21255: Add a debugging option to avoid rid decryption in the samsync output.

Andrew Bartlett
(This used to be commit 9e15a51579157405b2013b9b948d279fefd0eda6)

12 years agor21254: Fix running the testsuite outside pidl/
Jelmer Vernooij [Fri, 9 Feb 2007 00:18:06 +0000 (00:18 +0000)]
r21254: Fix running the testsuite outside pidl/
(This used to be commit 84c15158b986ef9e82f46d6c01d5cbf04300cda5)

12 years agor21253: Merge some pidl fixes:
Jelmer Vernooij [Thu, 8 Feb 2007 23:54:31 +0000 (23:54 +0000)]
r21253: Merge some pidl fixes:

 * Add tests for wireshark dissector generator
 * Add tests for the header code
 * Some cleanups
 * Fix handling of elements without [in] or [out]
(This used to be commit 1aecba7100685ed291ea13b0ae47fb0cf9e6a6c8)

12 years agor21252: Add operation works now, still to do mod and del ops
Simo Sorce [Thu, 8 Feb 2007 23:19:31 +0000 (23:19 +0000)]
r21252: Add operation works now, still to do mod and del ops
Need to implement tree refresh as well
(This used to be commit ed7e00e2a0ab2d2254959f53f7a4f661f9a4c1b3)

12 years agor21249: Now the ldif is generated correctly, but still rpc is not happy
Simo Sorce [Thu, 8 Feb 2007 21:26:44 +0000 (21:26 +0000)]
r21249: Now the ldif is generated correctly, but still rpc is not happy
(This used to be commit 3367f2fe69b8d2c3abad8718991378986d2d85dd)

12 years agor21248: - Test for Simo: no change of selected row(s) upon open/close click. Simo,
Derrell Lipman [Thu, 8 Feb 2007 21:02:42 +0000 (21:02 +0000)]
r21248: - Test for Simo: no change of selected row(s) upon open/close click.  Simo,
  I haven't done any testing of this.  Please let me know the side effects, if
  any, of this change.
(This used to be commit c3c3030a7d9a9f6a67c09f1b260ea0589638c0a5)

12 years agor21245: combination "c[3] s[1] e[1] d[0]..." was successful!!!!!!!!!!!!!!!!!!!!!...
Stefan Metzmacher [Thu, 8 Feb 2007 19:40:54 +0000 (19:40 +0000)]
r21245: combination "c[3] s[1] e[1] d[0]..." was successful!!!!!!!!!!!!!!!!!!!!!!!!!!

we now support the session depended password attribute encryption
used in DsGetNCChanges()

the static encryption (something like sam_rid_crypt() is assumed for some attributes
and the format of the attribute values isn't known yet, but some contain
some unicode strings...

(This used to be commit daee739ebba38e54fbdbd7e53d16fd30bc0dd54e)

12 years agor21244: fix typo...
Stefan Metzmacher [Thu, 8 Feb 2007 18:24:48 +0000 (18:24 +0000)]
r21244: fix typo...

I was wondering why I got allways 4 times the same decrypted buffer...

(This used to be commit ba168479d5c5710de4252d9b8970ab41b7979411)

12 years agor21243: create the infrastructure to test a lot of crypto combinations
Stefan Metzmacher [Thu, 8 Feb 2007 18:23:32 +0000 (18:23 +0000)]
r21243: create the infrastructure to test a lot of crypto combinations
to solve the drsuapi password encryption automaticly.

