14 years agor19166: better bug and reporting and proper authentication tracking
Derrell Lipman [Sat, 7 Oct 2006 20:33:09 +0000 (20:33 +0000)]
r19166: better bug and reporting and proper authentication tracking

14 years agor19165: handle errors better for jsonrpc. generate an error object whenever possible
Derrell Lipman [Sat, 7 Oct 2006 20:31:27 +0000 (20:31 +0000)]
r19165: handle errors better for jsonrpc.  generate an error object whenever possible

14 years agor19162: Merge WERR code from Samba 3.
Günther Deschner [Sat, 7 Oct 2006 05:28:14 +0000 (05:28 +0000)]
r19162: Merge WERR code from Samba 3.


14 years agor19153: Simple consistency check for enumprinters level 2 as well
Volker Lendecke [Fri, 6 Oct 2006 19:02:02 +0000 (19:02 +0000)]
r19153: Simple consistency check for enumprinters level 2 as well

14 years agor19152: A simple consistency check for enumprinters
Volker Lendecke [Fri, 6 Oct 2006 18:50:48 +0000 (18:50 +0000)]
r19152: A simple consistency check for enumprinters

14 years agor19150: add a usefull tool to test the gdb_backtrace script
Stefan Metzmacher [Fri, 6 Oct 2006 18:04:48 +0000 (18:04 +0000)]
r19150: add a usefull tool to test the gdb_backtrace script

just compile it with
cc -g -o gdb_backtrace_test gdb_backtrace_test.c

and run it in the same directory where your gdb_backtrace script is.


14 years agor19149: ladebug needs to have the stoponattach flag set
Stefan Metzmacher [Fri, 6 Oct 2006 17:55:17 +0000 (17:55 +0000)]
r19149: ladebug needs to have the stoponattach flag set
to generate the current backtrace without waiting for a signal...


14 years agor19146: merge from samba3:
Stefan Metzmacher [Fri, 6 Oct 2006 16:17:19 +0000 (16:17 +0000)]
r19146: merge from samba3:

talloc_reference() can fail


14 years agor19142: ensure no race conditions during installation by having same name in source...
Derrell Lipman [Fri, 6 Oct 2006 15:50:26 +0000 (15:50 +0000)]
r19142: ensure no race conditions during installation by having same name in source and swat directory; install new apps and services

14 years agor19141: add a reasonable subset of the qooxdoo runtime environment, and example appli...
Derrell Lipman [Fri, 6 Oct 2006 15:40:18 +0000 (15:40 +0000)]
r19141: add a reasonable subset of the qooxdoo runtime environment, and example applications

14 years agor19140: allow own authentication handling by jsonrpc apps
Derrell Lipman [Fri, 6 Oct 2006 15:38:43 +0000 (15:38 +0000)]
r19140: allow own authentication handling by jsonrpc apps

14 years agor19139: typos; consitent ordering of operations
Derrell Lipman [Fri, 6 Oct 2006 15:37:39 +0000 (15:37 +0000)]
r19139: typos; consitent ordering of operations

14 years agor19138: add authentication capability; fix typos
Derrell Lipman [Fri, 6 Oct 2006 15:36:07 +0000 (15:36 +0000)]
r19138: add authentication capability; fix typos

14 years agor19136: No conflict, after "svn up" it applied correctly", merge 19126 from 3_0
Volker Lendecke [Fri, 6 Oct 2006 15:16:18 +0000 (15:16 +0000)]
r19136: No conflict, after "svn up" it applied correctly", merge 19126 from 3_0

14 years agor19134: Merge the second set of C++ warning fixes from Samba3. I'll leave r19132 to
Volker Lendecke [Fri, 6 Oct 2006 15:03:41 +0000 (15:03 +0000)]
r19134: Merge the second set of C++ warning fixes from Samba3. I'll leave r19132 to
metze to merge until the questions have been answered.


