summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2007-10-10r18979: With these extra indexes (also added for the normal case) and aAndrew Bartlett3-2/+35
DB_CONFIG file, we now get reasonable enought performance to pass 'make test' against OpenLDAP. We do have to double the maximum runtime for the torture client however. Andrew Bartlett (This used to be commit 5b3c8cc036c1180c9e96d9aaacd3f2e0a83460e5)
2007-10-10r18978: Fix bug found by:Andrew Bartlett1-3/+13
http://www.ee.oulu.fi/research/ouspg/protos/testing/c06/ldapv3/ The issue here is that if the UTF8 conversion fails, because this isn't actually UTF8 data, then we need to do a binary compare instead. Andrew Bartlett (This used to be commit a113e47784157ec6086b014c1fc998e8a23e7382)
2007-10-10r18977: Seperate these asserts, so we know which fired.Andrew Bartlett1-1/+2
Andrew Bartlett (This used to be commit 9b2003618b28cb045e74937803e9aad773781803)
2007-10-10r18975: follow the advice from "William Jojo" <jojowil@hvcc.edu>Stefan Metzmacher1-1/+1
and pass -bbigtoc also to the linker when building shared libs metze (This used to be commit 9c77de282107d28362497aff60fed81e77620a87)
2007-10-10r18974: we don't have the configure check that setsStefan Metzmacher1-5/+1
ac_cv_gnu_ld_no_default_allow_shlib_undefined and as -Wl,--allow-shlib-undefined is supported on SuSE 7.3 I don't see a problem is always pass it explicit metze (This used to be commit 79db6bddc5fc1b97f0165348f72b6a2489b62be0)
2007-10-10r18972: we don't need this now all builds in the build farm are limited to 150Andrew Tridgell1-12/+0
file descriptors. We'll pretty quickly find leaks :) (This used to be commit be70992d71d7b8db64702f9be2b29aac3e463590)
2007-10-10r18971: avoid strndup is a few places. Fixes a minor memory leak, and shouldAndrew Tridgell4-11/+21
fix RPC-LSA on AIX. (This used to be commit 6cce709d08579f4e00b44b692332a557b0ea3b86)
2007-10-10r18970: avoid strndup and strnlen on AIX. They are quite broken. SeeAndrew Tridgell2-2/+6
http://lists.samba.org/archive/samba-technical/2004-August/036915.html (This used to be commit c178c84f01166609e6bd3393d39fb0034130167b)
2007-10-10r18968: EWOULDBLOCK should also be mapped to STATUS_MORE_ENTRIESAndrew Tridgell1-1/+4
(This used to be commit 27114fe1752f20c58948b34264e38db263f7a0ea)
2007-10-10r18965: fixed the bug with RPC-NETLOGON and solaris sparc machines. This bugAndrew Tridgell1-0/+6
took a _LONG_ time to find. The problem was that when encoding/decoding password buffers we use the pull/push string functions, which by default align unicode strings. But on solaris sparc the buffer is not aligned always (its a stack variable, an array of uint8_t). That perfectly OK in C, so we just tell the pull/push functions not to auto-align. (This used to be commit bb7835eced00607eb6b1725be6d96a6dcb842049)
2007-10-10r18960: fix caching of some configure tests AC_CACHE_CHECK()Stefan Metzmacher2-20/+20
*needs* '_cv_' in the shell var to do caching... metze (This used to be commit cb334632fae28bff107200a7b5643289f383eaad)
2007-10-10r18957: we need to include unistd.h in system/network.h becauseStefan Metzmacher1-0/+4
it contains the ioctl() prototype on some systems and we need to make sure it gets included before socket_wrapper defines ioctl swrap_ioctl metze (This used to be commit f37b13cbe65404f27d31b8020afb2f542313f4d3)
2007-10-10r18948: fix the build on solaris, the ioctl() prototypeStefan Metzmacher1-1/+1
in unistd.h was included after libreplace defined ioctl swrap_ioctl metze (This used to be commit 55b25a9fcfa5ddf290e371c2bcad8ad07969f19a)
2007-10-10r18947: overload listen() and ioctl() in socket_wrapperStefan Metzmacher3-0/+46
metze (This used to be commit dfaccdca1b6954cd61828749d7b000f804f3b066)
2007-10-10r18946: fix compiler warningStefan Metzmacher1-1/+1
metze (This used to be commit 450595822441ebaa199d6b10d568274940b1a91b)
