Commit Graph

14569 Commits

Author SHA1 Message Date
Pierangelo Masarati
6a293c65b3 line up network-timeout with back-meta 2006-04-07 09:08:37 +00:00
Howard Chu
c1019fbfbd More for ITS#4323 fix 2006-04-07 04:17:44 +00:00
Howard Chu
4c754ec249 Fix typo 2006-04-07 03:08:43 +00:00
Howard Chu
c6d0edc3b7 Use TLS context stuff in syncrepl 2006-04-07 02:57:39 +00:00
Howard Chu
571ac24b33 Fix destruct sequencing 2006-04-07 02:41:58 +00:00
Howard Chu
056b193ed7 Don't implicitly inherit the default SSL_CTX, tls.c:alloc_handle will
set it explicitly and handle its refcount.
2006-04-07 01:52:32 +00:00
Howard Chu
7a5d12cc51 Add TLS options to bindconf struct 2006-04-07 01:49:19 +00:00
Pierangelo Masarati
8c1b8d3f7b actually, if a connection is already in the AVL tree, use it if not binding; otherwise use a tainted one; taint connections that must be freed when refcnt goes to zero 2006-04-07 01:28:56 +00:00
Howard Chu
76bf5d4c76 Use new TLS SSL_CTX support 2006-04-07 01:25:46 +00:00
Howard Chu
9693c800bf Free/decrement SSL_CTX refcount when (re)setting it 2006-04-07 01:15:56 +00:00
Pierangelo Masarati
301394c28d allow a build tree different from the source tree (ITS#4480) 2006-04-07 01:15:07 +00:00
Howard Chu
7709d4d89e Bump SSL_CTX refcount whenever it gets retrieved 2006-04-07 01:13:31 +00:00
Howard Chu
d18277eac9 ITS#4422, #4475
Move TLS options into struct ldapoptions.
  Added ldap_int_tls_destroy()
  Added LDAP_OPT_X_TLS_NEWCTX to generate new SSL_CTX
2006-04-07 00:52:38 +00:00
Howard Chu
51d5db03ff ITS#4422, rearrange options to allow passing NULL args to TLS/Cyrus 2006-04-07 00:48:41 +00:00
Pierangelo Masarati
0f1023b6c7 pedantic cleanup 2006-04-06 23:55:09 +00:00
Pierangelo Masarati
267486bfad fix ITS#4479 2006-04-06 23:48:55 +00:00
Kurt Zeilenga
8fb8b17c04 Code which doesn't hanlding allocation failure should use
ch_ routines, not SLAP_ macros
2006-04-06 21:44:05 +00:00
Kurt Zeilenga
1b9d68d6ba cleanup 2006-04-06 21:29:38 +00:00
Pierangelo Masarati
b03d8abcde don't leak if realloc fails (ITS#4477) 2006-04-06 19:13:41 +00:00
Pierangelo Masarati
c6ab0790d1 don't leak if realloc fails (ITS#4477); s/new/newptr/ 2006-04-06 19:07:48 +00:00
Kurt Zeilenga
721e1c87a8 Have slapd-bind reuse LDAP session (no init). 2006-04-06 19:01:08 +00:00
Hallvard Furuseth
a070e6c05f Avoid sprintf buffer overrun if huge hostname 2006-04-06 09:28:19 +00:00
Howard Chu
daec2febfb ITS#4476 revert connection_write processing back into main thread 2006-04-06 04:34:37 +00:00
Pierangelo Masarati
8395383d2c fix ITS#4435 2006-04-06 01:27:16 +00:00
Hallvard Furuseth
02cba98c0b Protect from sprintf buffer overrun in ldapsearch -f file "(cn=%100000s)" 2006-04-05 20:48:15 +00:00
Hallvard Furuseth
b70d6dd8e6 LDAP_RDWR_DEBUG update:
- Avoid buffer overrun when too many readers.
- When remoinv reader ID, move one ID instead of all later IDs in array.
- assert(correct lock owner) before unlocking a write lock.
2006-04-05 20:08:34 +00:00
Hallvard Furuseth
ca4a4456e8 LDAP_THREAD_DEBUG revamping, cleanup, bug fixes.
Expand usage struct with magic numbers and state info.
Add options "noreinit", "nosync", "nomem", "threadID".  Remove "dupinit".
Detect address changes.  Turn off thread ID tracking by default.
2006-04-05 19:52:07 +00:00
Kurt Zeilenga
3462baf8bc Use standard compliant filter strings 2006-04-05 18:57:46 +00:00
Pierangelo Masarati
7420472b42 warn in case of error 2006-04-05 18:20:00 +00:00
Pierangelo Masarati
ffc69eadde add missing "break" to previous commit 2006-04-05 18:15:30 +00:00
Hallvard Furuseth
b4586b2f84 Fix for the previous ITS#4323 fix: for(expr;...) -> for(i=expr;...) 2006-04-05 14:15:34 +00:00
Howard Chu
c76eb501ab Revert cache-purge task, go back to purging inline 2006-04-05 07:17:11 +00:00
Howard Chu
0ad714d61e Additional fix for ITS#4323, must propagate search result controls 2006-04-05 07:03:25 +00:00
Kurt Zeilenga
45d0479d37 Reverse last commit. Wrong tree. 2006-04-05 00:40:53 +00:00
Kurt Zeilenga
31d64d4642 No LogTest in re23 2006-04-05 00:39:46 +00:00
Howard Chu
2bf9afd828 ITS#4323 allow lower overlays to work on master database 2006-04-05 00:27:16 +00:00
Hallvard Furuseth
eca5f7bf2d ITS#4472: Make t_rwlock a lock pointer instead of a lock, and malloc/free it.
That way, realloc(qm->templates...) won't move the ldap_pvt_thread_rdwr_t data.
2006-04-04 23:12:58 +00:00
Howard Chu
04dae5a020 ITS#4442 just log refint failures, client only cares about main request 2006-04-04 22:16:24 +00:00
Pierangelo Masarati
582ee9675f silence warning 2006-04-04 21:07:53 +00:00
Pierangelo Masarati
674c98fa7d honor stop on error 2006-04-04 19:48:22 +00:00
Pierangelo Masarati
e21f57cf05 complete previous commit (protect binds) 2006-04-04 18:45:59 +00:00
Pierangelo Masarati
6d357288d6 allow time/size limit errors as soon as some data is returned; don't divide by zero if no entries were found 2006-04-04 17:55:27 +00:00
Pierangelo Masarati
543396c804 make sure only on bind at a time occurs; release resources only when appropriate 2006-04-04 17:41:50 +00:00
Pierangelo Masarati
2e9f2995ec only wait if no entry was returned 2006-04-04 17:41:13 +00:00
Howard Chu
3bfe4a7f77 plug leaks 2006-04-04 10:58:08 +00:00
Howard Chu
abb01fa437 cleanup 2006-04-04 10:47:42 +00:00
Howard Chu
fc61aab12b plug leak 2006-04-04 10:31:01 +00:00
Howard Chu
6a50826676 plug leaks 2006-04-04 10:27:28 +00:00
Howard Chu
548270776d Plug leak 2006-04-04 10:12:55 +00:00
Howard Chu
272211fbcd Same as ITS#4464, don't call cache_return_entry with NULL entry 2006-04-04 09:56:34 +00:00