14 years agor19131: merge from samba3:
Stefan Metzmacher [Fri, 6 Oct 2006 14:04:05 +0000 (14:04 +0000)]
r19131: merge from samba3:

fix the logic for the AC_PROG_CC_C99 test


14 years agor19129: Add comment to clarify behavior
Simo Sorce [Fri, 6 Oct 2006 13:47:52 +0000 (13:47 +0000)]
r19129: Add comment to clarify behavior

14 years agor19128: - make the gdb_backtrace script more portable
Stefan Metzmacher [Fri, 6 Oct 2006 13:23:42 +0000 (13:23 +0000)]
r19128: - make the gdb_backtrace script more portable
- try /proc/${PID}/exe first
- fallback to the binary given on the commandline
- fallback searching the binary with 'which' from the
  commandline argument
- use 'ladebug' debugger on Tru64


14 years agor19123: - pass LDFLAGS from configure to Makefile
Stefan Metzmacher [Fri, 6 Oct 2006 11:02:59 +0000 (11:02 +0000)]
r19123: - pass LDFLAGS from configure to Makefile
  which should fix the build on Tru64 in the build farm.

Thanks to Bjoern Jacke for finding this!


14 years agor19122: merge from samba3:
Stefan Metzmacher [Fri, 6 Oct 2006 11:01:05 +0000 (11:01 +0000)]
r19122: merge from samba3:

sync system/iconv.h


14 years agor19120: silence a warning about a test function
Andrew Tridgell [Fri, 6 Oct 2006 10:17:52 +0000 (10:17 +0000)]
r19120: silence a warning about a test function

14 years agor19118: get rid of a bunch of bool misuse warnings
Andrew Tridgell [Fri, 6 Oct 2006 09:26:29 +0000 (09:26 +0000)]
r19118: get rid of a bunch of bool misuse warnings

14 years agor19117: I've changed the checker wrapper on snab to allow for a
Andrew Tridgell [Fri, 6 Oct 2006 07:45:31 +0000 (07:45 +0000)]
r19117: I've changed the checker wrapper on snab to allow for a
.checker_innocent file, similar to our .valgrind_suppressions file.

14 years agor19116: fixed a checker warning
Andrew Tridgell [Fri, 6 Oct 2006 07:44:36 +0000 (07:44 +0000)]
r19116: fixed a checker warning

14 years agor19115: Add the mapping required between Samba4's AD schema OIDs and what
Andrew Bartlett [Fri, 6 Oct 2006 07:09:22 +0000 (07:09 +0000)]
r19115: Add the mapping required between Samba4's AD schema OIDs and what
Fedora DS can support.

Andrew Bartlett

14 years agor19114: fixed another checker warning and a possible error on allocation
Andrew Tridgell [Fri, 6 Oct 2006 06:52:30 +0000 (06:52 +0000)]
r19114: fixed another checker warning and a possible error on allocation

14 years agor19113: fixed another checker warning.
Andrew Tridgell [Fri, 6 Oct 2006 06:42:05 +0000 (06:42 +0000)]
r19113: fixed another checker warning.

14 years agor19112: fixed a checker warning.
Andrew Tridgell [Fri, 6 Oct 2006 06:38:26 +0000 (06:38 +0000)]
r19112: fixed a checker warning.

In case you haven't noticed, lots of our packages now run the IBM
checker in the build farm on 'snab'

14 years agor19111: fixed basic web server operation
Andrew Tridgell [Fri, 6 Oct 2006 05:36:29 +0000 (05:36 +0000)]
r19111: fixed basic web server operation

14 years agor19098: with
Stefan Metzmacher [Thu, 5 Oct 2006 11:46:31 +0000 (11:46 +0000)]
r19098: with

make test will use SOCKET_WRAPPER_PCAP_FILE support
and will run each test with a different file

the capture will be removed unless the test fails.

the TEST PCAP: line will show the location of the file


14 years agor19097: add support for SUCKET_WRAPPER_PCAP_FILE
Stefan Metzmacher [Thu, 5 Oct 2006 11:39:59 +0000 (11:39 +0000)]
r19097: add support for SUCKET_WRAPPER_PCAP_FILE

we may need to rework the code to use something like PUSH_U32()
into a data blob instead of the structs for the header layout.