(This used to be commit 7b331b95bf43384e571ecc88ac827f46acd8c20c)

12 years agor21242: prepare trying to decrypt the encrypted attributes
Stefan Metzmacher [Thu, 8 Feb 2007 18:15:44 +0000 (18:15 +0000)]
r21242: prepare trying to decrypt the encrypted attributes

(This used to be commit 65a55c8fdecbc212ba2fc2a01c580f30dcc7b54a)

12 years agor21236: store and orint the dcerpc session key, gensec session key and user nthash
Stefan Metzmacher [Thu, 8 Feb 2007 10:44:54 +0000 (10:44 +0000)]
r21236: store and orint the dcerpc session key, gensec session key and user nthash
values for each drsuapi connection in RPC-DSSYNC

(This used to be commit df74c4a88108bd4387d9be4f64619c46c23493e6)

12 years agor21235: fix crash bug
Stefan Metzmacher [Thu, 8 Feb 2007 09:10:59 +0000 (09:10 +0000)]
r21235: fix crash bug

(This used to be commit 29c44957a0ea16af3e5e3717d2cb0c9180d394c2)

12 years agor21234: fixed a subtle bug with talloc reference counting and async ntvfs
Andrew Tridgell [Thu, 8 Feb 2007 03:01:47 +0000 (03:01 +0000)]
r21234: fixed a subtle bug with talloc reference counting and async ntvfs
(This used to be commit 6a2ba01f155266add71319c1dbc3442eb0c2c751)

12 years agor21233: first version of samba4 messaging using ctdb is working. This means we
Andrew Tridgell [Thu, 8 Feb 2007 02:59:58 +0000 (02:59 +0000)]
r21233: first version of samba4 messaging using ctdb is working. This means we
should now work on a real cluster, and not just a localhost simulator
(This used to be commit f05072ad74fb08fd906bc500c5e89930bcc3387f)

12 years agor21232: added a raw ctdb messaging api - allowing ctdb applications to take
Andrew Tridgell [Thu, 8 Feb 2007 02:57:08 +0000 (02:57 +0000)]
r21232: added a raw ctdb messaging api - allowing ctdb applications to take
advantage of the ctdb messaging layer for their own data
(This used to be commit b288ba05e5dc2aa5c8cd26eaee1c41b12e4996da)

12 years agor21230: added the hooks needed in the cluster layer and the messaging code for
Andrew Tridgell [Thu, 8 Feb 2007 00:58:17 +0000 (00:58 +0000)]
r21230: added the hooks needed in the cluster layer and the messaging code for
handling messages to remote nodes. Implemented dummy functions in the
'local' cluster backend for the messaging hooks, and modified the
messaging layer to check if the destination is remote and redirect
messages via the cluster layer
(This used to be commit 4474552e8fb73efebef32ad8480d7fe9a1e379ef)

12 years agor21222: Merge a couple of pidl fixes:
Jelmer Vernooij [Wed, 7 Feb 2007 19:03:19 +0000 (19:03 +0000)]
r21222: Merge a couple of pidl fixes:

* Pidl will now warn when trying to use pointers as integers in expressions.
* "subcontext()" is now marked as deprecated. The alternatives,
  transmit_as() / represent_as() should be available soon.
* More tests.
* Remove some unused code in smbtorture.
(This used to be commit 37c0da541e3962164d5af3e3c9560803a733f3b7)

12 years agor21221: 'make build' should be the default target for us
Derrell Lipman [Wed, 7 Feb 2007 18:52:11 +0000 (18:52 +0000)]
r21221: 'make build' should be the default target for us
(This used to be commit ca3b806070b924a455954cdfe124d726b38da87a)

12 years agor21220: enable 'make source' (and 'make') to work properly
Derrell Lipman [Wed, 7 Feb 2007 18:31:28 +0000 (18:31 +0000)]
r21220: enable 'make source' (and 'make') to work properly
(This used to be commit 341e12496ced6f3b0f110be7465538d4548b1b46)

12 years agor21216: fail the RAW-BENCH-LOCK test if the locking doesn't happen evenly
Andrew Tridgell [Wed, 7 Feb 2007 07:11:20 +0000 (07:11 +0000)]
r21216: fail the RAW-BENCH-LOCK test if the locking doesn't happen evenly
across the connections
(This used to be commit 01ef699981f3bcbbbdd4a47d9b85324aec783451)

