summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2010-10-01s3:events: Call all ready fd event handlers on each iteration of the main loopSteven Danneman5-33/+38
Previously, only one fd handler was being called per main message loop in all smbd child processes. In the case where multiple fds are available for reading the fd corresponding to the event closest to the beginning of the event list would be run. Obviously this is arbitrary and could cause unfairness. Usually, the first event fd is the network socket, meaning heavy load of client requests can starve out other fd events such as oplock or notify upcalls from the kernel. In this patch, I have changed the behavior of run_events() to unset any fd that it has already called a handler function, as well as decrement the number of fds that were returned from select(). This allows the caller of run_events() to iterate it, until all available fds have been handled. I then changed the main loop in smbd child processes to iterate run_events(). This way, all available fds are handled on each wake of select, while still checking for timed or signalled events between each handler function call. I also added an explicit check for EINTR from select(), which previously was masked by the fact that run_events() would handle any signal event before the return code was checked. This required a signature change to run_events() but all other callers should have no change in their behavior. I also fixed a bug in run_events() where it could be called with a selrtn value of -1, doing unecessary looping through the fd_event list when no fds were available. Also, remove the temporary echo handler hack, as all fds should be treated fairly now.
2010-10-01testprogs: print architecture used in win32 spoolss testsuite.Günther Deschner1-5/+9
Guenther
2010-10-01s3-dcerpc: no point for printing NDR twice for internal pipes in log level 10.Günther Deschner1-1/+1
Guenther
2010-10-01samba: share readline wrappers among all buildsystems.Günther Deschner21-35/+35
Guenther
2010-10-01s3-readline: move cmd_history to smbclient, the only user.Günther Deschner3-22/+20
Guenther
2010-10-01samba: share select wrappers.Günther Deschner19-65/+60
Guenther
2010-10-01s4-auth: fixed a vagrind error when creating keytabsAndrew Tridgell1-0/+3
Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2010-10-01autobuild: revert a bit more of the subunit changesAndrew Tridgell1-2/+2
still not quite right - we can enable this again once its fully tested
2010-10-01autobuild: disable the subuit changes for now - they break error checkingAndrew Tridgell1-2/+2
2010-10-01autobuild: fixed the tuples in the retry_taskAndrew Tridgell1-2/+3
Pair-Programmed-With: Jelmer Vernooij <jelmer@samba.org> Autobuild-User: Andrew Tridgell <tridge@samba.org> Autobuild-Date: Fri Oct 1 18:22:22 UTC 2010 on sn-devel-104
2010-10-01autobuild: cwd is needed on all command typesAndrew Tridgell1-2/+2
2010-10-01s4-rpmd: fixed a use after realloc bugAndrew Tridgell1-2/+8
we could use old_el after the base message had been re allocated, due to adding timestamps. We need to re-find the element before using it Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2010-10-01s4-dsdb: fail the transaction instead of asserting on errorAndrew Tridgell1-2/+10
It is more useful to fail the transaction and give the user an error message than to assert when we have an error in the repl_meta_data module Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2010-10-01autobuild: Avoid unnecessary chdir() calls.Jelmer Vernooij1-11/+7
2010-10-01autobuild: Output test results in subunit.Jelmer Vernooij1-3/+5
2010-10-01s3-selftest: fix prefix in subunit output.Jelmer Vernooij1-3/+3
2010-10-01s3: Add subunit-test target.Jelmer Vernooij2-1/+7
2010-10-01s3-configure: Require at least version 1.2.6 of external TDB, which hasJelmer Vernooij1-1/+1
TDB_INCOMPATIBLE_HASH.
2010-10-01s4: Add 'subunit-test' make target.Jelmer Vernooij1-0/+3
2010-10-01autobuild: Provide more information about build sequence, stage name and ↵Jelmer Vernooij1-51/+54
output mime type (all plain text for now).