14 years agor19092: active replica records cannot expire!
Stefan Metzmacher [Thu, 5 Oct 2006 07:51:51 +0000 (07:51 +0000)]
r19092: active replica records cannot expire!

we handle that in the scavenging code, but were missing that
in the wins server name query code...


14 years agor19091: merge from samba3:
Stefan Metzmacher [Thu, 5 Oct 2006 07:47:56 +0000 (07:47 +0000)]
r19091: merge from samba3:

remove some -Wl,-rpath combinations with /usr/lib


14 years agor19084: make socket_wrapper more portable (standalone) and not use uintX_t
Stefan Metzmacher [Thu, 5 Oct 2006 07:03:41 +0000 (07:03 +0000)]
r19084: make socket_wrapper more portable (standalone) and not use uintX_t


14 years agor19082: add the source vs. destination address and tcp logic
Stefan Metzmacher [Thu, 5 Oct 2006 06:27:41 +0000 (06:27 +0000)]
r19082: add the source vs. destination address and tcp logic


14 years agor19079: talloc_array can fail
Volker Lendecke [Wed, 4 Oct 2006 20:59:06 +0000 (20:59 +0000)]
r19079: talloc_array can fail

14 years agor19076: ldb_dn_copy_partial can fail
Volker Lendecke [Wed, 4 Oct 2006 20:22:08 +0000 (20:22 +0000)]
r19076: ldb_dn_copy_partial can fail

14 years agor19075: Fix a potential NULL dereference
Volker Lendecke [Wed, 4 Oct 2006 20:15:23 +0000 (20:15 +0000)]
r19075: Fix a potential NULL dereference

14 years agor19074: Merge from 3_0: Fix a potential NULL dereference.
Volker Lendecke [Wed, 4 Oct 2006 19:45:34 +0000 (19:45 +0000)]
r19074: Merge from 3_0: Fix a potential NULL dereference.


14 years agor19069: The sequence number is a 64 bit unsigned integer
Simo Sorce [Wed, 4 Oct 2006 19:08:36 +0000 (19:08 +0000)]
r19069: The sequence number is a 64 bit unsigned integer
Well spotted Volker

14 years agor19068: Fix a potential NULL dereference
Volker Lendecke [Wed, 4 Oct 2006 19:03:29 +0000 (19:03 +0000)]
r19068: Fix a potential NULL dereference

14 years agor19063: merge from samba3
Stefan Metzmacher [Wed, 4 Oct 2006 12:33:36 +0000 (12:33 +0000)]
r19063: merge from samba3

needed for dl* errors in iconv_open()


14 years agor19059: allow dash in URLs
Derrell Lipman [Tue, 3 Oct 2006 18:07:46 +0000 (18:07 +0000)]
r19059: allow dash in URLs

14 years agor19055: JSON-RPC is working! It passes the small qooxdoo JSON-RPC test suite
Derrell Lipman [Tue, 3 Oct 2006 02:38:08 +0000 (02:38 +0000)]
r19055: JSON-RPC is working!  It passes the small qooxdoo JSON-RPC test suite

This check in covers the remaining fixes and enhancements to complete the
server.  In a subsequent check-in, I need to learn about the new qooxdoo
packaging mechanism.  Although it'd be nice to have the whole source tree,
that's huge and needn't be in samba svn.  It's now supposed to be very easy to
package qooxdoo, so I'll figure out how that's done and then check in the
relevant small number of files.

Steps after that involve generating the stubs to allow various ejs functions
to be called via JSON-RPC.


