jra/samba/.git
16 years agor20280: if the uint8_t array starts with 0xFF, then it's not a OID
Stefan Metzmacher [Wed, 20 Dec 2006 14:44:14 +0000 (14:44 +0000)]
r20280: if the uint8_t array starts with 0xFF, then it's not a OID
and we return the array as hexstring. this is always in the last
array member of the meppings array, and I is always 21 bytes long

w2k in mixed mode:   FF0000000000000000000000000000000000000000
w2k3 in mixed mode:  FF00000002A5DA73B101C43B449028E2F832FE466F
w2k3 in native mode: FF00000001635D94BDE62E2C4C9BAC9D6AFA4F69F8

I assume it's some kind of schema version...

metze
(This used to be commit db16f6382da50167b6cefdaeb8488b00a45fb928)

16 years agor20278: make it possible to use oid strings in the drsuapi instead of uint8_t arrays
Stefan Metzmacher [Wed, 20 Dec 2006 13:05:13 +0000 (13:05 +0000)]
r20278: make it possible to use oid strings in the drsuapi instead of uint8_t arrays
with ber encoded oid's

metze
(This used to be commit bc13b3690166b29df246cc4778b378e9cf2a22d7)

16 years agor20277: make sure the asn1 structure has a welldefined state after a asn1_free()
Stefan Metzmacher [Wed, 20 Dec 2006 12:53:58 +0000 (12:53 +0000)]
r20277: make sure the asn1 structure has a welldefined state after a asn1_free()

metze
(This used to be commit 7e7d1a1da2d64d28915a06c399072de1caf108c4)

16 years agor20276: remove unneeded talloc_strdup()
Stefan Metzmacher [Wed, 20 Dec 2006 12:51:29 +0000 (12:51 +0000)]
r20276: remove unneeded talloc_strdup()

metze
(This used to be commit c4733b20c7c113dc1abfb2219a899b80bbcc2875)

16 years agor20275: we should check for the oid the caller gave us!
Stefan Metzmacher [Wed, 20 Dec 2006 12:50:03 +0000 (12:50 +0000)]
r20275: we should check for the oid the caller gave us!

metze
(This used to be commit 4b9e196288f2deb3594db9ba2dd36d774e774574)

16 years agor20274: add missing return statement and make it more explicit that we return a NULL...
Stefan Metzmacher [Wed, 20 Dec 2006 12:49:11 +0000 (12:49 +0000)]
r20274: add missing return statement and make it more explicit that we return a NULL DATA_BLOB

metze
(This used to be commit 7256481f08b5e860308e73c2b51926b55b1f4c43)

16 years agor20272: Fix for BASE-BENCH-READWRITE from Mathias Dietz <MDIETZ@de.ibm.com>
Volker Lendecke [Wed, 20 Dec 2006 07:46:22 +0000 (07:46 +0000)]
r20272: Fix for BASE-BENCH-READWRITE from Mathias Dietz <MDIETZ@de.ibm.com>
(This used to be commit b053a7df8453527a9c2dfc03f3c35487fe75676f)

16 years agor20259: add function to start a fake tag so that asn1_read_OID_String()
Stefan Metzmacher [Tue, 19 Dec 2006 19:28:49 +0000 (19:28 +0000)]
r20259: add function to start a fake tag so that asn1_read_OID_String()
can work alone

metze
(This used to be commit ea70f6ff07930951d98a952b03963d0ba358fec4)

16 years agor20258: add functions to read and write asn1 encoded OID strings without leading tag
Stefan Metzmacher [Tue, 19 Dec 2006 19:25:49 +0000 (19:25 +0000)]
r20258: add functions to read and write asn1 encoded OID strings without leading tag

metze
(This used to be commit 576d4c54cca844164b90e5d6ec71fe44b59607b7)

16 years agor20251: I found out that the oid-prefix to uint32-id-prefix mapping is transferred
Stefan Metzmacher [Tue, 19 Dec 2006 13:38:42 +0000 (13:38 +0000)]
r20251: I found out that the oid-prefix to uint32-id-prefix mapping is transferred
in replication replies, but I don't know the exact encoding.

for example the oids are transferred as:

2.5.4 => uint8_t v[] = { 0x55, 0x04 };
2.5.5 => uint8_t v[] = { 0x55, 0x05 };
2.5.6 => uint8_t v[] = { 0x55, 0x06 };
2.5.18 => uint8_t v[] = { 0x55, 0x12 };
2.5.20 => uint8_t v[] = { 0x55, 0x14 };
2.5.21 => uint8_t v[] = { 0x55, 0x15 };
1.2.840.113556.1.2 => uint8_t v[] = { 0x2a, 0x86, 0x48, 0x86, 0xf7, 0x14, 0x01, 0x02 };
1.2.840.113556.1.3 => uint8_t v[] = { 0x2a, 0x86, 0x48, 0x86, 0xf7, 0x14, 0x01, 0x03 };
1.2.840.113556.1.4 => uint8_t v[] = { 0x2a, 0x86, 0x48, 0x86, 0xf7, 0x14, 0x01, 0x04 };
1.2.840.113556.1.5 => uint8_t v[] = { 0x2a, 0x86, 0x48, 0x86, 0xf7, 0x14, 0x01, 0x05 };
1.2.840.113556.1.5.7000 => uint8_t v[] = { 0x2a, 0x86, 0x48, 0x86, 0xf7, 0x14, 0x01, 0x05, 0xb6, 0x58 };
1.2.840.113549.1.9 => uint8_t v[] = { 0x2a, 0x86, 0x48, 0x86, 0xf7, 0x0d, 0x01, 0x09 };
2.16.840.1.113730.3 => uint8_t v[] = { 0x60, 0x86, 0x48, 0x01, 0x86, 0xf8, 0x42, 0x03 };
2.16.840.1.113730.3.1 => uint8_t v[] = { 0x60, 0x86, 0x48, 0x01, 0x86, 0xf8, 0x42, 0x03, 0x01 };
2.16.840.1.113730.3.2 => uint8_t v[] = { 0x60, 0x86, 0x48, 0x01, 0x86, 0xf8, 0x42, 0x03, 0x02 };
0.9.2342.19200300.100.1 => uint8_t v[] = { 0x09, 0x92, 0x26, 0x89, 0x93, 0xf2, 0x2c, 0x64, 0x01 };
0.9.2342.19200300.100.4 => uint8_t v[] = { 0x09, 0x92, 0x26, 0x89, 0x93, 0xf2, 0x2c, 0x64, 0x04 };
1.3.6.1.4.1.250.1 => uint8_t v[] = { 0x2b, 0x06, 0x01, 0x04, 0x01, 0x81, 0x7a, 0x01 };
1.3.6.1.4.1.1466.101.119=> uint8_t v[] = { 0x2b, 0x06, 0x01, 0x04, 0x01, 0x8b, 0x3a, 0x65, 0x77 };

if someone knows how the encoding works, please tell me:-)
I assume some ASN.1 encoding...

metze
(This used to be commit aa720a15319392fee5c532959192d0df5bf4c718)

16 years agor20234: metze pointed out that we are re-loading the cache records on each
Andrew Tridgell [Sun, 17 Dec 2006 23:11:46 +0000 (23:11 +0000)]
r20234: metze pointed out that we are re-loading the cache records on each
write. We should only be doing this if another process writes and
changes the seqnum. This avoids the extra cache loads
(This used to be commit 65858ebb68c25a672e9284e8cba9a6675902f1df)

16 years agor20233: add functions to create a dsdb_attribute and dsdb_class from
Stefan Metzmacher [Sun, 17 Dec 2006 21:58:49 +0000 (21:58 +0000)]
r20233: add functions to create a dsdb_attribute and dsdb_class from
a drsuapi_DsReplicaObject

metze
(This used to be commit 9168236ddd3d8ed9daa1e47be1e6b12d2bc348b6)

16 years agor20232: specify attribute name and structure element name always together
Stefan Metzmacher [Sun, 17 Dec 2006 20:19:06 +0000 (20:19 +0000)]
r20232: specify attribute name and structure element name always together
there're a view elements where both values doesn't match

LDAP-SCHEMA now successful loads the schema

metze
(This used to be commit 5d7827f45ec05acaca28c94fd6a5d25f148d011c)

16 years agor20231: - add more oid => uint32 id mappings
Stefan Metzmacher [Sun, 17 Dec 2006 20:11:40 +0000 (20:11 +0000)]
r20231: - add more oid => uint32 id mappings
- handle the case where mappings for 1.2.3.4. and 1.2.3.4.5. prefixes
  exist
- extend test suite for the mappings

metze
(This used to be commit 3af0a9b2c490ed1e4cd0af7faf814823bc60c264)

16 years agor20230: I found more mappings of OID string to uint32 id
Stefan Metzmacher [Sun, 17 Dec 2006 19:36:09 +0000 (19:36 +0000)]
r20230: I found more mappings of OID string to uint32 id

metze
(This used to be commit 3d51ce92c272582e19046337ea6b8b2acd60997d)

16 years agor20226: rename macros with a _LDB at the end, because I'll add macros with _DS
Stefan Metzmacher [Sun, 17 Dec 2006 13:34:56 +0000 (13:34 +0000)]
r20226: rename macros with a _LDB at the end, because I'll add macros with _DS
soon

metze
(This used to be commit e5b67046c255c7dec07acf3ddc158278997502e3)

16 years agor20225: we can't use composite_error() in a _recv() function, as that would
Stefan Metzmacher [Sun, 17 Dec 2006 13:33:43 +0000 (13:33 +0000)]
r20225: we can't use composite_error() in a _recv() function, as that would
trigger the caller to call the _recv() function again and will be an endless
loop.

this is just a fix the to prevent this, and use a more usefull error code
than NT_STATUS_UNSUCCESSFUL

I think we should move the checks about valid responses into the function
which receives the the response (here continue_name_found()),
so that the _recv() function only needs to transfer the output vars to the caller
without any logic to analyse the network response.

metze
(This used to be commit c02048f4800fe14e975f1cf0b5bd45f14261ac18)

16 years agor20224: the struct element msDs_Schema_Extensions and the attribute name "msDs-Schema...
Stefan Metzmacher [Sun, 17 Dec 2006 13:20:57 +0000 (13:20 +0000)]
r20224: the struct element msDs_Schema_Extensions and the attribute name "msDs-Schema-Extensions"
doesn't match so we need to pass 2 parameters to the GET_BLOB() macro

metze
(This used to be commit e50cf96fea1318853a8cfc0061afcad9cbad7e2c)

