summaryrefslogtreecommitdiff
path: root/source3
AgeCommit message (Collapse)AuthorFilesLines
2010-04-06build: check for headers needed to generate stack tracesKai Blin1-0/+2
2010-04-06build: Check for krb5 headersKai Blin1-0/+8
2010-04-06build: Check for LDAP librariesKai Blin1-0/+4
2010-04-06build: Check for CUPSKai Blin1-0/+5
2010-04-06build: Add a basic wscript file for Samba3Kai Blin1-0/+41
2010-04-05Add FSCTL_QUERY_ALLOCATED_RANGES to the list of Windows ioctls we support.Jeremy Allison1-0/+52
Based on a patch reported and tested by Ira Cooper <samba@ira.wakeful.net>. Jeremy.
2010-04-05Fix issue with aio where r/w lock wasn't kept across aio read operations.Jeremy Allison3-51/+110
Change schedule_aio_read_and_X/schedule_aio_write_and_X to return NTSTATUS. Move the grant and release of the lock into the aio code. Jeremy
2010-04-05Fix related to bug #7326 - can't read/write any files with full_auditVolker Lendecke1-2/+3
2010-04-05s3: Fix a cut&paste error in winbindd_list_groups_doneVolker Lendecke1-1/+1
2010-04-04s3: Fix the GNU ld version detection on SLESVolker Lendecke1-1/+1
On SLES 11 "ld -v" gives GNU ld (GNU Binutils; SUSE Linux Enterprise 11) 2.19 The regexp to intended to detect the "2.19" is confused by the "11" in the version string. I'm not really into regexps, and awk '{print $NF}' to me is a simple (and hopefully portable) way to print the last field.
2010-04-02s3: Fix a commentVolker Lendecke1-1/+1
2010-04-01Move initialize_async_io_handler() inside of smbd/aio.c.Jeremy Allison3-38/+42
Call from actual aio read or write. No reason to call this globally on startup. Jeremy.
2010-04-01Add torture test for bug 7310 - DOS attribute inconsistency with MS OfficeJeremy Allison1-2/+20
Ensure we don't regress. Jeremy.
2010-04-01s3:winbindd: remove unused variablesStefan Metzmacher1-2/+0
metze
2010-04-01s3:winbindd: fix problems with SIGCHLD handling (bug #7317)Stefan Metzmacher3-17/+6
The main problem is that we call CatchChild() within the parent winbindd, which overwrites the signal handler that was registered by winbindd_setup_sig_chld_handler(). That means winbindd_sig_chld_handler() and winbind_child_died() are never triggered when a winbindd domain child dies. As a result will get "broken pipe" for all requests to that domain. To reduce the risk of similar bugs in future we call CatchChild() in winbindd_reinit_after_fork() now. We also use a full winbindd_reinit_after_fork() in the cache validation child now instead instead of just resetting the SIGCHLD handler by hand. This will also fix possible tdb problems on systems without pread/pwrite and disabled mmap as we now correctly reopen the tdb handle for the child. metze
2010-04-01s3: Ensure NULL termination before printing in winbindd_pam_logoffVolker Lendecke1-3/+3
2010-04-01s3: Fix a typo in winbindd_pam_logoffVolker Lendecke1-1/+1
2010-04-01s3:winbindd: correctly invalidate the cached connectionStefan Metzmacher1-6/+11
There're maybe additional TCP connection for ncacn_ip_tcp. metze
2010-04-01s3:winbindd: only set child_domain in the childStefan Metzmacher1-1/+1
metze
2010-04-01s3:winbindd: make sure we don't try rpc requests against unaccessable domainsStefan Metzmacher1-5/+28
This makes sure we don't crash while trying to dereference domain->conn.cli->foo while trying to establish a rpc connection to the server. metze
2010-04-01s3: fix a typo in winbind_client_response_writtenVolker Lendecke1-2/+2
2010-03-31Make smbd_lock_socket/smbd_unlock_socket recursive with a ref_count.Jeremy Allison6-56/+68
As these always call exit_server, make that part of the function. Use _internal functions for the echo client. Metze please check ! Jeremy.