14 years agor19053: match the flag values with the actual equivalents in samr.idl
Rafal Szczesniak [Mon, 2 Oct 2006 22:18:36 +0000 (22:18 +0000)]
r19053: match the flag values with the actual equivalents in samr.idl


14 years agor19052: clarify what actually does what
Derrell Lipman [Mon, 2 Oct 2006 20:44:19 +0000 (20:44 +0000)]
r19052: clarify what actually does what

14 years agor19051: JSON-RPC server work-in-progress. It's almost working.
Derrell Lipman [Mon, 2 Oct 2006 20:39:31 +0000 (20:39 +0000)]
r19051: JSON-RPC server work-in-progress.  It's almost working.

14 years agor19049: Samba 3 client fixes
Gerald Carter [Mon, 2 Oct 2006 16:19:05 +0000 (16:19 +0000)]
r19049: Samba 3 client fixes

* Make sure to check for NULL pointers when dealing with
  [out,unique] before assigning valoues
* Detect RPC faults and return immediately

14 years agor19048: ensure our memory gets cleaned up even upon exception
Derrell Lipman [Mon, 2 Oct 2006 15:23:03 +0000 (15:23 +0000)]
r19048: ensure our memory gets cleaned up even upon exception

14 years agor19036: add LOCAL-REPLACE and reorder the testlist
Stefan Metzmacher [Mon, 2 Oct 2006 10:45:00 +0000 (10:45 +0000)]
r19036: add LOCAL-REPLACE and reorder the testlist


14 years agor19035: add LOCAL-REPLACE test
Stefan Metzmacher [Mon, 2 Oct 2006 10:38:24 +0000 (10:38 +0000)]
r19035: add LOCAL-REPLACE test


14 years agor19033: prepare to use the libreplace testsuite as LOCAL-REPLACE in smbtorture
Stefan Metzmacher [Mon, 2 Oct 2006 10:35:40 +0000 (10:35 +0000)]
r19033: prepare to use the libreplace testsuite as LOCAL-REPLACE in smbtorture


14 years agor19032: let the getdents replacement fail and see what the build-farm says.
Stefan Metzmacher [Mon, 2 Oct 2006 09:26:05 +0000 (09:26 +0000)]
r19032: let the getdents replacement fail and see what the build-farm says.


14 years agor19029: test if opendir() on a file sets errno = ENOTDIR
Stefan Metzmacher [Mon, 2 Oct 2006 08:56:35 +0000 (08:56 +0000)]
r19029: test if opendir() on a file sets errno = ENOTDIR


14 years agor19027: Modify RPC-SAMBA3-SPOOLSS so that it can be activated in the build farm
Volker Lendecke [Mon, 2 Oct 2006 08:35:19 +0000 (08:35 +0000)]
r19027: Modify RPC-SAMBA3-SPOOLSS so that it can be activated in the build farm

14 years agor19025: - remove read-only and policy dependent fields and flags
Rafal Szczesniak [Mon, 2 Oct 2006 06:00:14 +0000 (06:00 +0000)]
r19025: - remove read-only and policy dependent fields and flags
- do not stop NET-API-USERMODIFY test when a single field fails
- add account flags to the fields tested
- separate "cleanup" step from closing domain at the end


14 years agor19024: remove read-only and policy dependent fields and flags
Rafal Szczesniak [Mon, 2 Oct 2006 05:53:45 +0000 (05:53 +0000)]
r19024: remove read-only and policy dependent fields and flags
from user modify routines.


14 years agor19023: lets see what the build-farm says about trying the
Stefan Metzmacher [Mon, 2 Oct 2006 05:52:42 +0000 (05:52 +0000)]
r19023: lets see what the build-farm says about trying the
getdirentries() based readdir() replacement first.


14 years agor19022: eliminate a warning
Derrell Lipman [Mon, 2 Oct 2006 02:39:05 +0000 (02:39 +0000)]
r19022: eliminate a warning

14 years agor19021: match the prototype for a trigger function
Derrell Lipman [Mon, 2 Oct 2006 02:38:43 +0000 (02:38 +0000)]
r19021: match the prototype for a trigger function

14 years agor19020: move jsonrpc_include to a more appropriate location in the file, and call...
Derrell Lipman [Mon, 2 Oct 2006 02:38:02 +0000 (02:38 +0000)]
r19020: move jsonrpc_include to a more appropriate location in the file, and call the setup function for parsing literals.