12 years agor21215: extend the RAW-MUX test to test cancelling a async lock request using
Andrew Tridgell [Wed, 7 Feb 2007 07:10:23 +0000 (07:10 +0000)]
r21215: extend the RAW-MUX test to test cancelling a async lock request using
a SMBexit
(This used to be commit 753a85eff6a19af2c7b0884f705e9f66b5f605fe)

12 years agor21214: fixed a valgrind error that can be caused by a semi-async call inside
Andrew Tridgell [Wed, 7 Feb 2007 07:06:28 +0000 (07:06 +0000)]
r21214: fixed a valgrind error that can be caused by a semi-async call inside
a nested ntvfs call. The req structure can go away while processing a
ntvfs request
(This used to be commit f62b3c505f71f37a86a76d152d643926e19eb148)

12 years agor21213: the build farm seems to show that FreeBSD 6.2 has two different error
Andrew Tridgell [Wed, 7 Feb 2007 03:21:56 +0000 (03:21 +0000)]
r21213: the build farm seems to show that FreeBSD 6.2 has two different error
codes for "Invalid argument". This change should help track that down
(This used to be commit 608aac7804754fed1c68b847676d7a71a586e4bb)

12 years agor21212: detect if the kernel does not support integrated epoll/aio when the
Andrew Tridgell [Wed, 7 Feb 2007 03:16:10 +0000 (03:16 +0000)]
r21212: detect if the kernel does not support integrated epoll/aio when the
event context is created. This allows the LOCAL-EVENT test to pass on
systems with have libaio but not the necessary kernel patches
(This used to be commit 2ff8abf0022824e6ae93019ee1b3391e651a8ee7)

12 years agor21211: fixed 3 places that assumed that torture_fail() didn't call
Andrew Tridgell [Wed, 7 Feb 2007 02:57:44 +0000 (02:57 +0000)]
r21211: fixed 3 places that assumed that torture_fail() didn't call
return. Having a lowercase macro that returns from a function is
probably a bad idea - it looks too much like a C function but doesn't
behave like one.
(This used to be commit e680656419f00fd8b4ddb996fd247bada0f0a51d)

12 years agor21208: fix a crash bug caused by r21205
Andrew Tridgell [Wed, 7 Feb 2007 00:54:33 +0000 (00:54 +0000)]
r21208: fix a crash bug caused by r21205

please test patches, even if they look trivial!
(This used to be commit 8134a628f06f0df2ee94612854e0fb3e2efc2eb8)

12 years agor21206: - a couple more nttrans places were a memcpy() should be used
Andrew Tridgell [Wed, 7 Feb 2007 00:21:29 +0000 (00:21 +0000)]
r21206: - a couple more nttrans places were a memcpy() should be used

- changed the setup array in nttrans to be a uint8 array, instead of a
  uint16 array. This makes it clearer that it is the job of the caller
  to do any byte swapping within that data
(This used to be commit fa3c9b29ae1c7c6a9bc97484cc386aeec9b5d266)

12 years agor21205: Fix warning.
Jelmer Vernooij [Wed, 7 Feb 2007 00:16:21 +0000 (00:16 +0000)]
r21205: Fix warning.

Trivial patch from "Lunar Child <>"
(This used to be commit 1b75a9ff2d7f91f13fbdfd136b40eca14867bf6f)

12 years agor21187: - Convert LdbBrowse to use TreeVirtual. This adds the following capabilities
Derrell Lipman [Tue, 6 Feb 2007 21:00:23 +0000 (21:00 +0000)]
r21187: - Convert LdbBrowse to use TreeVirtual.  This adds the following capabilities
  that Simo asked for:

  * One can now open a branch without selecting the row.  The previously
    selected row (if any) remains selected.

  * Selecting a row does not automatically open the branch.  That can be done
    by either clicking on the open/close button or pressing Enter while that
    row is selected.

