1 %define with_systemd %{?_with_systemd: 1} %{?!_with_systemd: 0}
2 %define initdir %{_sysconfdir}/init.d
6 Packager: Samba Team <samba@samba.org>
10 License: GNU GPL version 3
11 Group: System Environment/Daemons
12 URL: http://ctdb.samba.org/
14 Source: ctdb-%{version}.tar.gz
17 Requires: coreutils, sed, gawk, iptables, iproute, procps, ethtool, sudo
18 # Commands - package name might vary
19 Requires: /usr/bin/killall, /bin/kill, /bin/netstat
21 Provides: ctdb = %{version}
24 BuildRoot: %{_tmppath}/%{name}-%{version}-root
26 # Allow build with system libraries
27 # To enable, run rpmbuild with,
28 # "--with system_talloc"
30 # "--with system_tevent"
31 %define with_included_talloc %{?_with_system_talloc: 0} %{?!_with_system_talloc: 1}
32 %define with_included_tdb %{?_with_system_tdb: 0} %{?!_with_system_tdb: 1}
33 %define with_included_tevent %{?_with_system_tevent: 0} %{?!_with_system_tevent: 1}
35 # Required minimum library versions when building with system libraries
36 %define libtalloc_version 2.0.8
37 %define libtdb_version 1.2.11
38 %define libtevent_version 0.9.18
40 %if ! %with_included_talloc
41 BuildRequires: libtalloc-devel >= %{libtalloc_version}
42 Requires: libtalloc >= %{libtalloc_version}
44 %if ! %with_included_tdb
45 BuildRequires: libtdb-devel >= %{libtdb_version}
46 Requires: libtdb >= %{libtdb_version}
48 %if ! %with_included_tevent
49 BuildRequires: libtevent-devel >= %{libtevent_version}
50 Requires: libtevent >= %{libtevent_version}
53 # To build the ctdb-pcp-pmda package, run rpmbuild with "--with pmda"
54 %define with_pcp_pmda %{?_with_pmda: 1} %{?!_with_pmda: 0}
56 BuildRequires: pcp-libs-devel
60 BuildRequires: systemd-units
64 ctdb is the clustered database used by samba
66 #######################################################################
72 # setup the init script and sysconfig file
73 %setup -T -D -n ctdb-%{version} -q
78 if ccache -h >/dev/null 2>&1 ; then
86 ## always run autogen.sh
89 CFLAGS="$RPM_OPT_FLAGS $EXTRA -D_GNU_SOURCE" ./configure \
90 %if %with_included_talloc
91 --with-included-talloc \
93 %if %with_included_tdb
96 %if %with_included_tevent
97 --with-included-tevent \
102 --prefix=%{_prefix} \
103 --sysconfdir=%{_sysconfdir} \
104 --mandir=%{_mandir} \
105 --localstatedir="/var"
107 make docdir=%{_docdir} showflags
108 make docdir=%{_docdir}
111 # Clean up in case there is trash left from a previous build
112 rm -rf $RPM_BUILD_ROOT
114 # Create the target build directory hierarchy
115 mkdir -p $RPM_BUILD_ROOT%{_sysconfdir}/sysconfig
116 mkdir -p $RPM_BUILD_ROOT%{_sysconfdir}/sudoers.d
118 make DESTDIR=$RPM_BUILD_ROOT docdir=%{_docdir} install install_tests
120 install -m644 config/ctdb.sysconfig $RPM_BUILD_ROOT%{_sysconfdir}/sysconfig/ctdb
123 mkdir -p $RPM_BUILD_ROOT%{_unitdir}
124 install -m 755 config/ctdb.service $RPM_BUILD_ROOT%{_unitdir}
126 mkdir -p $RPM_BUILD_ROOT%{initdir}
127 install -m755 config/ctdb.init $RPM_BUILD_ROOT%{initdir}/ctdb
130 cp config/events.d/README README.eventscripts
131 cp config/notify.d.README README.notify.d
133 # Remove "*.old" files