14 years agor19019: Add a parser for javascript literals. This is implemented for parsing JSON
Derrell Lipman [Mon, 2 Oct 2006 02:36:27 +0000 (02:36 +0000)]
r19019: Add a parser for javascript literals.  This is implemented for parsing JSON
strings, but is generally useful for creating ejs variables from object or
array literals, which aren't supported in the old version of ejs that we're

This parser is implemented in C, but is callable from an ejs script via

14 years agor19017: Merge from Samba3:
Volker Lendecke [Sun, 1 Oct 2006 20:45:46 +0000 (20:45 +0000)]
r19017: Merge from Samba3:

Attempt to fix the BSD RAW-CHKPATH failures: In Samba3 we rely on opendir
returning ENOTDIR when the file opened is not a directory. Can we merge this
back into Samba4?

This includes the tabs->spaces fix requested by Stefan.


14 years agor19009: ensure that data values from ldap libs are null terminated, to allow
Andrew Tridgell [Sat, 30 Sep 2006 07:54:20 +0000 (07:54 +0000)]
r19009: ensure that data values from ldap libs are null terminated, to allow
ldb_msg_find_attr_as_string() to work correctly.

Thanks to Jim Myers for spotting this!

14 years agor18999: merge from samba3:
Stefan Metzmacher [Fri, 29 Sep 2006 12:38:51 +0000 (12:38 +0000)]
r18999: merge from samba3:

- make more usage of PRINTF_ATTRIBUTE()
- vsyslog takes a 'const char *format'


14 years agor18997: we have a replacement for vsyslog in libreplace so libroken doesn't need...
Stefan Metzmacher [Fri, 29 Sep 2006 12:24:23 +0000 (12:24 +0000)]
r18997: we have a replacement for vsyslog in libreplace so libroken doesn't need to care about it


14 years agor18995: - fix bug 4078
Stefan Metzmacher [Fri, 29 Sep 2006 10:50:12 +0000 (10:50 +0000)]
r18995: - fix bug 4078
- talloc_free(talloc_autofree_context()); should not result
  in a SIGABORT on exit
- add a test for this, but this test can also pass in the standalone build
  and samba3, as samba4 uses talloc_autofree_context()


14 years agor18994: - torture_register_suite should be public
Stefan Metzmacher [Fri, 29 Sep 2006 10:20:51 +0000 (10:20 +0000)]
r18994: - torture_register_suite should be public
- some whitespace cleanups


14 years agor18992: prepare SOCKET_WRAPPER_PCAP_FILE support
Stefan Metzmacher [Fri, 29 Sep 2006 09:31:20 +0000 (09:31 +0000)]
r18992: prepare SOCKET_WRAPPER_PCAP_FILE support


14 years agor18991: Fix bug 4133 -- Thanks to Olaf Flebbe
Volker Lendecke [Fri, 29 Sep 2006 09:12:55 +0000 (09:12 +0000)]
r18991: Fix bug 4133 -- Thanks to Olaf Flebbe

14 years agor18989: Fixes found by these two LDAP testsuites:
Andrew Bartlett [Fri, 29 Sep 2006 04:45:15 +0000 (04:45 +0000)]
r18989: Fixes found by these two LDAP testsuites:
- http://www.ee.oulu.fi/research/ouspg/protos/testing/c06/ldapv3/
- http://gleg.net/protover_ldap_sample.shtml

Also fixes found by a subsequent audit of the code for similar issues.

14 years agor18979: With these extra indexes (also added for the normal case) and a
Andrew Bartlett [Thu, 28 Sep 2006 17:08:17 +0000 (17:08 +0000)]
r18979: With these extra indexes (also added for the normal case) and a
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

14 years agor18978: Fix bug found by:
Andrew Bartlett [Thu, 28 Sep 2006 17:06:38 +0000 (17:06 +0000)]
r18978: Fix bug found by:

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

14 years agor18977: Seperate these asserts, so we know which fired.
Andrew Bartlett [Thu, 28 Sep 2006 16:58:28 +0000 (16:58 +0000)]
r18977: Seperate these asserts, so we know which fired.