  * The entire tree can now be disabled.  In LdbBrowse, the fsm is configured
    to disable the tree during each remote procedure call, so that a pile of
    requests don't get queued during RPC.

  The most obvious flaw with TreeVirtual right now, is that the tree does not
  properly resize (or even initially size itself) to its container.  Fixing
  that is my next task.  It's actually a generic Table issue that people have
  been asking for for ages: having columns resize themselves automagically.
(This used to be commit 23a8b777638f7f40d95e125de97d077ae964e796)

12 years agor21185: Fix PNG images in TreeVirtual, for when using IE
Derrell Lipman [Tue, 6 Feb 2007 20:57:22 +0000 (20:57 +0000)]
r21185: Fix PNG images in TreeVirtual, for when using IE
(This used to be commit 1dde7f477b1e887fab7a24845673510eb2c1c084)

12 years agor21183: Commit the big/little endian fix for nttrans setup[] handling.
Volker Lendecke [Tue, 6 Feb 2007 18:16:10 +0000 (18:16 +0000)]
r21183: Commit the big/little endian fix for nttrans setup[] handling.

Please check!

For code symmetry reasons, trans2 might also see this fix, but this is left to
the interested reader :-)

(This used to be commit f2d364ebe69fc5d624e456cc7d941071811a5567)

12 years agor21179: Anything more complex than this causes the keytab never to be updated...
Andrew Bartlett [Tue, 6 Feb 2007 12:27:23 +0000 (12:27 +0000)]
r21179: Anything more complex than this causes the keytab never to be updated...

Andrew Bartlett
(This used to be commit c3977b4bae1e1b5e4ff4a64c7146534536685e91)

12 years agor21178: Adapt RPC-SAMBA3-SHARESEC to new smbd behavior
Volker Lendecke [Tue, 6 Feb 2007 10:48:32 +0000 (10:48 +0000)]
r21178: Adapt RPC-SAMBA3-SHARESEC to new smbd behavior
(This used to be commit 38b688fd15a9103e5ef5c86e01b594169d6b6b9c)

12 years agor21175: Fix the kerberos keytab update code to handle deletes.
Andrew Bartlett [Tue, 6 Feb 2007 05:41:04 +0000 (05:41 +0000)]
r21175: Fix the kerberos keytab update code to handle deletes.

Fix the join code to know that the ldb layer handles the keytab update.

Andrew Bartlett
(This used to be commit d3fbc089f4161ae71b21077d50130fdabd8b2d77)

12 years agor21174: many thanks to Paul Wayper for pointing out that C99 requires a
Andrew Tridgell [Tue, 6 Feb 2007 05:26:25 +0000 (05:26 +0000)]
r21174: many thanks to Paul Wayper for pointing out that C99 requires a
matching va_end() for each va_copy(). This doesn't matter for most
architectures, but there could be some obscure ones where it does

some of this should be ported to Samba3
(This used to be commit 21eb316473486cb6b73bb3ff9c5f3a44ecd57e4a)

12 years agor21173: - keep the ctdb queue when reconnecting (this will need to be more
Andrew Tridgell [Tue, 6 Feb 2007 04:47:44 +0000 (04:47 +0000)]
r21173: - keep the ctdb queue when reconnecting (this will need to be more
  complex once we have code for changing the number of nodes)

- fixed a ctdb/tcp bug in the handling of multiple packets waiting on
  a socket
(This used to be commit faf9a1146cb5f33228a11e01b66cb3cbe18bd05d)

12 years agor21172: fixed a comment
Andrew Tridgell [Tue, 6 Feb 2007 04:45:26 +0000 (04:45 +0000)]
r21172: fixed a comment
(This used to be commit 2d6e082e813e18dcd43bc8d9a8f445f77e0a8452)

12 years agor21171: fixed a bug related to recursive event handling.
Andrew Tridgell [Tue, 6 Feb 2007 04:43:48 +0000 (04:43 +0000)]
r21171: fixed a bug related to recursive event handling.