16 years agor20223: this function is in the torture test now
Stefan Metzmacher [Sun, 17 Dec 2006 13:05:34 +0000 (13:05 +0000)]
r20223: this function is in the torture test now

metze
(This used to be commit 66c4b42f5f4c48e9c288781208ecdd4849c2f939)

16 years agor20222: return status unsuccessful when null pointers are returned
Rafal Szczesniak [Sun, 17 Dec 2006 13:01:35 +0000 (13:01 +0000)]
r20222: return status unsuccessful when null pointers are returned
from lookup call.

rafal
(This used to be commit 0eb605cf423faee2111b3d70cef1c578564a0fb5)

16 years agor20221: - move the dsdb_schema, dsdb_attribute, dsdb_class structs into
Stefan Metzmacher [Sun, 17 Dec 2006 02:19:56 +0000 (02:19 +0000)]
r20221: - move the dsdb_schema, dsdb_attribute, dsdb_class structs into
  a generic location
- this struct should be also used by the dsdb/samdb/ldb_modules/schema.c
  module later
- add functions to map from LDAP OID values to uint32_t values
- add torture test for the OID => uint32 mapping code

metze
(This used to be commit 58ba6ec195ce3a5e2352866809f6cb57f8d8260a)

16 years agor20220: Apply patch from Brad Henry with some additional qooxdoo/json-rpc SWAT
Derrell Lipman [Sat, 16 Dec 2006 19:07:37 +0000 (19:07 +0000)]
r20220: Apply patch from Brad Henry with some additional qooxdoo/json-rpc SWAT
functionality: basic admin console.
(This used to be commit 23f8dd44cbf85e05e4b3d38de9ed7798dd21defb)

16 years agor20219: add enough knowledge to display the content of attributeSchema objects
Stefan Metzmacher [Sat, 16 Dec 2006 19:05:50 +0000 (19:05 +0000)]
r20219: add enough knowledge to display the content of attributeSchema objects

metze
(This used to be commit 38ae6ce7b9b5dd733e4e838da04d1c570c450c97)

16 years agor20213: add 2 error codes related to the msDs-IntId attribute
Stefan Metzmacher [Sat, 16 Dec 2006 17:21:53 +0000 (17:21 +0000)]
r20213: add 2 error codes related to the msDs-IntId attribute

metze
(This used to be commit dd5df84cccd10e9880648eecbff6faf7252d5e73)

16 years agor20205: remove unused var
Stefan Metzmacher [Sat, 16 Dec 2006 01:24:43 +0000 (01:24 +0000)]
r20205: remove unused var

metze
(This used to be commit a77e1bd1b48e953773db47a5469b1712794f26cf)

16 years agor20204: use symbolic constant
Stefan Metzmacher [Sat, 16 Dec 2006 01:21:59 +0000 (01:21 +0000)]
r20204: use symbolic constant

metze
(This used to be commit a816c2973a5cded5d2e520f35a79dc519205f350)

16 years agor20198: let the IBM checker ignore the warning about a missing break
Stefan Metzmacher [Fri, 15 Dec 2006 23:00:24 +0000 (23:00 +0000)]
r20198: let the IBM checker ignore the warning about a missing break
statement...

metze
(This used to be commit 6b20123c10b8812759b0876990766adc207bc5b4)

16 years agor20195: fix typo
Stefan Metzmacher [Fri, 15 Dec 2006 22:51:31 +0000 (22:51 +0000)]
r20195: fix typo

metze
(This used to be commit 9d06a4b6e22e18bdaa4383c7b9ab8af058c157c0)

16 years agor20194: inline all static talloc functions
Stefan Metzmacher [Fri, 15 Dec 2006 22:47:50 +0000 (22:47 +0000)]
r20194: inline all static talloc functions

metze
(This used to be commit ee47b404412299ac0e341f0d269adc21a7ebc603)

16 years agor20193: - let talloc_parent_chunk() handle a NULL pointer
Stefan Metzmacher [Fri, 15 Dec 2006 22:45:36 +0000 (22:45 +0000)]
r20193: - let talloc_parent_chunk() handle a NULL pointer
- use talloc_parent_chunk() in talloc_parent_name()
- add prototype of talloc_parent_name()

metze
(This used to be commit 85fc18f047cd2132d0c455f739ee76ce5005d7ed)

16 years agor20192: I assume a 'break' is the correct thing to do here,
Stefan Metzmacher [Fri, 15 Dec 2006 22:37:06 +0000 (22:37 +0000)]
r20192: I assume a 'break' is the correct thing to do here,

simo,tridge: please check this.

found by the IBM checker

metze
(This used to be commit 5ac373c8b853d4527a095111253f3cb10522f5e8)

16 years agor20191: fix bug found by the IBM checker
Stefan Metzmacher [Fri, 15 Dec 2006 22:00:56 +0000 (22:00 +0000)]
r20191: fix bug found by the IBM checker

metze
(This used to be commit 4c1e4bfeef8d93583b6d4345dbb1b8fa90368308)

16 years agor20190: fix the ldb_ldap backend
Stefan Metzmacher [Fri, 15 Dec 2006 21:04:01 +0000 (21:04 +0000)]
r20190: fix the ldb_ldap backend

metze
(This used to be commit 6b7eb5d68e769fd5450c5d92c3ece4c471aaeae9)