134 find $RPM_BUILD_ROOT -name "*.old" -exec rm -f {} \;
137 rm -rf $RPM_BUILD_ROOT
140 #######################################################################
142 #######################################################################
145 %defattr(-,root,root)
147 %config(noreplace) %{_sysconfdir}/sysconfig/ctdb
148 %config(noreplace) %{_sysconfdir}/ctdb/notify.sh
149 %config(noreplace) %{_sysconfdir}/ctdb/debug-hung-script.sh
150 %config(noreplace) %{_sysconfdir}/ctdb/ctdb-crash-cleanup.sh
151 %config(noreplace) %{_sysconfdir}/ctdb/gcore_trace.sh
152 %config(noreplace) %{_sysconfdir}/ctdb/debug_locks.sh
155 %{_unitdir}/ctdb.service
157 %attr(755,root,root) %{initdir}/ctdb
160 %attr(755,root,root) %{_sysconfdir}/ctdb/notify.d
162 %doc README COPYING NEWS
163 %doc README.eventscripts README.notify.d
164 %doc doc/recovery-process.txt
167 %{_sysconfdir}/sudoers.d/ctdb
168 %{_sysconfdir}/ctdb/functions
169 %{_sysconfdir}/ctdb/events.d/00.ctdb
170 %{_sysconfdir}/ctdb/events.d/01.reclock
171 %{_sysconfdir}/ctdb/events.d/10.interface
172 %{_sysconfdir}/ctdb/events.d/13.per_ip_routing
173 %{_sysconfdir}/ctdb/events.d/11.natgw
174 %{_sysconfdir}/ctdb/events.d/11.routing
175 %{_sysconfdir}/ctdb/events.d/20.multipathd
176 %{_sysconfdir}/ctdb/events.d/31.clamd
177 %{_sysconfdir}/ctdb/events.d/40.fs_use
178 %{_sysconfdir}/ctdb/events.d/40.vsftpd
179 %{_sysconfdir}/ctdb/events.d/41.httpd
180 %{_sysconfdir}/ctdb/events.d/49.winbind
181 %{_sysconfdir}/ctdb/events.d/50.samba
182 %{_sysconfdir}/ctdb/events.d/60.nfs
183 %{_sysconfdir}/ctdb/events.d/60.ganesha
184 %{_sysconfdir}/ctdb/events.d/62.cnfs
185 %{_sysconfdir}/ctdb/events.d/70.iscsi
186 %{_sysconfdir}/ctdb/events.d/91.lvs
187 %{_sysconfdir}/ctdb/events.d/99.timeout
188 %config(noreplace) %{_sysconfdir}/ctdb/nfs-rpc-checks.d/10.statd.check
189 %config(noreplace) %{_sysconfdir}/ctdb/nfs-rpc-checks.d/20.nfsd.check
190 %config(noreplace) %{_sysconfdir}/ctdb/nfs-rpc-checks.d/30.lockd.check
191 %config(noreplace) %{_sysconfdir}/ctdb/nfs-rpc-checks.d/40.mountd.check
192 %config(noreplace) %{_sysconfdir}/ctdb/nfs-rpc-checks.d/50.rquotad.check
193 %{_sysconfdir}/ctdb/statd-callout
195 %{_sbindir}/ctdbd_wrapper
197 %{_bindir}/ctdb_lock_helper
201 %{_bindir}/ctdb_diagnostics
203 %{_mandir}/man1/ctdb.1.gz
204 %{_mandir}/man1/ctdbd.1.gz
205 %{_mandir}/man1/ctdbd_wrapper.1.gz
206 %{_mandir}/man1/onnode.1.gz
207 %{_mandir}/man1/ltdbtool.1.gz
208 %{_mandir}/man1/ping_pong.1.gz
209 %{_mandir}/man5/ctdbd.conf.5.gz
210 %{_mandir}/man7/ctdb.7.gz
211 %{_mandir}/man7/ctdb-tunables.7.gz
212 %{_libdir}/pkgconfig/ctdb.pc
216 Summary: CTDB development libraries
217 Group: Development/Libraries
220 development libraries for ctdb
223 %defattr(-,root,root)
224 %{_includedir}/ctdb.h
225 %{_includedir}/ctdb_client.h
226 %{_includedir}/ctdb_protocol.h
227 %{_includedir}/ctdb_private.h
228 %{_includedir}/ctdb_typesafe_cb.h
231 Summary: CTDB test suite
232 Group: Development/Tools
233 Requires: ctdb = %{version}
240 %defattr(-,root,root)
241 %dir %{_datadir}/%{name}-tests
242 %{_datadir}/%{name}-tests/*