If this happens:

  - two sockets are readable, and select/epoll/aio returns both of
  - read event on socket1 is called
  - inside that read event an event_loop_once is called, this returns that
    socket2 is readable
  - read event on socket2 is called
  - event_loop_once returns
  - top level event handler then calls read event on socket2 (as it
    still has that listed as readable)
  - read handler for socket2 returns zero byte read, which is
    interpreted as end of file
  - socket is incorrectly closed

this happened with ctdb, but it could happen anywhere (just
rarely). The fix is trivial - ensure we break out of the event loop
when we have been called recursively.
(This used to be commit e042002bb5ee8974220e1ade56b64389571f75a6)

12 years agor21170: Latest qooxdoo required a minor server change
Derrell Lipman [Tue, 6 Feb 2007 04:42:52 +0000 (04:42 +0000)]
r21170: Latest qooxdoo required a minor server change
(This used to be commit a4a394c9bb42ed1883c95c10f9c24987d4ab4294)

12 years agor21169: - Step 3: Upgrade qooxdoo release code with latest qooxdoo svn's Table and
Derrell Lipman [Tue, 6 Feb 2007 04:35:30 +0000 (04:35 +0000)]
r21169: - Step 3: Upgrade qooxdoo release code with latest qooxdoo svn's Table and
  TreeVirtual.  The TreeVirtual that was in the release was a very early
  pre-release, and had a number of problems and missing features.  This brings
  it up-to-date and provides what seems to be a pretty well-working and highly
  useful widget.
(This used to be commit 194880f1ffde7f01997c120136854acd145189e9)

12 years agor21168: - Step 2: Update swat for the latest qooxdoo version. The build now copies
Derrell Lipman [Tue, 6 Feb 2007 03:31:23 +0000 (03:31 +0000)]
r21168: - Step 2: Update swat for the latest qooxdoo version.  The build now copies
  only those icons which are actually used.  This means that instead of a few
  thousand files getting copied to /usr/local/samba/share/webapps (or
  wherever), the number is less than an order of magnitude fewer.  This should
  make Andrew happier yet.
(This used to be commit 754c4fe01f91ecdbf156c757804118ef301bffe3)

12 years agor21167: - Upgrade to latest released qooxdoo. This is step 1 of the upgrade process,
Derrell Lipman [Tue, 6 Feb 2007 03:23:52 +0000 (03:23 +0000)]
r21167: - Upgrade to latest released qooxdoo.  This is step 1 of the upgrade process,
  which involves adding the latest SDK.  This new version has not 2, but 4
  icon themes.  In order to quiet Andrew B. who thinks that 9000+ files in a
  gui api is too many (come on, disk space is free these days :-), I have
  removed 3 of the 4 icon themes, leaving only the default, Nuvola.  That
  lowers the file count by 1/3.  I'm sure Andrew still isn't happy, but I hope
  he's a bit happier...  Step 2 will make him happier yet.
(This used to be commit d161c1382c36238105b85d0499d5a1011f580f52)

12 years agor21142: fix compiler warnings
Stefan Metzmacher [Mon, 5 Feb 2007 14:20:35 +0000 (14:20 +0000)]
r21142: fix compiler warnings

(This used to be commit 1f8a037ac4f592d29f7d66e1f924efe1c5d8c2b0)

12 years agor21141: Attempt to fix the build
Volker Lendecke [Mon, 5 Feb 2007 13:42:54 +0000 (13:42 +0000)]
r21141: Attempt to fix the build
(This used to be commit 3f41a4b014a47b3c751066f5d1bef7f76f4be831)

12 years agor21140: enable the pending lock notify code in the ctdb backend so that timed
Andrew Tridgell [Mon, 5 Feb 2007 03:23:45 +0000 (03:23 +0000)]
r21140: enable the pending lock notify code in the ctdb backend so that timed
locks retry immediately when another blocking lock is removed.
(This used to be commit 169920a5341392e84d5b5fb5f39c2b9d5243bb3b)