2007-10-10r18945: fix compiler warnings and end-of-non-void function bugsStefan Metzmacher2-11/+17
metze (This used to be commit ed195999c0c7d89cdc61e980576d191fc05d65d7)
2007-10-10r18944: talloc_reference() can fail...Stefan Metzmacher1-1/+1
(remove compiler warning) metze (This used to be commit 515f4eb1cef11d8f942aeffc72c9da1bbd4d78cd)
2007-10-10r18943: Samba4 wants its databases private to root. I know some of them couldAndrew Tridgell1-0/+5
potentially be public, but we'd need to think about them carefully. (This used to be commit 142a235f663f707eee3f16d0c1265b01fe1ecf32)
2007-10-10r18942: add a ldb_set_create_perms() function in ldb. I didn't call itAndrew Tridgell4-1/+16
ldb_set_umask() (which is what we had discussed) as it doesn't actually set the umask (in effect it sets the inverse of the umask - the perms to be used for the file) (This used to be commit 7e2ec875908c112d5c3b0f6d18f9a8bbacf33539)
2007-10-10r18939: don't rely on the umask being right in ldb creation. Both Samba3 andAndrew Tridgell1-1/+1
Samba4 smbd force the umask to 0, which meant we ended up with ldb being world writable. This isn't really an ideal fix, as it means ldb no longer honors umask (as it should do, like all good libraries). Unfortunately the 'proper' fix is too complex for now this also merges a tiny code style fix from s4 to s3 (This used to be commit 1a42f38dfdc55d7429a1f8d3e811f3d857195a58)
2007-10-10r18936: hopefully fix the test for negative enum values. When a compilerAndrew Tridgell1-9/+3
truncates the enums, the test was passing. (This used to be commit c6216f7dbf68012fa8351dcae17fe4c7ab165b58)
2007-10-10r18933: Add helpful emacs markerAndrew Bartlett1-0/+1
(This used to be commit 5fd53ea6a3a6d7638a9f542f46623a19143ff81d)
2007-10-10r18925: Add current snapshot of the ejs-2.0 code. Tridge, will you be ↵Derrell Lipman80-0/+33339
incorporating this? (This used to be commit 917af234a8d517f82bd42256a940608a16b988f4)
2007-10-10r18918: - bail out with unsupported option to socket()Stefan Metzmacher1-18/+53
- don't reuse portnumbers in the autobind code - use if (!...) return; logic instead of if (...) { do everything } return for swrap_close metze (This used to be commit 68a65aa4490c12ca70db335f928f3ac507902337)
2007-10-10r18917: having 255 virtual interfaces available in socket wrapper means weAndrew Tridgell3-8/+10
stat() 240 files that don't exist on every broadcast. That's a bit excessive! reduce max virtual interfaces to 16 (This used to be commit 3c4100027c7d3806a2021cb4d70ec6adf9dd2dc6)
2007-10-10r18916: fixed the messaging layer on *BSD systems. When a socket was full weAndrew Tridgell1-1/+1
were getting ENOBUFS, which mapped to NT_STATUS_NO_MEMORY, which in turn caused the messaging code to loop trying until it gave up. Now it correctly falls back to select. Messaging speed goes from 3 messages per second to over 7000 on my test vmware box. Not bad for a one line change :) (This used to be commit 6568f30adf980c572f9ffd6ff884336ebe652f90)
2007-10-10r18915: fix usage of popt, do make -W setting the recursive flag on SuSE 10.1Stefan Metzmacher1-19/+58
metze (This used to be commit 0fef3f807dd3e156970e315d83f5cb4a45e20d12)
2007-10-10r18914: this bug fix needs to be for just hpux 11.11Andrew Tridgell1-1/+4
(This used to be commit 2cccede13d34011767159c3345bb6da24ed09bd3)
2007-10-10r18913: an attempt to get tdb/ldb working on the HPUX box 'gwen'. This ideaAndrew Tridgell1-0/+1
come from Don McCall. Don may well be able to provide us with a configure test in the future which does this in a nicer way, I just want to see if it works now. (This used to be commit f7f548a1c79a78a1b15e96732994135cba94aa3d)
2007-10-10r18911: add a JSON encoderDerrell Lipman1-0/+213
(This used to be commit 774e87974951a03663a1efd15dfde3d012bcf526)
2007-10-10r18910: Change ldb_msg_add_string() to not actually add an attribute if theAndrew Tridgell1-0/+5
