Stefan Metzmacher [Wed, 27 Feb 2008 21:22:56 +0000 (22:22 +0100)]
pvfs_open: fix callers of odb_get_delete_on_close()
metze
Stefan Metzmacher [Wed, 27 Feb 2008 21:16:55 +0000 (22:16 +0100)]
opendb: only return delete_on_close on odb_get_delete_on_close()
metze
Stefan Metzmacher [Wed, 27 Feb 2008 21:22:10 +0000 (22:22 +0100)]
pvfs: remove unused args from pvfs_delete_on_close_set()
metze
Stefan Metzmacher [Wed, 27 Feb 2008 20:52:52 +0000 (21:52 +0100)]
pvfs_open: use the delete_path of odb_close_file()
metze
Stefan Metzmacher [Wed, 27 Feb 2008 20:50:51 +0000 (21:50 +0100)]
opendb: return the path that should be deleted in odb_close_file()
That means the last close returns the path name if the
delete_on_close flag is set.
metze
Andrew Bartlett [Wed, 27 Feb 2008 23:43:57 +0000 (10:43 +1100)]
Ensure we don't try and set the acct_flags if they are 0 (meaning
unchanged in this interface).
We seem to have two very similar interfaces here, and this is the
poorer interface, for this reason.
Andrew Bartlett
Andrew Bartlett [Wed, 27 Feb 2008 23:33:54 +0000 (10:33 +1100)]
Merge branch 'v4-0-test' of git://git.samba.org/samba into 4-0-local
Andrew Bartlett [Wed, 27 Feb 2008 23:32:06 +0000 (10:32 +1100)]
Fix up the libnet tests.
These tests still do not pass against windows, but at least now it
doesn't reset acct_flags to 0.
This is enough to have Samba4 pass it's own tests for the moment.
Andrew Bartlett
Andrew Bartlett [Wed, 27 Feb 2008 23:05:32 +0000 (10:05 +1100)]
Check for and reject invalid account flags.
(lest we have an account set with 0 flags)
Andrew Bartlett
Andrew Bartlett [Wed, 27 Feb 2008 21:50:00 +0000 (08:50 +1100)]
Generate ACB_PW_EXPIRED correctly
More correctly handle expired passwords, and do not expire machine accounts.
Test that the behaviour is consistant with windows, using the RPC-SAMR test.
Change NETLOGON to directly query the userAccountControl, just because
we don't want to do the extra expiry processing here.
Andrew Bartlett
Andrew Bartlett [Wed, 27 Feb 2008 21:43:10 +0000 (08:43 +1100)]
Users and computers now share the same template.
Slowly work away at the samldb module again, it is clear that AD does
not use much of a templating system. samAccountType is managed, as
far as I can tell, when groupType or userAccountControl changes.
Andrew Bartlett
Andrew Bartlett [Wed, 27 Feb 2008 21:39:45 +0000 (08:39 +1100)]
Reorder modules to have rdn_name before objectclass.
This ensures the relative DN is placed in the correct case into the
DB.
Andrew Bartlett
Andrew Bartlett [Wed, 27 Feb 2008 21:38:53 +0000 (08:38 +1100)]
Fix rdn_name errors.
Return the correct error when the DN is mismatched with it's RDN
attribute (now matches AD).
Andrew Bartlett
Andrew Bartlett [Wed, 27 Feb 2008 21:37:11 +0000 (08:37 +1100)]
Extend the ldap.js test
Now we verify sAMAccountType and userAccountControl, as well as rename and
DN mismatch semantics.
Andrew Bartlett
Andrew Bartlett [Wed, 27 Feb 2008 21:36:13 +0000 (08:36 +1100)]
Do not re-randomise in an individual test.
smbtorture calls srandom() once for the whole suite, with command line
override to allow reproduceable random tests.
Andrew Bartlett
Stefan Metzmacher [Wed, 27 Feb 2008 18:47:23 +0000 (19:47 +0100)]
unixinfo.idl: don't use utf8string anymore
Note:
This changes the on wire format!!!
But it's a very bad idea to keep utf8string
as it's not valid NDR encoding and noone
would be able to write a compatible IDL
file using another compiler than pidl.
Maybe we should change the uuid,
but as this isn't really used yet
I keep the old one.
metze
Jelmer Vernooij [Wed, 27 Feb 2008 18:37:45 +0000 (19:37 +0100)]
Merge branch 'v4-0-test' of ssh://git.samba.org/data/git/samba into v4-0-test
Jelmer Vernooij [Thu, 21 Feb 2008 15:18:01 +0000 (16:18 +0100)]
Clarify comments.