12 years agor21139: make the RAW-MUX test fail if the locking subsystem doesn't trigger
Andrew Tridgell [Mon, 5 Feb 2007 03:04:01 +0000 (03:04 +0000)]
r21139: make the RAW-MUX test fail if the locking subsystem doesn't trigger
early byte range lock retries (ie. PENDING locks in brlock)
(This used to be commit 09237e3acf524f7e772acce2e652ef30b8ba4d65)

12 years agor21137: If you change the name of a constant in one file, it helps to ensure that...
Derrell Lipman [Sun, 4 Feb 2007 19:58:16 +0000 (19:58 +0000)]
r21137: If you change the name of a constant in one file, it helps to ensure that the use of the constant in other files is corrected as well.
(This used to be commit 1773d430176bc3cab37431c6a9991bc5c9c66dcd)

12 years agor21136: First attempt at implementing add/modify rpc calls.
Simo Sorce [Sun, 4 Feb 2007 19:21:17 +0000 (19:21 +0000)]
r21136: First attempt at implementing add/modify rpc calls.
The code does not work, but Derrell asked me to commit to analyze the problem.

(This used to be commit 433591c430891fd5f0859fe3b685287b4fb166e5)

12 years agor21135: Instead of having hooks to update keytabs as an explicit thing, update
Andrew Bartlett [Sun, 4 Feb 2007 07:17:03 +0000 (07:17 +0000)]
r21135: Instead of having hooks to update keytabs as an explicit thing, update
them as a hook on ldb modify, via a module.

This should allow the secrets.ldb to be edited by the admin, and to
have things update in the on-disk keytab just as an in-memory keytab

This isn't really a dsdb plugin, but I don't have any other good ideas
about where to put it.

Andrew Bartlett
(This used to be commit 6ce557a1aff4754d2622be8f1c6695d9ee788d54)

12 years agor21134: Attempt to quieten the build farm for Samba3
Volker Lendecke [Sat, 3 Feb 2007 17:46:32 +0000 (17:46 +0000)]
r21134: Attempt to quieten the build farm for Samba3
(This used to be commit 966c30ca591c410a4a229396c7919fe8f98bd066)

12 years agor21126: Enhance the modal window so that it gets focus and just presing enter works.
Simo Sorce [Fri, 2 Feb 2007 17:38:04 +0000 (17:38 +0000)]
r21126: Enhance the modal window so that it gets focus and just presing enter works.
Still I don't see the cursor :(

Swap -/+ buttons, you risk to delete an attribute by mistake when instead you want to add a new one,
and you might even not notice it.

(This used to be commit afa61d59714686ceef1e33c46e8b504d5030eb10)

12 years agor21125: Tridge, please check this closely if this relaxed notify test is
Volker Lendecke [Fri, 2 Feb 2007 17:30:10 +0000 (17:30 +0000)]
r21125: Tridge, please check this closely if this relaxed notify test is

In Samba3 with inotify we have the problem that the MODIFIED action is sent
second. I think this could also happen in Samba4, depending on the order when
the different sockets (unix domain dgram / inotify) are looked at by the
events system.

This patch accepts the NEW/OLD/MODIFIED entries in any order.

Comments welcome!

(This used to be commit 72a7771b757c3aca637eb632abebab60bdd2da9f)

12 years agor21124: Samba3 does not support create times correctly.
Volker Lendecke [Fri, 2 Feb 2007 15:36:17 +0000 (15:36 +0000)]
r21124: Samba3 does not support create times correctly.