16 years agor20189: remove unused struct element
Stefan Metzmacher [Fri, 15 Dec 2006 20:22:17 +0000 (20:22 +0000)]
r20189: remove unused struct element

metze
(This used to be commit d20d1872d5ed1176928b85ef9811c6a5177d0148)

16 years agor20188: move back to an default attribute handler and not use the '*' attribute
Stefan Metzmacher [Fri, 15 Dec 2006 18:56:56 +0000 (18:56 +0000)]
r20188: move back to an default attribute handler and not use the '*' attribute
to not conflict with the one that maybe added via the @ATTRIBUTES object

this is just to make the test-tdb-feature.sh torture test happy

There's still a bug when a attribute is registered multiple time
without removing old ldb_schema_attribute instances. But this bug
was there before my changes too and was just triggered by my changes

metze
(This used to be commit 70c4a367433f8c54bdd940eb0a6a24ab976a4063)

16 years agor20187: fix the build on systems without sqlite3
Stefan Metzmacher [Fri, 15 Dec 2006 17:54:47 +0000 (17:54 +0000)]
r20187: fix the build on systems without sqlite3

metze
(This used to be commit 216713d9a034984ffb40889b42837b91abae862a)

16 years agor20185: - SMB_ENABLE() needs upper case YES/NO
Stefan Metzmacher [Fri, 15 Dec 2006 13:12:18 +0000 (13:12 +0000)]
r20185: - SMB_ENABLE() needs upper case YES/NO
- fix compilation of the ldb_sqlite3 module

metze
(This used to be commit 39c41be0ac5bfa39c4ff3267b1a8291857e3a034)

16 years agor20184: change ldb_attrib_handler into ldb_schema_attribute, which has a pointer
Stefan Metzmacher [Fri, 15 Dec 2006 13:08:57 +0000 (13:08 +0000)]
r20184: change ldb_attrib_handler into ldb_schema_attribute, which has a pointer
to a ldb_schema_syntax struct.

the default attribute handler is now registered dynamicly as "*"
attribute, instead of having its own code path.

ldb_schema_attribute's can be added to the ldb_schema given a
ldb_schema_syntax struct or the syntax name

we may also need to introduce a ldb_schema_matching_rule,
and add a pointer to a default ldb_schema_matching_rule
in the ldb_schema_syntax.

metze
(This used to be commit b97b8f5dcbce006f005e53ca79df3330e62f117b)

16 years agor20182: make the composite api usage more consistant (only cosmetic change)
Stefan Metzmacher [Fri, 15 Dec 2006 11:07:35 +0000 (11:07 +0000)]
r20182: make the composite api usage more consistant (only cosmetic change)

metze
(This used to be commit 1f67433914bfab83bf1950263809c9e152ad6dcc)

16 years agor20177: return the actual function status code.
Rafal Szczesniak [Thu, 14 Dec 2006 22:45:12 +0000 (22:45 +0000)]
r20177: return the actual function status code.

rafal
(This used to be commit 2d046cc0df820edc141a74e7d204e4fba462e037)

16 years agor20176: Info method returns null object if the user is not found.
Rafal Szczesniak [Thu, 14 Dec 2006 22:14:07 +0000 (22:14 +0000)]
r20176: Info method returns null object if the user is not found.

rafal
(This used to be commit a20b05183d274a3a780ae197dc7014428739cc7a)

16 years agor20175: use libnet context instead mem_ctx as the latter gets
Rafal Szczesniak [Thu, 14 Dec 2006 22:12:53 +0000 (22:12 +0000)]
r20175: use libnet context instead mem_ctx as the latter gets
freed just before the function returns.

rafal
(This used to be commit ca8259fd27d1e92bcb650441461afdfd613e93bf)

16 years agor20168: start separating attributes and syntaxes
Stefan Metzmacher [Thu, 14 Dec 2006 10:03:21 +0000 (10:03 +0000)]
r20168: start separating attributes and syntaxes

metze
(This used to be commit 8dda4342f648aa71878ac9eeb7941710e2813aee)

16 years agor20167: reenable the ejsnet tests
Stefan Metzmacher [Thu, 14 Dec 2006 08:25:24 +0000 (08:25 +0000)]
r20167: reenable the ejsnet tests

metze
(This used to be commit 1e4289fae4fb71725d9c37ed772eab0ff5853e65)

16 years agor20166: we have a dom_sid_add_rid() function that adds the rid after allocating
Stefan Metzmacher [Thu, 14 Dec 2006 08:23:55 +0000 (08:23 +0000)]
r20166: we have a dom_sid_add_rid() function that adds the rid after allocating
enough memory for the new sub_auth element.

the old version wrote behind the buffer.

also make the output sid a pointer.

metze
(This used to be commit b9901d5f8c9aa777bafd3a670d27209552bc56b7)

16 years agor20164: Disable the test till tomorrow when I find out what caused
Rafal Szczesniak [Wed, 13 Dec 2006 23:53:42 +0000 (23:53 +0000)]
r20164: Disable the test till tomorrow when I find out what caused
the crash.

rafal
(This used to be commit 5d7908038b6508f53cc8714f73ae7c5bfc6bce58)

16 years agor20163: Return null object if no user is found.
Rafal Szczesniak [Wed, 13 Dec 2006 23:37:22 +0000 (23:37 +0000)]
r20163: Return null object if no user is found.

rafal
(This used to be commit af539fec4acb3ad78b8f0fa4f7c2f4e6cebea0b7)