243 %dir %{_libdir}/%{name}-tests
244 %{_libdir}/%{name}-tests/*
245 %{_bindir}/ctdb_run_tests
246 %{_bindir}/ctdb_run_cluster_tests
252 Summary: CTDB PCP pmda support
253 Group: Development/Tools
254 Requires: ctdb = %{version}
257 %description pcp-pmda
258 Performance Co-Pilot (PCP) support for CTDB
261 %dir /var/lib/pcp/pmdas/ctdb
262 /var/lib/pcp/pmdas/ctdb/*
269 * Thu Mar 1 2012 : Version 1.13
270 - This is the new stable branch for modern features for ctdb.
271 Main new features are performance/scaling improvements for
272 concurrnet fetch and fetch_lock operations.
273 * Tue Nov 8 2011 : Version 1.12
274 - Add new tunable : AllowClientDBAttach that can be used to stop
275 client db access during maintenance operations
276 - Updated logging for interfaces that are missing or dont exist but are
277 configured to be used.
278 - Add timeout argument to ctdb_cmdline_client
280 - Initial support for 'readonly' delegations for ctdb databases
281 This will when finished greatly improve performance for contended hot
282 records that are used for just read-access.
283 - New 'ctdb cattdb' command
284 - Massive updates to tests and eventscripts
285 - LCP2 ip allocation algorithm
286 - Record Fetch collapse. Collapse multiple fetch-lock requests from cients
287 to a single network fetch and defer other concurrent requests until the
288 initial fetch completes, and then service the deferred calls locally.
289 This will greatly improve performance for contended hot records
290 where clients request write-locks.
291 * Thu Sep 1 2011 : Version 1.11
293 * Tue May 25 2010 : Version 1.10
295 * Tue May 25 2010 : Version 1.9
297 * Wed Mar 25 2010 : Version 1.0.114
298 - Lots of changes from Metze
299 * Wed Jan 13 2010 : Version 1.0.113
300 - Incorrect use of dup2() could cause ctdb to spin eating 100% cpu.
301 * Tue Jan 12 2010 : Version 1.0.112
302 - Revert the use of wbinfo --ping-dc as it is proving too unreliable.
303 - Minor testsuite changes.
304 * Fri Dec 18 2009 : Version 1.0.111
305 - Fix a logging bug when an eventscript is aborted that could cause a crash.
306 - Add back cb_status that was lost in a previous commit.
307 * Fri Dec 18 2009 : Version 1.0.110
308 - Metxe: fix for filedescriptor leak in the new eventscript code.
309 - Rusty: fix for a crash bug in the eventscript code.
310 * Thu Dec 17 2009 : Version 1.0.109
311 - Massive eventscript updates. (bz58828)
312 - Nice the daemon instead of using realtime scheduler, also use mlockall() to
313 reduce the risk of blockign due to paging.
314 - Workarounds for valgrind when forking once for each script. Valgrind consumes
315 massive cpu when terminating the scripts on virtual systems.
316 - Sync the tdb library with upstream, and use the new TDB_DISALLOW_NESTING flag.
317 - Add new command "ctdb dumpdbbackup"
318 - Start using the new tdb check framework to validate tdb files upon startup.
319 - A new framework where we can control health for individual tdb databases.
320 - Fix a crash bug in the logging code.
321 - New transaction code for persistent databases.