2010-10-01autobuild: Add --always-email option.Jelmer Vernooij1-0/+33
2010-10-01autobuild-remote: Support autobuild.py rather than land.py.Jelmer Vernooij2-50/+70
Autobuild-User: Jelmer Vernooij <jelmer@samba.org> Autobuild-Date: Fri Oct 1 09:46:37 UTC 2010 on sn-devel-104
2010-10-01Remove land.py - it's been obsoleted by autobuild.py.Jelmer Vernooij1-118/+0
2010-10-01autobuild: Add --daemon option.Jelmer Vernooij1-0/+31
2010-10-01autobuild: Remove autogen step for projects that have checked in configure.Jelmer Vernooij1-6/+3
2010-10-01autobuild: Simplify find_git_root.Jelmer Vernooij1-11/+6
2010-10-01pidl: Fix handling of typedefs of typedefs.Jelmer Vernooij1-3/+4
2010-10-01s3-spoolss: fix do_drv_upgrade_printer() which must have been broken since theGünther Deschner1-3/+6
days we moved away from fstrings. Guenther
2010-10-01s3-net: better handle obscure 0x80070002 error reply when trying to update anGünther Deschner1-3/+7
not yet published printer. Guenther
2010-10-01s3-net: make sure we dont crash when publishing a single printer.Günther Deschner1-0/+5
Guenther
2010-10-01s3-spoolss: make sure we dont crash on NULL setprinter level2 elements as ↵Günther Deschner1-7/+13
seen from win7. Guenther
2010-10-01s3-spoolss: dont overwrite location change notify.Günther Deschner3-3/+19
Guenther
2010-10-01s3-selftest: added samba3.posix_s3.rpc.spoolss.printer to knownfailAndrew Tridgell1-0/+1
this fails intermittently on sn-devel. Guenther suggested adding it to knownfail Autobuild-User: Andrew Tridgell <tridge@samba.org> Autobuild-Date: Fri Oct 1 04:37:36 UTC 2010 on sn-devel-104
2010-09-30heimdal: added verbose logging of hemimdal crypto errorsAndrew Bartlett1-2/+15
2010-09-30autobuild: fixed the --tail option for new log locationsAndrew Tridgell1-1/+1
2010-09-30s4-rodc: don't set SPECIAL_SECRET_PROCESSING on EXOP_REPL_SECRETAndrew Tridgell1-0/+3
otherwise we don't get the secrets! Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2010-09-30s4-spn: don't try and send an empty SPN listAndrew Tridgell1-0/+2
Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2010-10-01selftest: Let selftest provide the tempdir, rather than creating it as ↵Jelmer Vernooij3-6/+6
sideeffect of tests.py.
2010-09-30selftest: fixed a selftest error on snAndrew Tridgell1-1/+1
Pair-Programmed-With: Jelmer Vernooij <jelmer@samba.org>
2010-10-01delete_object: Remove unnecessary pass calls.Jelmer Vernooij1-7/+0
2010-10-01s4-selftest: Remove unnecessary PYTHONPATH overrides.Jelmer Vernooij1-6/+6
2010-10-01s4-selftest: Normalize paths.Jelmer Vernooij1-5/+5
2010-10-01s4-selftest: Finish conversion of selftest.sh to Python.Jelmer Vernooij2-104/+104
2010-10-01s4-selftest: Convert tests.sh to Python.Jelmer Vernooij2-544/+510
2010-09-30autobuild: push of ref/notes/commits isn't allowed in masterAndrew Tridgell1-1/+3
metze may enable this later Autobuild-User: Andrew Tridgell <tridge@samba.org> Autobuild-Date: Thu Sep 30 22:25:02 UTC 2010 on sn-devel-104
2010-09-30s4-provision: wipe the old keytabs when provisioningAndrew Tridgell2-7/+29
Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2010-09-30s4-rodc: fixed the keyVersionNumber on the RODC account in secrets.keytabAndrew Tridgell1-2/+5
we need to fetch the msDS-keyVersionNumber from the writeable DC Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2010-09-30s4-drs: put the GCSPN flag into the repsTo if requestedAndrew Tridgell2-0/+8
Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2010-09-30s4-libnet: wipe the old keytab when exportingAndrew Tridgell1-0/+2
this prevents confusion with old keytab entries Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
2010-09-30s4-dsdb: silence the domainFunctionality not setup warningAndrew Tridgell1-1/+2