16 years agor20162: Always return some error string.
Rafal Szczesniak [Wed, 13 Dec 2006 23:36:34 +0000 (23:36 +0000)]
r20162: Always return some error string.

rafal
(This used to be commit 0c373605776eb166ffe46db0dd21395efc3bb186)

16 years agor20161: Prevent potential segfault in case account is unknown.
Rafal Szczesniak [Wed, 13 Dec 2006 23:35:50 +0000 (23:35 +0000)]
r20161: Prevent potential segfault in case account is unknown.

rafal
(This used to be commit 9f5f9ee9b4b815715c0b580e21ab0906fa15ae96)

16 years agor20152: Commit missing files from last night's commit. We no longer maintain
Andrew Bartlett [Wed, 13 Dec 2006 20:47:24 +0000 (20:47 +0000)]
r20152: Commit missing files from last night's commit.  We no longer maintain
a distinction between PDC and BDC in the configuration files, only as
an entry in the ldb.

Andrew Bartlett
(This used to be commit dc9eee7cb37e4a6828c2cba23b0d836df9eac7b5)

16 years agor20149: Remove the smb.conf distinction between PDC and BDC. Now the correct
Andrew Bartlett [Wed, 13 Dec 2006 11:19:51 +0000 (11:19 +0000)]
r20149: Remove the smb.conf distinction between PDC and BDC.  Now the correct
way to setup a Samba4 DC is to set 'server role = domain controller'.

We use the fSMORoleOwner attribute in the base DN to determine the PDC.

This patch is quite large, as I have corrected a number of places that
assumed taht we are always the PDC, or that used the smb.conf
lp_server_role() to determine that.

Also included is a warning fix in the SAMR code, where the IDL has
seperated a couple of types for group display enumeration.

We also now use the ldb database to determine if we should run the
global catalog service.

In the near future, I will complete the DRSUAPI
DsGetDomainControllerInfo server-side on the same basis.

Andrew Bartlett
(This used to be commit 67d8365e831adf3eaecd8b34dcc481fc82565893)

16 years agor20145: fix whitespaces
Stefan Metzmacher [Tue, 12 Dec 2006 23:48:28 +0000 (23:48 +0000)]
r20145: fix whitespaces

metze
(This used to be commit d2d9842914bb049189f9349ff96b32c088cd02a8)

16 years agor20144: add NET-API-BECOME-DC test that tests libnet_BecomeDC()/libnet_UnbecomeDC()
Stefan Metzmacher [Tue, 12 Dec 2006 23:23:50 +0000 (23:23 +0000)]
r20144: add NET-API-BECOME-DC test that tests libnet_BecomeDC()/libnet_UnbecomeDC()

metze
(This used to be commit 0359d9a00b3e30e2bd67b89d6b11236e5af0106b)

16 years agor20143: a bit of experiments before doing serious changes in ejsnet.
Rafal Szczesniak [Tue, 12 Dec 2006 23:01:51 +0000 (23:01 +0000)]
r20143: a bit of experiments before doing serious changes in ejsnet.

rafal
(This used to be commit 3cdd43ebcedfa7d1d59bd33b3bfed4e80560ec82)

16 years agor20142: only the DDNS updates are missing for the full libnet_BecomeDC(),
Stefan Metzmacher [Tue, 12 Dec 2006 22:57:43 +0000 (22:57 +0000)]
r20142: only the DDNS updates are missing for the full libnet_BecomeDC(),
but that's not needed untill we can support the DsGetNCChanges()
in the server...

metze
(This used to be commit 9c3267e17f984a0a47abd35cecb56c43cf97efcd)

16 years agor20141: use the gensec_features of the cli_credentials for ildap connections,
Stefan Metzmacher [Tue, 12 Dec 2006 22:43:35 +0000 (22:43 +0000)]
r20141: use the gensec_features of the cli_credentials for ildap connections,
instead of hardcoded GENSEC_FEATURE_SEAL.

That means plain LDAP is now the default.

metze
(This used to be commit b69471866c2a6c61002147938f233f2f63963ba4)

16 years agor20139: only add GSS_C_CONF_FLAG and GSS_C_INTEG_FLAG if the caller requested it!
Stefan Metzmacher [Tue, 12 Dec 2006 22:38:23 +0000 (22:38 +0000)]
r20139: only add GSS_C_CONF_FLAG and GSS_C_INTEG_FLAG if the caller requested it!

this is needed to create plain, singed or sealed LDAP connections.

this should go into lorikeet and main heimdal...

metze
(This used to be commit 75c037cae21714e394a63f2506387e1049eb4406)

16 years agor20138: return the proper status for the other functions as well.
Rafal Szczesniak [Tue, 12 Dec 2006 22:34:35 +0000 (22:34 +0000)]
r20138: return the proper status for the other functions as well.

rafal
(This used to be commit 0b82b9dba1f67b1d340b7cf2b35a826135c101d4)

16 years agor20137: return the proper status.
Rafal Szczesniak [Tue, 12 Dec 2006 22:28:33 +0000 (22:28 +0000)]
r20137: return the proper status.

rafal
(This used to be commit e635f5aeaf7cd722e55bd99e52a8de166ae0ecd5)

16 years agor20135: attach default gensec features to the cli_credentials structure,
Stefan Metzmacher [Tue, 12 Dec 2006 21:47:56 +0000 (21:47 +0000)]
r20135: attach default gensec features to the cli_credentials structure,
so make it possible to force encryption or signing.

metze
(This used to be commit a91dc4a02a46370c52f59cbd4dea9580fa6efafa)

16 years agor20134: The IBM Checker correctly notes that *p cannot be \0 and still satisfy
Andrew Bartlett [Tue, 12 Dec 2006 21:47:47 +0000 (21:47 +0000)]
r20134: The IBM Checker correctly notes that *p cannot be \0 and still satisfy
the loop entry condition.