322 - Various other smaller fixes.
323 * Mon Dec 7 2009 : Version 1.0.108
324 - Transaction updates from Michael Adam.
325 - Use the new wbinfo --ping-dc instead of -p in the eventscript for samba
326 the check if winbindd is ok.
327 - Add a better "process-exist" for samba so it will automatically
328 reap smbd's on stopped and banned nodes to reclaim subrecords.
329 This will be done a bit differently in the next release.
330 - Use a statically allocated buffer for the 'first-time' capture buffer
331 to reduce the pressure on malloc/free.
332 * Wed Dec 2 2009 : Version 1.0.107
333 - fix for rusty to solve a double-free that can happen when there are
334 multiple packets queued and the connection is destroyed before
335 all packets are processed.
336 * Tue Dec 1 2009 : Version 1.0.106
337 - Buildscript changes from Michael Adam
338 - Dont do a full recovery when there is a mismatch detected for ip addresses,
339 just do a less disruptive ip-reallocation
340 - When starting ctdbd, wait until all initial recoveries have finished
341 before we issue the "startup" event.
342 So dont start services or monitoring until the cluster has
344 - Major eventscript overhaul by Ronnie, Rusty and Martins and fixes of a few
346 * Thu Nov 19 2009 : Version 1.0.105
347 - Fix a bug where we could SEGV if multiple concurrent "ctdb eventscript ..."
348 are used and some of them block.
349 - Monitor the daemon from the syslog child process so we shutdown cleanly when
350 the main daemon terminates.
351 - Add a 500k line ringbuffer in memory where all log messages are stored.
352 - Add a "ctdb getlog <level>" command to pull log messages from the in memory
354 - From martin : fixes to cifs and nfs autotests
355 - from michael a : fix a bashism in 11.natgw
356 * Fri Nov 6 2009 : Version 1.0.104
357 - Suggestion from Metze, we can now use killtcp to kill local connections
358 for nfs so change the killtcp script to kill both directions of an NFS
360 We used to deliberately only kill one direction in these cases due to
362 - Suggestion from christian Ambach, when using natgw, try to avoid using a
363 UNHEALTHY node as the natgw master.
364 - From Michael Adam: Fix a SEGV bug in the recent change to the eventscripts
365 to allow the timeout to apply to each individual script.
366 - fix a talloc bug in teh vacuuming code that produced nasty valgrind
368 - From Rusty: Set up ulimit to create core files for ctdb, and spawned
369 processes by default. This is useful for debugging and testing but can be
370 disabled by setting CTDB_SUPRESS_COREFILE=yes in the sysconfig file.
371 - Remove the wbinfo -t check from the startup check that winbindd is happy.
372 - Enhance the test for bond devices so we also check if the sysadmin have
373 disabled all slave devices using "ifdown".
374 * Tue Nov 3 2009 : Version 1.0.103
375 - Dont use vacuuming on persistent databases
376 - Michael A : transaction updates to persistent databases
377 - Dont activate service automatically when installing the RPM. Leave this to the admin.
378 - Create a child process to send all log messages to, to prevent a hung/slow syslogd
379 from blocking the main daemon. In this case, discard log messages instead and let the child
381 - Michael A: updates to log messages
382 * Thu Oct 29 2009 : Version 1.0.102
383 - Wolfgang: fix for the vacuuming code
384 - Wolfgang: stronger tests for persistent database filename tests
385 - Improve the log message when we refuse to startup since wbinfo -t fails
386 to make it easier to spot in the log.
387 - Update the uptime command output and the man page to indicate that
388 "time since last ..." if from either the last recovery OR the last failover
389 - Michael A: transaction updates
390 * Wed Oct 28 2009 : Version 1.0.101
391 - create a separate context for non-monitoring events so they dont interfere with the monitor event
392 - make sure to return status 0 in teh callback when we abort an event
393 * Wed Oct 28 2009 : Version 1.0.100
394 - Change eventscript handling to allow EventScriptTimeout for each individual script instead of for all scripts as a whole.