Jelmer Vernooij [Wed, 27 Feb 2008 16:36:49 +0000 (17:36 +0100)]
Set lp_ctx properly.
Stefan Metzmacher [Wed, 27 Feb 2008 15:23:57 +0000 (16:23 +0100)]
RAW-OPLOCK: be more strict with share modes against windows and samba4
But still allow samba3 behavior.
metze
Stefan Metzmacher [Wed, 27 Feb 2008 15:22:45 +0000 (16:22 +0100)]
selftest: samba4 doesn't pass RAW-OPLOCK BATCH19 und 20 yet
metze
Stefan Metzmacher [Wed, 27 Feb 2008 15:22:04 +0000 (16:22 +0100)]
RAW-OPLOCK: use torture_result(TORTURE_FAIL) so that the knownfailures file work
metze
Stefan Metzmacher [Wed, 27 Feb 2008 14:45:08 +0000 (15:45 +0100)]
RAW-OPLOCK: add BATCH20 test renaming via setpathinfo() and setfileinfo()
metze
Stefan Metzmacher [Wed, 27 Feb 2008 14:43:04 +0000 (15:43 +0100)]
RAW-OPLOCK: add BATCH19 test renaming via setpathinfo
This is really strange as, it seems to complete ignore
oplocks and share modes and doesn't match the behavior
of RENAME and NTRENAME.
metze
Stefan Metzmacher [Wed, 27 Feb 2008 10:07:53 +0000 (11:07 +0100)]
RAW-OPLOCK: add BATCH18 test a ntrename with FLAG_RENAME
metze
Stefan Metzmacher [Wed, 27 Feb 2008 09:41:47 +0000 (10:41 +0100)]
RAW-OPLOCK: add BATCH17 test rename on a file with oplock
metze
Stefan Metzmacher [Wed, 27 Feb 2008 09:40:19 +0000 (10:40 +0100)]
RAW-OPLOCK: add EXCLISIVE6 test a rename on a file with an oplock
metze
Stefan Metzmacher [Wed, 27 Feb 2008 09:02:52 +0000 (10:02 +0100)]
RAW-OPLOCK: add EXCLUSIVE5 to test attribute only with OVERWRITE_IF
metze
Stefan Metzmacher [Wed, 27 Feb 2008 08:41:44 +0000 (09:41 +0100)]
RAW-OPLOCK: add BATCH16 and tests attribute only with OVERWRITE_IF
metze
Stefan Metzmacher [Wed, 27 Feb 2008 10:25:33 +0000 (11:25 +0100)]
pvfs_rename: we need a do a odb_rename() after pvfs_do_rename()
metze
Stefan Metzmacher [Wed, 27 Feb 2008 10:24:37 +0000 (11:24 +0100)]
pvfs_rename: add retry logic after sharing violation or non granted oplock
metze
Stefan Metzmacher [Wed, 27 Feb 2008 08:40:49 +0000 (09:40 +0100)]
opendb_tdb: EXCLUSIVE oplock use the same matching logic
metze
Stefan Metzmacher [Wed, 27 Feb 2008 07:51:25 +0000 (08:51 +0100)]
RAW-OPLOCK: add EXCLUSIVE4 a attribute only open doesn't break an exclusive oplock
metze
Stefan Metzmacher [Tue, 26 Feb 2008 15:16:31 +0000 (16:16 +0100)]
RAW-OPLOCK: add EXCLUSIVE3
This tests that a setpathinfo EOF breaks
an exclusive oplock to none.
metze
Stefan Metzmacher [Tue, 26 Feb 2008 14:51:17 +0000 (15:51 +0100)]
RAW-OPLOCK: add EXCLUSIVE2
Exclusive oplocks break to LEVEL2
metze
Stefan Metzmacher [Tue, 26 Feb 2008 13:58:00 +0000 (14:58 +0100)]
RAW-OPLOCK: rename test (NORMAL => EXCLUSIVE1)
metze
Matthias Dieter Wallnöfer [Wed, 27 Feb 2008 12:47:34 +0000 (13:47 +0100)]
Fix cases, add version number. (#4935)
Jelmer Vernooij [Wed, 27 Feb 2008 12:17:08 +0000 (13:17 +0100)]
Merge branch 'v4-0-test' of ssh://git.samba.org/data/git/samba into v4-0-gmake-simple
Michael Adam [Wed, 27 Feb 2008 09:33:32 +0000 (10:33 +0100)]
libreplace: standalone build: use -I$srcdir instead of -I.