Andrew Bartlett
(This used to be commit c1182751313290fc38af5b1c44bc66b5d1654977)

16 years agor20130: remove one more talloc_get_type()
Stefan Metzmacher [Tue, 12 Dec 2006 19:02:14 +0000 (19:02 +0000)]
r20130: remove one more talloc_get_type()

metze
(This used to be commit 024dacb3b0a32bcaaef0b7a7598070b9034a7648)

16 years agor20129: remove unused structure elements
Stefan Metzmacher [Tue, 12 Dec 2006 18:58:21 +0000 (18:58 +0000)]
r20129: remove unused structure elements

metze
(This used to be commit 53805a8562c59116e458dedd695e061eabf78620)

16 years agor20128: get rid of more talloc_get_type() calls
Stefan Metzmacher [Tue, 12 Dec 2006 18:52:25 +0000 (18:52 +0000)]
r20128: get rid of more talloc_get_type() calls

metze
(This used to be commit cb89f0b8d5a64433374887bcd44e04ad63e4857e)

16 years agor20127: - allocate ldb_handle under ldb_request! fix a mem leak
Stefan Metzmacher [Tue, 12 Dec 2006 18:45:12 +0000 (18:45 +0000)]
r20127: - allocate ldb_handle under ldb_request! fix a mem leak
- pass ldb_request to init_lldb_handle()
- remove some useless talloc_get_type() calls

metze
(This used to be commit a7397c4d2bed181c96863e985727c8dad0894df7)

16 years agor20126: fix talloc hierachy and make lldb a child of module
Stefan Metzmacher [Tue, 12 Dec 2006 18:25:19 +0000 (18:25 +0000)]
r20126: fix talloc hierachy and make lldb a child of module

metze
(This used to be commit 1e3bb180261643900c7bdacef818add5349c7a30)

16 years agor20125: fix some ugly mem leaks in the ldb_ildb backend
Stefan Metzmacher [Tue, 12 Dec 2006 18:14:31 +0000 (18:14 +0000)]
r20125: fix some ugly mem leaks in the ldb_ildb backend

metze
(This used to be commit db85b7840c1022665d410516d9a8989474d67a0f)

16 years agor20123: - avoid some more talloc_get_type() calls
Stefan Metzmacher [Tue, 12 Dec 2006 17:37:41 +0000 (17:37 +0000)]
r20123: - avoid some more talloc_get_type() calls
- pass down ldb_request

metze
(This used to be commit b0b9e83fed0145bac6058d11e9b1b2c57d091649)

16 years agor20122: pass ildb to ildb_request_noop()
Stefan Metzmacher [Tue, 12 Dec 2006 17:20:38 +0000 (17:20 +0000)]
r20122: pass ildb to ildb_request_noop()

metze
(This used to be commit cf9aade216930980bf47ea8b97db976c4907cbca)

16 years agor20121: pass down the ldb_request struct to ildb_request_send(),
Stefan Metzmacher [Tue, 12 Dec 2006 17:14:58 +0000 (17:14 +0000)]
r20121: pass down the ldb_request struct to ildb_request_send(),
also pass ildb instead of module, to avoid multiple talloc_get_type() calls

metze
(This used to be commit 03029d4fed86b1ca5fb19a408312607d8790d110)

16 years agor20120: fix the talloc hierachy and make ildb a child of module
Stefan Metzmacher [Tue, 12 Dec 2006 16:56:53 +0000 (16:56 +0000)]
r20120: fix the talloc hierachy and make ildb a child of module

metze
(This used to be commit b85d5cb7a4931d1d43a0ec73f1de1519c720f1af)

16 years agor20115: from Julien Kerihuel <j.kerihuel@openchange.org>, thanks!:
Stefan Metzmacher [Tue, 12 Dec 2006 07:53:12 +0000 (07:53 +0000)]
r20115: from Julien Kerihuel <j.kerihuel@openchange.org>, thanks!:

   Here is a patch that adds a torture:dc_binding parametric option to
   torture_create_testuser in testjoin.c. This patch is used in OpenChange
   torture tests when the AD is different from the Exchange Server. This generic
   option could also be used for later 'member server' tests in Samba4.

metze
(This used to be commit 9ddabfc60e1a4f5e471991253141ce458c9a683d)

16 years agor20114: from Julien Kerihuel <j.kerihuel@openchange.org>, thanks!:
Stefan Metzmacher [Tue, 12 Dec 2006 07:46:06 +0000 (07:46 +0000)]
r20114: from Julien Kerihuel <j.kerihuel@openchange.org>, thanks!:

  I've attached the patch which fix this problem. I've only added
  DCERPC_NDR_REF_ALLOC to the connection flags. This way it is processed
  correctly by ndr_pull_init_flags and added to the ndr flags of the pull
  structure.

metze
(This used to be commit ed4c7ce547c61907291d19c172d5eb6f4c4981fe)

16 years agor20113: Update the DRSUAPI CrackNames test to explore a few more cases, and in
Andrew Bartlett [Mon, 11 Dec 2006 23:59:03 +0000 (23:59 +0000)]
r20113: Update the DRSUAPI CrackNames test to explore a few more cases, and in
particular to verify more expected results.