395 - Enhanced logging from the eventscripts, log the name and the duration for each script as it finishes.
396 - Add a check to use wbinfo -t for the startup event of samba
397 - TEMP: allow clients to attach to databases even when teh node is in recovery mode
398 - dont run the monitor event as frequently after an event has failed
399 - DEBUG: in the eventloops, check the local time and warn if the time changes backward or rapidly forward
400 - From Metze, fix a bug where recovery master becoming unhealthy did not trigger an ip failover.
401 - Disable the multipath script by default
402 - Automatically re-activate the reclock checking if the reclock file is specified at runtime. Update manpage to reflect this.
403 - Add a mechanism where samba can register a SRVID and if samba unexpectedly disconnects, a message will be broadcasted to all other samba daemons.
404 - Log the pstree on hung scripts to a file in /tmp isntead of /var/log/messages
405 - change ban count before unhealthy/banned to 10
406 * Thu Oct 22 2009 : Version 1.0.99
407 - Fix a SEGV in the new db priority code.
408 - From Wolfgang : eliminate a ctdb_fatal() if there is a dmaster violation detected.
409 - During testing we often add/delete eventscripts at runtime. This could cause an eventscript to fail and mark the node unhealthy if an eventscript was deleted while we were listing the names. Handle the errorcode and make sure the node does not becomne unhealthy in this case.
410 - Lower the debuglevel for the messages when ctdb creates a filedescruiptor so we dont spam the logs with these messages.
411 - Dont have the RPM automatically restart ctdb
412 - Volker : add a missing transaction_cancel() in the handling of persistent databases
413 - Treat interfaces with the anme ethX* as bond devices in 10.interfaces so we do the correct test for if they are up or not.
414 * Tue Oct 20 2009 : Version 1.0.98
415 - Fix for the vacuuming database from Wolfgang M
416 - Create a directory where the test framework can put temporary overrides
417 to variables and functions.
418 - Wait a lot longer before shutting down the node when the reclock file
419 is incorrectly configured, and log where it is configured.
420 - Try to avoid running the "monitor" event when databases are frozen.
421 - Add logging for every time we create a filedescriptor so we can trap
423 * Thu Oct 14 2009 : Version 1.0.97
424 - From martins : update onnode.
425 Update onnode to allow specifying an alternative nodes file from
426 the command line and also to be able to specify hostnames on the
428 onnode host1,host2,...
429 * Tue Oct 13 2009 : Version 1.0.96
430 - Add more debugging output when eventscripts have trouble. Print a
431 "pstree -p" to the log when scripts have hung.
432 - Update the initscript, only print the "No reclock file used" warning
433 when we do "service ctdb start", dont also print them for all other
435 - When changing between unhealthy/healthy state, push a request to the
436 recovery master to perform an ip reallocation instead of waiting for the
437 recovery master to pull and check the state change.
438 - Fix a bug in the new db-priority handling where a pre-.95 recovery master
439 could no longer lock the databases on a post-.95 daemon.
440 - Always create the nfs state directories during the "monitor" event.
441 This makes it easier to configure and enable nfs at runtime.
442 - From Volker, forward-port a simper deadlock avoiding patch from the 1.0.82
443 branch. This is a simpler versionof the "db priority lock order" patch
444 that went into 1.0.95, and will be kept for a few versions until samba
445 has been updated to use the functionality from 1.0.95.
446 * Mon Oct 12 2009 : Version 1.0.95
447 - Add database priorities. Allow samba to set the priority of databases
448 and lock the databases in priority order during recovery
449 to avoid a deadlock when samba locks one database then blocks indefinitely
450 while waiting for the second databaso to become locked.
451 - Be aggressive and ban nodes where the recovery transaction start call
453 * Thu Oct 10 2009 : Version 1.0.94
454 - Be very aggressive and quickly ban nodes that can not freeze their databases
455 * Tue Oct 8 2009 : Version 1.0.93
456 - When adding an ip, make sure to update this assignment on all nodes
457 so it wont show up as -1 on other nodes.
