summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2009-12-21s4-repl: only try to replicate for NCs that we are a master forAndrew Tridgell1-4/+3
2009-12-21s4-torture: another unsigned comparison bugAndrew Tridgell1-1/+1
2009-12-21s4-schema: a unsigned comparison bug in the schema codeAndrew Tridgell1-1/+2
2009-12-21s4-drs: another two unsigned comparison bugsAndrew Tridgell1-3/+5
2009-12-21librpc: fixed the GUID_compare() functionAndrew Tridgell1-5/+5
When comparing two unsigned values you can't just subtract them. Imagine you are comparing: "uint32_t u1" and "uint32_t u2". If you use "u1 - u2" and u2 is zero, then the signed integer result will depend on the top bit of u1. This error occurs in a few places in Samba. For DRS replication it resulted in corrupt uptodateness vectors.
2009-12-21s4-repl: lower debug level of a common messageAndrew Tridgell1-1/+1
2009-12-21s4-dsdb: don't use a non-constant format string for a printf formatAndrew Tridgell1-10/+3
2009-12-21s4-dsdb: added DSDB_MODIFY_RELAX flag to the dsdb_module_*() callsAndrew Tridgell2-0/+8
2009-12-21s4-dsdb: added dsdb_get_extended_dn_uint64()Andrew Tridgell1-3/+11
2009-12-21s4-dsdb: use varargs expression in dsdb_module_search()Andrew Tridgell1-1/+7
Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2009-12-21s4-dsdb: added two new dsdb_get_extended_dn_*() helper functionsAndrew Tridgell1-0/+43
Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2009-12-21s4-provision: added a note about where invocationIDs come fromAndrew Tridgell1-0/+2
Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2009-12-21s4-dsdb: give us an invocationID when in standalone modeAndrew Tridgell1-1/+79
To allow us to use the repl_meta_data module in standalone mode (and thus not have two module stacks to test), we need a invocationID stored somewhere when standalone. This creates a random one, and stores it in @SAMBA_DSDB. Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2009-12-20s3: Fix an error case in cli_negprotVolker Lendecke1-0/+1
2009-12-20tevent: prefix types and defined with tevent_ and TEVENT_Stefan Metzmacher1-37/+39
This fixes the build warnings on some build-farm hosts. metze
2009-12-19Output %p as unsigned in snprintf replacement.Peter Rosin1-0/+1
2009-12-19s4:kdc: setup the local and remote tsocket_address at accept timeStefan Metzmacher1-44/+49
metze
2009-12-19s4:kdc: convert UDP based communication to tdgram_contextStefan Metzmacher2-177/+138
metze
2009-12-18Actually explain the twisty paths of tortured logic behindJeremy Allison1-14/+42
reply_doserror(), reply_nterror(), and reply_nterror(NT_STATUS_DOS()). Fix the call in rely_openerror() to actually force a DOS error for "too many open files". Jeremy.
2009-12-18reply_doserror() doesn't force DOS errors on the wire.Jeremy Allison1-10/+10
Start migrating uses of reply_doserror() to reply_nterror() with the correct mapping. Eventually we'll get to the point where we can change reply_doserror() to force a DOS error code on the wire, and can change calls to reply_nterror(req, NT_STATUS_DOS()) - which *does* force DOS errors on the wire - to reply_doserror(). Which might actually make the server code look like it's making sense. Jeremy.
2009-12-18reply_force_nterror() is not used anywhere. Remove it.Jeremy Allison2-3/+0
Jeremy.
2009-12-18s4 torture: Add test to show archive bit behavior with directoriesZachary Loafman1-0/+153
Signed-off-by: Tim Prouty <tprouty@samba.org>
2009-12-18s4 torture: Fix RAW-STREAMS-DELETE to pass against samba3Tim Prouty1-1/+2
2009-12-18Added freeing a successful req so it doesnt croud the ldb contextNadezhda Ivanova1-2/+4
2009-12-18Added oid for AS_SYSTEM control, used to bypass access checks for system ↵Nadezhda Ivanova1-0/+6
operations.