Michael
Michael Adam [Wed, 27 Feb 2008 00:41:30 +0000 (01:41 +0100)]
libreplace: fix standalone build - add necessary libs.
The libs needed for getifaddrs replacements have to be added
to LIBS and used for the testsuite target.
Michael
Michael Adam [Wed, 27 Feb 2008 00:29:12 +0000 (01:29 +0100)]
libreplace: fix standalone build on some systems.
getifaddr tests include system/network.h, which does not
find getaddrinfo.h without "-I.".
Michael
Michael Adam [Tue, 26 Feb 2008 12:24:54 +0000 (13:24 +0100)]
libreplace: Add tests for connect and gethostbyname.
Provide dummy replacements when a function isnt found.
The functions are also searched for in certain libraries,
and variables SOCKET_LIBS and NSL_LIBS are set accordingly.
One purpose of this is to fix the getifaddrs tests on
systems where e.g. the socket calls require special libs
for linking.
Michael
Michael Adam [Tue, 26 Feb 2008 14:38:45 +0000 (15:38 +0100)]
pidl: Remove superfluous semicolon in perl code.
Michael
Stefan Metzmacher [Tue, 26 Feb 2008 13:52:50 +0000 (14:52 +0100)]
RAW-OPLOCK: BATCH15: make sure a qpathinfo doesn't break batch oplocks
metze
Jelmer Vernooij [Tue, 26 Feb 2008 13:20:31 +0000 (14:20 +0100)]
Rely on GNU make to build static libraries.
Conflicts:
source/build/smb_build/makefile.pm
Stefan Metzmacher [Tue, 26 Feb 2008 11:52:46 +0000 (12:52 +0100)]
opendb_tdb: with break_to_none attribute only opens also break oplocks
metze
Stefan Metzmacher [Tue, 26 Feb 2008 09:26:33 +0000 (10:26 +0100)]
pvfs_open: pass down an access mask to pvfs_can_update_file_size()
You just need SEC_FILE_WRITE_ATTRIBUTE to change
the filesize...
metze
Stefan Metzmacher [Tue, 26 Feb 2008 09:29:07 +0000 (10:29 +0100)]
pvfs_qfileinfo: down discard the return value of pvfs_can_stat()
The odb_can_open() code returns DELETE_PENDING if
a delete is really pending.
metze
Stefan Metzmacher [Tue, 26 Feb 2008 09:28:07 +0000 (10:28 +0100)]
pvfs_open: pass down an access mask in pvfs_can_stat()
metze
Stefan Metzmacher [Tue, 26 Feb 2008 09:21:39 +0000 (10:21 +0100)]
pvfs_open: pass NTCREATEX_DISP_OPEN to odb_can_open()
As 0 is NTCREATEX_DISP_SUPERSEDE and that's not what we want here.
metze
Michael Adam [Tue, 26 Feb 2008 10:18:51 +0000 (11:18 +0100)]
Fix build on VOS: socklen_t is needed for rep_inet_ntop: move it up.
Michael
Stefan Metzmacher [Tue, 26 Feb 2008 09:31:37 +0000 (10:31 +0100)]
BASE-DEFER_OPEN: be less strict with to late responses
The server might be just busy.
metze
Stefan Metzmacher [Sat, 23 Feb 2008 10:54:01 +0000 (11:54 +0100)]
selftest: Samba4 passes all oplocks tests now
The support for oplocks is not completely finished
I believe we will not pass some non existing test.
But it's a good start.