458 - When adding an ip and immediately deleting it, it was possible that
459 the daemon would crash accessing already freed memory.
460 Readjust the memory hierarchy so the destructors are called in the right order.
461 - Add a handshake to the recovery daemon to eliminate some rare cases where
462 addip/delip might cause a recovery to occur.
463 - updated onnode documenation from Martin S
464 - Updates to the natgw eventscript to allow disabling natgw at runtime
465 * Fri Oct 2 2009 : Version 1.0.92
466 - Test updates and merge from martin
467 - Add notification for "startup"
468 - Add documentation for notification
469 - from martin, a fix for restarting vsftpd in the eventscript
470 * Tue Sep 29 2009 : Version 1.0.91
471 - New vacuum and repack design from Wolgang Mueller.
472 - Add a new eventscript 01.reclock that will first mark a node unhealthy and later ban the node if the reclock file can not be accessed.
473 - Add machinereadable output to the ctdb getreclock command
474 - merge transaction updates from Michael Adam
475 - In the new banning code, reset the culprit count to 0 for all nodes that could successfully compelte a full recovery.
476 - dont mark the recovery master as a ban culprit because a node in the cluster needs a recovery. this happens naturally when using ctdb recover command so dont make this cause a node to be banned.
477 * Sat Sep 12 2009 : Version 1.0.90
478 - Be more forgiving for eventscripts that hang during startup
479 - Fix for a banning bug in the new banning logic
480 * Thu Sep 3 2009 : Version 1.0.89
481 - Make it possible to manage winbind independently of samba.
482 - Add new prototype banning code
483 - Overwrite the vsftpd state file instead of appending. This eliminates
484 annoying errors in the log.
485 - Redirect some iptables commands to dev null
486 - From Michael A, explicitely set the broadcast when we takeover a public ip
487 - Remove a reclock file check we no longer need
488 - Skip any persistent database files ending in .bak
489 * Mon Aug 17 2009 : Version 1.0.88
490 - Add a new state for eventscripts : DISABLED.
491 Add two new commands "ctdb enablescript/disablescript" to enable/disable
492 eventscripts at runtime.
493 - Bugfixes for TDB from rusty.
494 - Merge/Port changes from upstream TDB library by rusty.
495 - Additional new tests from MartinS. Tests for stop/continue.
496 - Initial patch to rework vacuuming/repacking process from Wolfgang Mueller.
497 - Updates from Michael Adam for persistent writes.
498 - Updates from MartinS to handle the new STOPPED bit in the test framework.
499 - Make it possible to enable/disable the RECMASTER and LMASTER roles
500 at runtime. Add two new commands
501 "ctdb setlmasterrole/setrecmasterrole on/off"
502 - Make it possible to enable/disable the natgw feature at runtime. Add
503 the command "ctdb setnatgwstate on/off"
504 * Fri Jul 17 2009 : Version 1.0.87
505 - Add a new event "stopped" that is called when a node is stopped.
506 - Documentation of the STOPPED flag and the stop/continue commands
507 - Make it possible to start a node in STOPPED mode.
508 - Add a new node flag : STOPPED and commands "ctdb stop" "ctdb continue"
509 These commands are similar to "diasble/enable" but will also remove the node from the vnnmap, while disable only fails all ip addresses over.
510 - tests for NFS , CIFS by martins
511 - major updates to the init script by martins
512 - Send gratious arps with a 1.1 second stride instead of a 1 second stride to workaround interesting "features" of common linux stacks.
513 - Various test enhancements from martins:
514 - additional other tests
515 - add tests for grat arp generation, ping during failover, ssh and failover
516 - New/updated tcp tickle tests and supprot functions
517 - provide better debugging when a test fails
518 - make ctdbd restarts more reliable in the tests
519 - update the "wait bar" to make the wait progress in tests more obvious
521 - when dispatching a message to a handler, make the message a real talloc object so that we can reparent the object in the tallic hierarchy.