2010-03-31s3: Merge French translation for pam_winbind after reviewChristian PERRIER1-14/+17
- some wording improvements ("user" is not "usager" but "utilisateur" - gender neutrality changes: by rewording: ("vous n'êtes pas autorisé" --> "vous n'avez pas l'autorisation") - bad translation corrections: "valide"-->"valable" (a very common false friend) - and several changes we consider to be improvements, mostly because they are common wording in computing translations in French Signed-off-by: Lars Müller <lars@samba.org>
2010-03-31s3: Fix an error message in winbindd_pam_chauthtok()Volker Lendecke1-1/+1
2010-03-31s3: Ensure null termination in winbindd_pam_chauthtok()Volker Lendecke1-0/+4
2010-03-31s3: Make check_info3_in_group staticVolker Lendecke2-4/+2
2010-03-31s3-winbind: Make append_auth_data() staticVolker Lendecke2-8/+4
2010-03-31Fix bug 7310 - DOS attribute inconsistency with MS OfficeJeremy Allison1-0/+15
On rename we need to set the archive bit on the renamed file. Jeremy
2010-03-31Merge branch 'master' of ssh://git.samba.org/data/git/sambaJeremy Allison2-2/+14
2010-03-31s3:smbd: handle SMB2 in deadtime_fn() and avoid disconnecting non idle clientsStefan Metzmacher1-1/+13
metze
2010-03-31selftest: Convert filter-subunit to Python so the subunit Python moduleJelmer Vernooij1-1/+1
can be used later.
2010-03-30Second part of fix for bug #6494 - Incorrect FileStatus returned in ↵Jeremy Allison1-1/+19
NT_CREATE_ANDX. Ensure we do this in nttrans create too. Jeremy.
2010-03-30Fix for bug #7233 - print fails with jobs >4GB from Win7 clients.Jeremy Allison3-4/+33
Contains for by Sebastian Kloska <oncaphillis@snafu.de>. Submitter confirms this fixes the problem. Jeremy.
2010-03-30If the timeout has been set to zero, don't register an end time.Jeremy Allison1-3/+5
The caller doesn't want a timeout. Jeremy.
2010-03-30s3:libads: retry with signing after getting LDAP_STRONG_AUTH_REQUIREDStefan Metzmacher1-0/+10
If server requires LDAP signing we're getting LDAP_STRONG_AUTH_REQUIRED, if "client ldap sasl wrapping = plain", instead of failing we now autoupgrade to "client ldap sasl wrapping = sign" for the given connection. metze
2010-03-30selftest: Replace perl subunit formatter with python subunit formatter,Jelmer Vernooij1-1/+1
so we can leverage the work happening in python-subunit.
2010-03-29s3:winbindd: correctly retry if the netlogon pipe gets disconnected during a ↵Stefan Metzmacher1-2/+2
logon call This fixes hopefully the last part of bug #7295. metze
2010-03-29s3:rpc_client: remove more unused codeStefan Metzmacher4-63/+0
metze
2010-03-29s3:rpc_client: remove unused code, we handle transport failures in the ↵Stefan Metzmacher1-14/+0
transport layer now metze
2010-03-29s3:winbindd_reconnect: don't only reconnect on NT_STATUS_UNSUCCESSFULStefan Metzmacher1-14/+59
metze
2010-03-29s3:winbindd_cm: invalidate connection if cm_connect_netlogon() failsStefan Metzmacher1-2/+2
metze
2010-03-29s3:winbindd: consistently use TALLOC_FREE(conn->foo_pipe) is we create a new ↵Stefan Metzmacher1-0/+5
connection metze
2010-03-29s3:winbindd_cm: use rpccli_is_connected() helper functionStefan Metzmacher1-4/+4
metze
2010-03-29s3:winbindd_cm: use cli_state_is_connected() helper functionStefan Metzmacher1-14/+4
metze
2010-03-29s3:rpc_client: return at least 10 sec as old timeout in rpccli_set_timeout() ↵Stefan Metzmacher1-3/+12
instead of 0 metze
2010-03-29s3:rpc_client: add set_timeout hook to rpc_cli_transportStefan Metzmacher5-20/+64
metze
2010-03-29s3:rpc_client: add rpccli_is_connected()Stefan Metzmacher6-18/+175
metze
2010-03-29s3:rpc_client: don't mix layers and keep a reference to cli_state in the callerStefan Metzmacher1-8/+17
We should not rely on the backend to have a reference to the cli_state. This will make it possible for the backend to set its cli_state reference to NULL, when the transport is dead. metze
2010-03-29s3:rpc_transport_np: add comment about bad usage in a destructorStefan Metzmacher1-1/+2
metze
2010-03-29s3:rpc_transport_np: use cli_state_is_connected() helperStefan Metzmacher1-1/+1
metze