(This used to be commit c9886ce63b8cc085e0032ad2f6193c547336856d)

12 years agor21118: Looks a bit better ...
Simo Sorce [Thu, 1 Feb 2007 23:53:27 +0000 (23:53 +0000)]
r21118: Looks a bit better ...
(This used to be commit e2d1f6d291602d56011c7099485ecd1a5767fca5)

12 years agor21104: Rename struct watch_context to struct inotify_watch_context
Volker Lendecke [Thu, 1 Feb 2007 10:59:35 +0000 (10:59 +0000)]
r21104: Rename struct watch_context to struct inotify_watch_context
(This used to be commit 3548b1f683a7f5ccc6563de7e099d3a5a86ed39c)

12 years agor21103: This seems to do the 'right thing' in applying the correct access
Andrew Bartlett [Thu, 1 Feb 2007 01:27:02 +0000 (01:27 +0000)]
r21103: This seems to do the 'right thing' in applying the correct access
control to the SWAT ldb browser.  This should show up as the
authenticated user, rather than anonymous now.

Andrew Bartlet
(This used to be commit 48a5105f6bb0afcd22feb9448aba3006e6646848)

12 years agor21095: Expose all naming contexts
Simo Sorce [Wed, 31 Jan 2007 17:57:44 +0000 (17:57 +0000)]
r21095: Expose all naming contexts
(This used to be commit d25e828dbe479347894b197c94199c5540ac34e5)

12 years agor21086: add some notes
Stefan Metzmacher [Wed, 31 Jan 2007 14:05:09 +0000 (14:05 +0000)]
r21086: add some notes

(This used to be commit dfbdb958419a041c1aff039810f7e9cb4320ec2c)

12 years agor21085: implement something like the parameter
Stefan Metzmacher [Wed, 31 Jan 2007 13:47:52 +0000 (13:47 +0000)]
r21085: implement something like the parameter


(This used to be commit eee635ab4d7ae8ffd8962b3d4518a7e6b5a90f97)

12 years agor21078: TP4 is just released
Stefan Metzmacher [Wed, 31 Jan 2007 12:28:05 +0000 (12:28 +0000)]
r21078: TP4 is just released

(This used to be commit 1c942b7d8ffb97b70f17fd901033f54771274b96)

12 years agor21075: Generate parameters in structs for the server side Samba 3 code. The current...
Jelmer Vernooij [Wed, 31 Jan 2007 11:54:01 +0000 (11:54 +0000)]
r21075: Generate parameters in structs for the server side Samba 3 code. The current code in Samba 3 is already generated using this pidl patch.
(This used to be commit 30a06273d28fb59c6f20154f97c67f4c0a63dee9)

12 years agor21072: fix compiler warning
Stefan Metzmacher [Wed, 31 Jan 2007 09:25:37 +0000 (09:25 +0000)]
r21072: fix compiler warning

(This used to be commit e788709835fa13b512fbf38951c9d0ca9bc3df18)

12 years agor21071: Move some of the key path variables into the script.
Andrew Bartlett [Wed, 31 Jan 2007 07:08:36 +0000 (07:08 +0000)]
r21071: Move some of the key path variables into the script.

Andrew Bartlett
(This used to be commit 1c466c6d6f02bf76f7f68e7dcd3d9a4c2a63179a)

12 years agor21069: Try to split up the script into parts to deal with each
Andrew Bartlett [Wed, 31 Jan 2007 05:11:19 +0000 (05:11 +0000)]
r21069: Try to split up the script into parts to deal with each
LDAP implementation, and another to hold the key blobs.

Also fix the OpenLDAP test.

Andrew Bartlett
(This used to be commit ec511c592bbb7df513fe6f62e06e1fa984431550)

12 years agor21068: Code to configure, start and stop Fedora DS.
Andrew Bartlett [Wed, 31 Jan 2007 01:28:55 +0000 (01:28 +0000)]
r21068: Code to configure, start and stop Fedora DS.

We can't actually use it, as we are stuck inside our socket_wrapper
network, and can't talk to it.

Andrew Bartlett
(This used to be commit bd564da6b9b45c438ef81b42dc156116ed81cbe7)