Andrew Bartlett

14 years agor18975: follow the advice from "William Jojo" <jojowil@hvcc.edu>
Stefan Metzmacher [Thu, 28 Sep 2006 15:12:44 +0000 (15:12 +0000)]
r18975: follow the advice from "William Jojo" <jojowil@hvcc.edu>
and pass -bbigtoc also to the linker when building shared libs


14 years agor18974: we don't have the configure check that sets
Stefan Metzmacher [Thu, 28 Sep 2006 15:10:11 +0000 (15:10 +0000)]
r18974: we don't have the configure check that sets
and as -Wl,--allow-shlib-undefined is supported on SuSE 7.3
I don't see a problem is always pass it explicit


14 years agor18972: we don't need this now all builds in the build farm are limited to 150
Andrew Tridgell [Thu, 28 Sep 2006 06:49:14 +0000 (06:49 +0000)]
r18972: we don't need this now all builds in the build farm are limited to 150
file descriptors. We'll pretty quickly find leaks :)

14 years agor18971: avoid strndup is a few places. Fixes a minor memory leak, and should
Andrew Tridgell [Thu, 28 Sep 2006 06:44:47 +0000 (06:44 +0000)]
r18971: avoid strndup is a few places. Fixes a minor memory leak, and should
fix RPC-LSA on AIX.

14 years agor18970: avoid strndup and strnlen on AIX. They are quite broken. See
Andrew Tridgell [Thu, 28 Sep 2006 06:43:27 +0000 (06:43 +0000)]
r18970: avoid strndup and strnlen on AIX. They are quite broken. See

14 years agor18968: EWOULDBLOCK should also be mapped to STATUS_MORE_ENTRIES
Andrew Tridgell [Thu, 28 Sep 2006 01:17:16 +0000 (01:17 +0000)]
r18968: EWOULDBLOCK should also be mapped to STATUS_MORE_ENTRIES

14 years agor18965: fixed the bug with RPC-NETLOGON and solaris sparc machines. This bug
Andrew Tridgell [Wed, 27 Sep 2006 23:24:36 +0000 (23:24 +0000)]
r18965: fixed the bug with RPC-NETLOGON and solaris sparc machines. This bug
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.

14 years agor18960: fix caching of some configure tests AC_CACHE_CHECK()
Stefan Metzmacher [Wed, 27 Sep 2006 14:47:00 +0000 (14:47 +0000)]
r18960: fix caching of some configure tests AC_CACHE_CHECK()
*needs* '_cv_' in the shell var to do caching...


14 years agor18957: we need to include unistd.h in system/network.h because
Stefan Metzmacher [Wed, 27 Sep 2006 14:16:32 +0000 (14:16 +0000)]
r18957: we need to include unistd.h in system/network.h because
it contains the ioctl() prototype on some systems
and we need to make sure it gets included before socket_wrapper
defines ioctl swrap_ioctl


14 years agor18948: fix the build on solaris, the ioctl() prototype
Stefan Metzmacher [Wed, 27 Sep 2006 11:15:03 +0000 (11:15 +0000)]
r18948: fix the build on solaris, the ioctl() prototype
in unistd.h was included after libreplace defined ioctl swrap_ioctl


14 years agor18947: overload listen() and ioctl() in socket_wrapper
Stefan Metzmacher [Wed, 27 Sep 2006 10:13:45 +0000 (10:13 +0000)]
r18947: overload listen() and ioctl() in socket_wrapper


14 years agor18946: fix compiler warning
Stefan Metzmacher [Wed, 27 Sep 2006 06:48:24 +0000 (06:48 +0000)]
r18946: fix compiler warning