Also return more details from the join process.  Now we also return
the machine account's GUID.

Andrew Bartlett
(This used to be commit 5b32f102af1fc7acb56bf7eaa40068d60a1ee396)

16 years agor20111: this is the total object count of the current replication cycle
Stefan Metzmacher [Mon, 11 Dec 2006 20:14:49 +0000 (20:14 +0000)]
r20111: this is the total object count of the current replication cycle

metze
(This used to be commit 4daa45f98a3bb6f3c245fe61e05681335ef8b5c6)

16 years agor20109: let libnet_BecomeDC() work against w2k servers
Stefan Metzmacher [Mon, 11 Dec 2006 17:04:43 +0000 (17:04 +0000)]
r20109: let libnet_BecomeDC() work against w2k servers

metze
(This used to be commit d295239861a282d06a9a93cdafa4e4d7616928b8)

16 years agor20108: match w2k3 and send 1.2.840.48018.1.2.2 before 1.2.840.113554.1.2.2
Stefan Metzmacher [Mon, 11 Dec 2006 16:45:31 +0000 (16:45 +0000)]
r20108: match w2k3 and send 1.2.840.48018.1.2.2 before 1.2.840.113554.1.2.2
to work better against w2k, so we don't get redirected from
1.2.840.113554.1.2.2 to 1.2.840.48018.1.2.2 by a w2k server, causing 2 additional
auth roundtrips.

metze
(This used to be commit fa5c942ee99d3b5779598aa75f71d0317ba3f622)

16 years agor20107: don't segfault when the server didn't send a supportedMech in a negTokenTarg
Stefan Metzmacher [Mon, 11 Dec 2006 16:36:30 +0000 (16:36 +0000)]
r20107: don't segfault when the server didn't send a supportedMech in a negTokenTarg

this happens because we send 1.2.840.113554.1.2.2 before 1.2.840.48018.1.2.2
in the negTokenInit. And w2k's spnego code redirects us to use 1.2.840.48018.1.2.2
and then we start the our spnego engine with 1.2.840.48018.1.2.2 and in the then following
negTokenTarg w2k don't send the supportedMech (which means it aggrees in what we've choosen)

metze
(This used to be commit 5af5488593991ab4a2a8e17d38501ad9ec539020)

16 years agor20106: Optional ONE Level indexing for ldb_tdb
Simo Sorce [Mon, 11 Dec 2006 15:49:39 +0000 (15:49 +0000)]
r20106: Optional ONE Level indexing for ldb_tdb
To activate it you must modify the @INDEXLIST object adding
the attribute @IDXONE: 1

Ldb test included

Simo.
(This used to be commit ea111795f4016916473ccc05d23c6655e6af1207)

16 years agor20105: remove useless comments
Stefan Metzmacher [Mon, 11 Dec 2006 13:17:36 +0000 (13:17 +0000)]
r20105: remove useless comments

metze
(This used to be commit d70140ed07bf4d0d1d0a0485ca73a29cf1cafe8c)

16 years agor20104: this is a alternative events backend, which uses a hybrid of aio and
Andrew Tridgell [Mon, 11 Dec 2006 08:53:29 +0000 (08:53 +0000)]
r20104: this is a alternative events backend, which uses a hybrid of aio and
epoll. It is not linked in anywhere yet - I'm committing it in case
anyone else wants to have a look at it.

The concept is quite strange really, but it seems to be the only way
that Linux 2.6.x can currently use a unified event model allowing for
AIO events and socket events to be waited for by a single unified
event wait function. You setup a epoll system, then setup a weird aio
event that points at the epoll system, then use io_getevents() to
actually do the waiting.

I'm hoping that kevents or a proper integration of epoll will allow us
to avoid ths rather hackish scheme, but meanwhile this is the only
path to proper AIO in Samba on Linux (without a horrible signals mess)

(NOTE: this code requires some kernel patches to work at the moment)
(This used to be commit 195051fdee341e8d8cb76e5c91dcc0f6c246a870)

16 years agor20103: Change the returned type of prerequisite checking functions,
Rafal Szczesniak [Sun, 10 Dec 2006 23:43:32 +0000 (23:43 +0000)]
r20103: Change the returned type of prerequisite checking functions,
as metze once suggested.

rafal
(This used to be commit 1f3fd5f85461d1e8c9cbdc3144d0a6533b170f76)

16 years agor20102: Do not reference remote_ldb before we initialise it. This should fix
Andrew Bartlett [Sun, 10 Dec 2006 23:28:36 +0000 (23:28 +0000)]
r20102: Do not reference remote_ldb before we initialise it.  This should fix
up many of the build farm failures.