metze
Stefan Metzmacher [Sat, 23 Feb 2008 10:50:19 +0000 (11:50 +0100)]
pvfs_setpathinfo: retry setpathinfo after oplock not granted on on setpathinfo ALLOCATION_INFO and END_OF_FILE_INFO
metze
Stefan Metzmacher [Sat, 23 Feb 2008 10:49:39 +0000 (11:49 +0100)]
pvfs_open: add pvfs_can_update_file_size()
TODO: this is not complete, we need more tests to trigger this
metze
Stefan Metzmacher [Fri, 22 Feb 2008 10:52:17 +0000 (11:52 +0100)]
pvfs_unlink: retry unlink after oplock not granted
metze
Stefan Metzmacher [Mon, 25 Feb 2008 18:17:45 +0000 (19:17 +0100)]
pvfs: add posix:oplocktimeout=30 option
metze
Stefan Metzmacher [Sat, 23 Feb 2008 10:46:43 +0000 (11:46 +0100)]
pvfs_setfileinfo: break level2 oplocks on setfileinfo() ALLOCATION_INFO and END_OF_FILE_INFO
metze
Stefan Metzmacher [Fri, 22 Feb 2008 15:34:50 +0000 (16:34 +0100)]
pvfs: send oplock breaks to none to level2 holders on write/lock requests
metze
Stefan Metzmacher [Fri, 22 Feb 2008 15:34:50 +0000 (16:34 +0100)]
pvfs_oplocks: add pvfs_break_level2_oplocks()
metze
Stefan Metzmacher [Mon, 25 Feb 2008 08:51:58 +0000 (09:51 +0100)]
pvfs: handle oplock releases in its own function pvfs_oplock_release()
metze
Stefan Metzmacher [Mon, 25 Feb 2008 07:39:13 +0000 (08:39 +0100)]
pvfs_open: call pvfs_setup_oplock() if an oplock was granted
This is needed to receive oplock breaks from other "processes"
metze
Stefan Metzmacher [Mon, 25 Feb 2008 07:32:35 +0000 (08:32 +0100)]
pvfs: add pvfs_setup_oplock() to receive oplock breaks and pass them to the client
metze
Stefan Metzmacher [Thu, 21 Feb 2008 16:48:13 +0000 (17:48 +0100)]
pvfs_open: make the retry logic indepdendent from open and sharing violations
metze
Stefan Metzmacher [Mon, 25 Feb 2008 16:50:22 +0000 (17:50 +0100)]
pvfs_open: fix odb_can_open() callers after prototype change
metze
Stefan Metzmacher [Mon, 25 Feb 2008 16:48:13 +0000 (17:48 +0100)]
opendb: unify the parameters of odb_open_file() and odb_can_open()
metze
Stefan Metzmacher [Mon, 25 Feb 2008 15:49:40 +0000 (16:49 +0100)]
pvfs_open: pass down open_disposition and break_to_none to odb_open_file()
metze
Stefan Metzmacher [Mon, 25 Feb 2008 15:47:48 +0000 (16:47 +0100)]
opendb: pass down open_disposition and break_to_none to odb_open_file()
metze
Stefan Metzmacher [Tue, 26 Feb 2008 08:14:54 +0000 (09:14 +0100)]
pvfs_setfileinfo_rename: map DELETE_PENDING to ACCESS_DENIED
This is needed as odb_can_open/pvfs_can_delete changed the return value.
metze
Stefan Metzmacher [Mon, 25 Feb 2008 19:20:35 +0000 (20:20 +0100)]
opendb_tdb: only file->delete_on_close == true should give DELETE_PENDING
metze
Stefan Metzmacher [Mon, 25 Feb 2008 15:26:04 +0000 (16:26 +0100)]
opendb_tdb: move sharemode, oplock logic into odb_tdb_open_can_internal()
metze
Stefan Metzmacher [Mon, 25 Feb 2008 15:14:23 +0000 (16:14 +0100)]
opendb_tdb: add force break to none logic
metze
Stefan Metzmacher [Fri, 22 Feb 2008 16:26:40 +0000 (17:26 +0100)]
opendb_tdb: attribute only opens doesn't conflict with BATCH oplocks
metze
Stefan Metzmacher [Fri, 22 Feb 2008 16:26:40 +0000 (17:26 +0100)]
opendb_tdb: grant level2 oplocks
metze
Stefan Metzmacher [Fri, 22 Feb 2008 15:30:13 +0000 (16:30 +0100)]
opendb: add odb_break_oplocks() function
This send breaks to none to all level2 holders
metze
Andrew Bartlett [Tue, 26 Feb 2008 05:24:09 +0000 (16:24 +1100)]
Remove more cruft about smbscript.
Andrew Bartlett
Andrew Bartlett [Tue, 26 Feb 2008 05:19:57 +0000 (16:19 +1100)]
Merge branch 'v4-0-test' of git://git.samba.org/samba into 4-0-local
Andrew Bartlett [Tue, 26 Feb 2008 05:19:42 +0000 (16:19 +1100)]
Make use of smbpython clear.