14 years agor18945: fix compiler warnings and end-of-non-void function bugs
Stefan Metzmacher [Wed, 27 Sep 2006 06:42:19 +0000 (06:42 +0000)]
r18945: fix compiler warnings and end-of-non-void function bugs


14 years agor18944: talloc_reference() can fail...
Stefan Metzmacher [Wed, 27 Sep 2006 06:39:56 +0000 (06:39 +0000)]
r18944: talloc_reference() can fail...
(remove compiler warning)


14 years agor18943: Samba4 wants its databases private to root. I know some of them could
Andrew Tridgell [Wed, 27 Sep 2006 06:02:07 +0000 (06:02 +0000)]
r18943: Samba4 wants its databases private to root. I know some of them could
potentially be public, but we'd need to think about them

14 years agor18942: add a ldb_set_create_perms() function in ldb. I didn't call it
Andrew Tridgell [Wed, 27 Sep 2006 05:57:41 +0000 (05:57 +0000)]
r18942: add a ldb_set_create_perms() function in ldb. I didn't call it
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)

14 years agor18939: don't rely on the umask being right in ldb creation. Both Samba3 and
Andrew Tridgell [Wed, 27 Sep 2006 03:34:50 +0000 (03:34 +0000)]
r18939: don't rely on the umask being right in ldb creation. Both Samba3 and
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

14 years agor18936: hopefully fix the test for negative enum values. When a compiler
Andrew Tridgell [Wed, 27 Sep 2006 02:00:13 +0000 (02:00 +0000)]
r18936: hopefully fix the test for negative enum values. When a compiler
truncates the enums, the test was passing.

14 years agor18933: Add helpful emacs marker
Andrew Bartlett [Tue, 26 Sep 2006 22:14:15 +0000 (22:14 +0000)]
r18933: Add helpful emacs marker

14 years agor18925: Add current snapshot of the ejs-2.0 code. Tridge, will you be incorporating...
Derrell Lipman [Tue, 26 Sep 2006 16:58:27 +0000 (16:58 +0000)]
r18925: Add current snapshot of the ejs-2.0 code.  Tridge, will you be incorporating this?

14 years agor18918: - bail out with unsupported option to socket()
Stefan Metzmacher [Tue, 26 Sep 2006 13:15:31 +0000 (13:15 +0000)]
r18918: - bail out with unsupported option to socket()
- don't reuse portnumbers in the autobind code
- use if (!...) return; logic instead of if (...) { do everything } return
  for swrap_close


14 years agor18917: having 255 virtual interfaces available in socket wrapper means we
Andrew Tridgell [Tue, 26 Sep 2006 11:31:14 +0000 (11:31 +0000)]
r18917: having 255 virtual interfaces available in socket wrapper means we
stat() 240 files that don't exist on every broadcast. That's a bit

reduce max virtual interfaces to 16

14 years agor18916: fixed the messaging layer on *BSD systems. When a socket was full we
Andrew Tridgell [Tue, 26 Sep 2006 11:27:09 +0000 (11:27 +0000)]
r18916: fixed the messaging layer on *BSD systems. When a socket was full we
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 :)

14 years agor18915: fix usage of popt, do make -W setting the recursive flag on SuSE 10.1
Stefan Metzmacher [Tue, 26 Sep 2006 10:34:40 +0000 (10:34 +0000)]
r18915: fix usage of popt, do make -W setting the recursive flag on SuSE 10.1


14 years agor18914: this bug fix needs to be for just hpux 11.11
Andrew Tridgell [Tue, 26 Sep 2006 05:40:57 +0000 (05:40 +0000)]
r18914: this bug fix needs to be for just hpux 11.11

14 years agor18913: an attempt to get tdb/ldb working on the HPUX box 'gwen'. This idea
Andrew Tridgell [Tue, 26 Sep 2006 03:11:31 +0000 (03:11 +0000)]
r18913: an attempt to get tdb/ldb working on the HPUX box 'gwen'. This idea
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.