2009-12-18s3-docs: mention pam_winbind.conf(5) manpage in pam_winbind(8) manpage.Günther Deschner1-1/+5
Guenther
2009-12-18s3-docs: add new pam_winbind.conf(5) manpage.Günther Deschner1-0/+190
Guenther
2009-12-18s4-dsdb: stop warnings about unknown struct GUID in prototypesAndrew Tridgell1-0/+1
Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2009-12-18s4-ldb: fixed a valgrind error in ldbtestAndrew Tridgell1-0/+2
we were using msg->dn after the ldb it contained had been freed Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2009-12-18s4-dsdb: greatly simplify the subtree_delete moduleAndrew Tridgell1-117/+20
We can use dsdb_module_search() to make this much simpler Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2009-12-18s4-dsdb: declare ldb_dn_update_components()Andrew Tridgell1-0/+1
2009-12-18s4-dsdb: added ldb_dn_update_components()Andrew Tridgell1-0/+23
This is used to udpate just the DN components of a ldb_dn, leaving the other extended fields alone. It is needed to prevent linked attribute updates from removing other extended components. Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2009-12-18s4-dsdb: fixed the sort in dsdb_find_nc_root()Andrew Tridgell1-1/+6
2009-12-18s4-ldb: display msDS-OptionalFeatureGUID as a GUIDAndrew Tridgell1-0/+1
Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2009-12-18s4-scripts: add a enablerecyclebin scriptAndrew Tridgell1-0/+54
This can be used to enable the recyclebin on a windows box. Once we properly implement this feature in samba we will use this to enable the feature on ourselves as well.
2009-12-18s4-ldb: canonicalise the message on ldb_addAndrew Tridgell1-0/+9
This canonicalise avoids a problem with an add that has multiple elements with the same el->name. That is allowed by MS servers, and by ldb, but it breaks things like the tdb backend and the repl_meta_data RPMD handling. Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2009-12-18s4-dsdb: Add a test for adding, deleting, and appending a posixAccount ↵Brendan Powers1-0/+30
objectClass to a user Signed-off-by: Andrew Bartlett <abartlet@samba.org>
2009-12-18s4-dsdb: fix handling of AUX classes in objectclass_sortBrendan Powers1-146/+133
This is done by sorting the classes by subClass_order, which will check if the last structural class is valid to add (in objectclass_do_add instead checking the last class in the list). They were being sorted by building a class tree, and adding the classes to the list in that order. However, AUX classes usually don't fit into that tree, so LDB_ERR_OBJECT_CLASS_VIOLATION was returned. I have changed the behavior to sort the classes by subClass_order instead. Signed-off-by: Andrew Bartlett <abartlet@samba.org>
2009-12-18s4-dsdb: return an error if samAccountName is not specified when creating a ↵Brendan Powers1-0/+7
user. Makes sure samAccountName has been specified before adding a user. This happened while I was trying to add a user with the posixAccount objectclass. I forgot to specify the user objectClass, and samba segfaulted. It now returns LDB_ERR_CONSTRAINT_VIOLATION. Signed-off-by: Andrew Bartlett <abartlet@samba.org>
2009-12-18s4-dsdb: Move get_last_structural class from descriptor.c to util.cBrendan Powers3-16/+30
It can now also be used by objectclass.c get_last_structural_class now ignores AUX classes, because they are not structural Signed-off-by: Andrew Bartlett <abartlet@samba.org>
2009-12-18s4-dsdb: Add a check to prevent acl_modify from debuging a NULL messageBrendan Powers1-1/+5
Check to see if there were any messages passed to acl_modify before debugging the first one. I think I caused this by some malformed LDIF. Signed-off-by: Andrew Bartlett <abartlet@samba.org>
2009-12-18Return NULL in strlower_talloc if src is NULLBrendan Powers1-0/+4
Prevents strlower_talloc from segfaulting if you pass it a NULL string. Signed-off-by: Andrew Bartlett <abartlet@samba.org>
2009-12-18s4:provision Give a more useful error message in guess_namesAndrew Bartlett1-3/+11
The problem here is that as we start to get 'real users' they still decide to do silly things, like load empty but existing smb.conf files. Let's give them a better chance to discover what they did wrong. Andrew Bartlett
2009-12-17Always map EMFILE to ERRDOS, ERRnofids, *NOT* NT_STATUS_TOO_MANY_OPENED_FILES.Jeremy Allison1-0/+4
This is what W2KR3 does for NTCreateX and openX calls. May be the correct fix for bug 6837 - "Too many open files" when trying to access large number of files. Jeremy.
2009-12-17Vector correctly through reply_openerror() (which uses the same logic).Jeremy Allison1-6/+1
Jeremy.
2009-12-17s4 torture: RAW-STREAMS-NAMES Make sure the create time of the streams are ↵Tim Prouty1-0/+24
different from the base file
2009-12-17s4 torture: Add RAW-STREAMS-RENAME3 to show error when trying to overwrite a ↵Tim Prouty1-0/+92
stream with an open handle A normal file overwrite in this case would return NT_STATUS_ACCESS_DENIED, but if a stream is the target NT_STATUS_INVALID_PARAMETER is returned.
2009-12-17s4 torture: Parameterize streams delete behavior rather than commenting out ↵Tim Prouty1-3/+7
the check.
2009-12-17s4 torture: RAW-STREAMS-CREATEDISP Use a normal deny modeTim Prouty1-1/+1
2009-12-17s4 torture: close handle instead of using exitTim Prouty1-1/+1
generated by cgit (git 2.34.1) at 2024-12-14 19:09:27 +0000