Andrew Bartlett
Kai Blin [Tue, 26 Feb 2008 05:17:04 +0000 (06:17 +0100)]
howto: We need bin/smbpython to run python provisioning
Simo Sorce [Mon, 25 Feb 2008 19:03:07 +0000 (14:03 -0500)]
Update homepages for talloc, tdb and ldb subprojects
Stefan Metzmacher [Thu, 21 Feb 2008 15:12:27 +0000 (16:12 +0100)]
opendb: add odb_update_oplock() call
metze
Stefan Metzmacher [Thu, 21 Feb 2008 11:20:31 +0000 (12:20 +0100)]
opendb: send also the oplock break level on MSG_NTVFS_OPLOCK_BREAK
metze
Stefan Metzmacher [Fri, 22 Feb 2008 09:18:13 +0000 (10:18 +0100)]
opendb: add odb_get_key() function to get the key back from a struct odb_lock
metze
Stefan Metzmacher [Fri, 22 Feb 2008 08:41:41 +0000 (09:41 +0100)]
pvfs_unlink: move stream logic into pvfs_unlink_one()
metze
Stefan Metzmacher [Fri, 22 Feb 2008 08:37:22 +0000 (09:37 +0100)]
pvfs_unlink: splitup the logic into generic and file specific functions
metze
Stefan Metzmacher [Fri, 22 Feb 2008 08:30:51 +0000 (09:30 +0100)]
pvfs_unlink: add a fast path for the non wildcard case
metze
Stefan Metzmacher [Fri, 22 Feb 2008 08:28:51 +0000 (09:28 +0100)]
pvfs_unlink: move !name->stream_exists into the caller
metze
Stefan Metzmacher [Fri, 22 Feb 2008 08:23:53 +0000 (09:23 +0100)]
pvfs_unlink: pass down union smb_unlink completely to sub functions
metze
Stefan Metzmacher [Thu, 21 Feb 2008 19:12:46 +0000 (20:12 +0100)]
pvfs_unlink: pass down a struct pvfs_filename to pvfs_unlink_one()
metze
Stefan Metzmacher [Thu, 21 Feb 2008 18:56:36 +0000 (19:56 +0100)]
pvfs_open: unify talloc behavior in pvfs_can_delete/rename/stat()
And also handle NULL for lckp in the error path
without crashing.
metze
Stefan Metzmacher [Thu, 21 Feb 2008 19:01:25 +0000 (20:01 +0100)]
pvfs: handle SHARING_VIOLATION and OPLOCK_NOT_GRANTED in pvfs_can_delete/rename()
If the caller asks for the odb_lock return it
also if we return NT_STATUS_SHARING_VIOLATION
or NT_STATUS_OPLOCK_NOT_GRANTED so that the caller
can add the pending notification.
metze
Stefan Metzmacher [Thu, 21 Feb 2008 18:59:13 +0000 (19:59 +0100)]
pvfs: pass NULL to pvfs_can_*() when no odb_lock is needed by the caller
metze
Stefan Metzmacher [Thu, 21 Feb 2008 08:02:17 +0000 (09:02 +0100)]
pvfs_wait: use struct pvfs_wait * instead of void *
metze
Stefan Metzmacher [Thu, 21 Feb 2008 07:59:21 +0000 (08:59 +0100)]
pvfs_wait: 'private' -> 'private_data' and use talloc_get_type()
metze
Stefan Metzmacher [Sat, 23 Feb 2008 10:51:43 +0000 (11:51 +0100)]
RAW-OPLOCK: remove unused var
Stefan Metzmacher [Sat, 23 Feb 2008 10:51:23 +0000 (11:51 +0100)]
RAW-OPLOCK: fix typo (send oplock releases on the correct connection)
metze
Stefan Metzmacher [Mon, 25 Feb 2008 06:42:38 +0000 (07:42 +0100)]
selftest: we pass the BASE-DEFER_OPEN test
We just need to pass the same sharedelay time
to smbtorture as we use for smbd.
metze
Stefan Metzmacher [Sat, 23 Feb 2008 23:13:36 +0000 (00:13 +0100)]
BASE-DEFER_OPEN: add torture:sharedelays option to run this test faster
metze
Volker Lendecke [Sat, 23 Feb 2008 09:26:52 +0000 (10:26 +0100)]
Respect the return value of asprintf
This removes tons of warnings
warning: ignoring return value of 'fwrite', declared with attribute warn_unused_result
for me. Lots of that kind left though :-)
Michael Adam [Fri, 22 Feb 2008 12:53:05 +0000 (13:53 +0100)]
libreplace: don't use socketwrapper for getifaddrs() implementations.
Michael
(cherry picked from commit
85eec1d1d6b674294c50eb912fbe7d5a1dd42909)