string is zero length. This allows callers to not have to worry about creating an invalid ldap attribute. See extensive discussion on samba-technical list :-) (This used to be commit 7a1db8c2a432b9ab59b29ee1bfce6c8fe8e981a2)
2007-10-10r18909: use newer functions that were introduced after this code was madeSimo Sorce1-7/+2
(This used to be commit 3ce1796eb4cca7fd78366ee540a998a1ca377866)
2007-10-10r18908: Store the schema structure into an opaque pointer so that it can be ↵Simo Sorce1-6/+14
reused by multiple connections (This used to be commit ca8827d8f9a9f6ec60afed29b0b85f491d725d1c)
2007-10-10r18903: merge from samba3:Stefan Metzmacher4-4/+6
define HAVE_LDB_LDAP and HAVE_LDB_SQLITE3 metze (This used to be commit 4e2f5ba490d717283ab089d90ebd5c108a8c53b0)
2007-10-10r18894: Merge const fixes from 3_0Volker Lendecke1-5/+5
(This used to be commit 4ce447223cc71b8b2a31c2da3b0afd8e3fcf0c01)
2007-10-10r18890: sort the flags out and remove unknown as there's stillRafal Szczesniak1-6/+4
quite a few of them (not sure if these are used actually). rafal (This used to be commit 1622d4608bc738b73d6f51c758828f96602b3e59)
2007-10-10r18887: add flags found while on the plane.Rafal Szczesniak1-14/+19
rafal (This used to be commit ec59441977205af9a38926b3d432ec0de6379573)
2007-10-10r18885: Fix bug in our IDL pointed out by RonnieSimo Sorce1-2/+2
(This used to be commit 77e8d96bc488a574b0b728d5b8facd51e69822d5)
2007-10-10r18884: client_test.exp getting changed to client_tests.exp was an ↵Brad Henry1-1/+1
intentional error that I inserted for testing earlier and forgot to take out before commiting. The other error was a real bug. :) (This used to be commit 0269f3101555daaa4b2c4080ab24c1ca75cdc2f6)
2007-10-10r18883: Start including test_functions.sh.Brad Henry1-113/+70
test_win.sh has been rewritten to use the testit() function from test_functions.sh. (This used to be commit d0ed99859f7cbed599abf7083e7fb20bd58b1556)
2007-10-10r18882: make style consistentSimo Sorce1-1/+3
(This used to be commit a141ee9473ae46c63dea247456f156b40f766d42)
2007-10-10r18881: remove wrong check and statement.Simo Sorce1-5/+0
to manipulate rootDSE we use ldb_dn_new() as base and that has 0 elements. (This used to be commit 3e131177dae3536c07632fe09e7ebe877bcd9332)
2007-10-10r18880: JSON-RPC work in progressDerrell Lipman8-26/+1152
(This used to be commit 34bffbaebf50c2a75c91285d5ec82e8f377981cc)
2007-10-10r18873: Use WERR_errors table rather than DOS_errors table for wireshark.Jelmer Vernooij1-2/+2
(This used to be commit 5a2e103ea0b31e80e4b2c2237b5fd9940820cec6)
2007-10-10r18872: when converting IDL to use out,ref pointers, you also must fix anyAndrew Tridgell1-1/+1
length_is() statements. (This used to be commit 2f9a1f1d62351e20274227c3068a2e8211148535)
2007-10-10r18862: as andrew pointed our on irc, if we are going to define _TRUE, thenAndrew Tridgell1-2/+2
better make it be true :) (This used to be commit 4bda5ed3f9fda437c3f9b7e8ce83853999601c74)
2007-10-10r18861: merge from samba3Andrew Tridgell1-0/+11
(This used to be commit c5e2fcdc10a0869264bb1c6a13ada5f12e1b655f)
2007-10-10r18860: fixed some code that bitrotted while we didn't have the safety checkAndrew Tridgell3-8/+8
with NTSTATUS, WERROR etc. (This used to be commit fddb85bfa8bb5df7812d226a7a6f6fb41ccf107f)
2007-10-10r18859: finally worked out what is going wrong with immediate structures. TheAndrew Tridgell1-0/+27
problem is that the AC_PROG_CC_C99 macro ends up selecting either -std=std99 or -std=gnu99 for gcc. Ironically enough, that breaks constant structure initialisers! So, simplest solution is to not try that configure test if we know we are using gcc (This used to be commit 331435daf3275acaf282c1032c6e9f7dc3e685bb)
2007-10-10r18851: Some C++ warningsVolker Lendecke1-8/+8
(This used to be commit 69f05f288ce7a8508760db861d0910495bd1d578)