Andrew Bartlett
(This used to be commit 924af98ffaab3735c3f31014059e7cd0abd71919)

16 years agor20101: Also rename a variable now that the unused parameter is gone
Simo Sorce [Sun, 10 Dec 2006 22:24:28 +0000 (22:24 +0000)]
r20101: Also rename a variable now that the unused parameter is gone
(This used to be commit a2520bcfa918977f2139a963e9817370789cf077)

16 years agor20100: Remove completely unused parameters
Simo Sorce [Sun, 10 Dec 2006 22:21:55 +0000 (22:21 +0000)]
r20100: Remove completely unused parameters
(This used to be commit cc1bcb814844e8a03dfa9a310d26ce3f3441e7bb)

16 years agor20099: Add some comments, and correct others.
Andrew Bartlett [Sun, 10 Dec 2006 22:21:20 +0000 (22:21 +0000)]
r20099: Add some comments, and correct others.

Andrew Bartlett
(This used to be commit d1b1a4c059bfa93bd55f9ffc718e802695c50cc9)

16 years agor20096: now I'm 100% sure:-)
Stefan Metzmacher [Sat, 9 Dec 2006 15:30:15 +0000 (15:30 +0000)]
r20096: now I'm 100% sure:-)

metze
(This used to be commit 3daa68e1c4212f1dad98b196a096ded3204b03ba)

16 years agor20095: make a complete copy of the binding string
Stefan Metzmacher [Sat, 9 Dec 2006 15:27:22 +0000 (15:27 +0000)]
r20095: make a complete copy of the binding string

metze
(This used to be commit b0ee1308928d4747639033807aa3f9441ba2ac14)

16 years agor20094: this should be non-const
Stefan Metzmacher [Sat, 9 Dec 2006 14:49:17 +0000 (14:49 +0000)]
r20094: this should be non-const

metze
(This used to be commit 1b6621814ba83724e05c0c2bea28d6eb295a3655)

16 years agor20093: some documentation updates
Stefan Metzmacher [Sat, 9 Dec 2006 13:54:30 +0000 (13:54 +0000)]
r20093: some documentation updates

metze
(This used to be commit 314308c0bf8eeb168d59dab2a867e6bf5f51266b)

16 years agor20092: added a locking benchmark that should be good for benchmarking
Andrew Tridgell [Sat, 9 Dec 2006 03:16:17 +0000 (03:16 +0000)]
r20092: added a locking benchmark that should be good for benchmarking
communitcation in a CIFS clustered server. It tries to keep the
connections full by setting up the next lock as each lock is done.

The locking pattern is similar to the local filesystem ping pong test
in junkcode, forcing a communication between nodes on each request
(This used to be commit d57b9fb29860bd03cfa970bcc52ef45d17775638)

16 years agor20091: remove blank lines at the end of text lines loaded from a file
Andrew Tridgell [Sat, 9 Dec 2006 03:12:33 +0000 (03:12 +0000)]
r20091: remove blank lines at the end of text lines loaded from a file
(This used to be commit afcc797e8a25d99269f67e05fe57e952dd5d65bc)

16 years agor20088: implement DsReplicaUpdateRefs() calls for all partitions
Stefan Metzmacher [Fri, 8 Dec 2006 18:52:29 +0000 (18:52 +0000)]
r20088: implement DsReplicaUpdateRefs() calls for all partitions

metze
(This used to be commit 557bb8008916fbba95dd3b3a1a872ddf5a143693)

16 years agor20086: pull domain partition
Stefan Metzmacher [Fri, 8 Dec 2006 18:29:44 +0000 (18:29 +0000)]
r20086: pull domain partition

metze
(This used to be commit 8b03b3a196d9c2b33344d8d0730596b7f24c61dc)

16 years agor20085: add _part to structure elements
Stefan Metzmacher [Fri, 8 Dec 2006 18:22:26 +0000 (18:22 +0000)]
r20085: add _part to structure elements

metze
(This used to be commit de2f9869b5ab97adcb018ede10941562eca518c7)

16 years agor20084: pull the config partition
Stefan Metzmacher [Fri, 8 Dec 2006 18:13:49 +0000 (18:13 +0000)]
r20084: pull the config partition

metze
(This used to be commit d770effb89a84b53fcdbe0ced29d76455321df9d)

16 years agor20083: move function together which belong to each other
Stefan Metzmacher [Fri, 8 Dec 2006 18:08:30 +0000 (18:08 +0000)]
r20083: move function together which belong to each other

metze
(This used to be commit 0c4a7d713da3a96ab1abb3ef65c2fecd110a0549)

16 years agor20081: pull the whole schema partition
Stefan Metzmacher [Fri, 8 Dec 2006 18:06:38 +0000 (18:06 +0000)]
r20081: pull the whole schema partition

metze
(This used to be commit 2f1ec24d776a7ecf4d9b4270f8969e57a4335050)

16 years agor20080: add error code that maps to NT_STATUS_INVALID_NETWORD_RESPONSE
Stefan Metzmacher [Fri, 8 Dec 2006 18:01:30 +0000 (18:01 +0000)]
r20080: add error code that maps to NT_STATUS_INVALID_NETWORD_RESPONSE

metze
(This used to be commit cb5c2e9dc6838145fe9f6f2e727a1df26e688467)

16 years agor20079: the 2nd guid is the source_dsa invocation_id not the objectGUID,
Stefan Metzmacher [Fri, 8 Dec 2006 17:34:15 +0000 (17:34 +0000)]
r20079: the 2nd guid is the source_dsa invocation_id not the objectGUID,
this wasn't noticed because on the 1st dc in the forest both have the
same value

metze
(This used to be commit 527bd9a0a361e19606e25e885b92da316e740bf9)