522 - document the ipreallocate command
523 - Updates to enable/disable to use the ipreallocate command to block until the following ipreallocation has completed.
524 - Update the main daemon and the tools to allow debug level to be a string instead of an integer.
525 - Update the sysconfig file to show using string literals instead of numeric values for the debuglevels used.
526 - If no debuglevel is specific, make "ctdb setdebug" show the available options.
527 - When trying to allocate network packets, add explicit checks if the network transport has been shutdown before trying and failing, to make log messages easier to read. Add this extra check and logging to every plave packets are allocated.
528 * Tue Jun 30 2009 : Version 1.0.86
529 - Do not access the reclock at all if VerifyRecoveryLock is zero, not even try to probe it.
530 - Allow setting the reclock file as "", which means that no reclock file at all should be used.
531 - Document that a reclock file is no longer required, but that it is dangerous.
532 - Add a control that can be used to set/clear/change the reclock file in the daemon during runtime.
533 - Update the recovery daemon to poll whether a reclock file should be sued and if so which file at runtime in each monitoring cycle.
534 - Automatically disable VerifyRecoveryLock everytime a user changes the location of the reclock file.
535 - do not allow the VerifyRecoveryLock to be set using ctdb setvar if there is no recovery lock file specified.
536 - Add two commands "ctdb getreclock" and "ctdb setreclock" to modify the reclock file.
537 * Tue Jun 23 2009 : Version 1.0.85
538 - From William Jojo : Dont use getopt on AIX
539 - Make it possible to use "ctdb listnodes" also when the daemon is not running
540 - Provide machinereadable output to "ctdb listnodes"
541 - Dont list DELETED nodes in the ctdb listnodes output
542 - Try to avoid causing a recovery for the average case when adding/deleting/moving an ip
543 - When banning a node, drop the IPs on that node only and not all nodes.
544 - Add tests for NFS and CIFS tickles
545 - Rename 99.routing to 11.routing so it executes before NFS and LVS scripts
546 - Increase the default timeout before we deem an unresponsive recovery daemon hung and shutdown
547 - Reduce the reclock timout to 5 seconds
548 - Spawn a child process in the recovery daemon ot check the reclock file to
549 avoid blocking the process if the underlying filesystem is unresponsive
550 - fix for filedescriptor leak when a child process timesout
551 - Dont log errors if waitpid() returns -1
552 - Onnode updates by Martins
553 - Test and initscript cleanups from Martin S
554 * Tue Jun 2 2009 : Version 1.0.84
555 - Fix a bug in onnode that could not handle dead nodes
556 * Tue Jun 2 2009 : Version 1.0.83
557 - Document how to remove a ndoe from a running cluster.
558 - Hide all deleted nodes from ctdb output.
559 - Lower the loglevel on some eventscript related items
560 - Dont queue packets to deleted nodes
561 - When building initial vnnmap, ignode any nonexisting nodes
562 - Add a new nodestate : DELETED that is used when deleting a node from an
564 - dont remove the ctdb socket when shutting down. This prevents a race in the
565 initscripts when restarting ctdb quickly after stopping it.
566 - TDB nesting reworked.
567 - Remove obsolete ipmux
568 - From Flavio Carmo Junior: Add eventscript and documentation for ClamAV antivirus engine
569 - From Sumit Bose: fix the regex in the test to handle the new ctdb
570 statistics output that was recently added.
571 - change the socket type we use for grauitious arps from the obsolete
572 AF_INET/SOCK_PACKET to instead use PF_PACKET/SOCK_RAW.
573 - Check return codes for some functions, from Sumit Bose, based on codereview by Jim Meyering.
574 - Sumit Bose: Remove structure memeber node_list_file that is no longer used.
575 - Sumit Bose: fix configure warning for netfilter.h
576 - Updates to the webpages by Volker.
577 - Remove error messages about missing /var/log/log.ctdb file from ctdb_diagnostics.sh from christian Ambach
578 - Additional error logs if hte eventscript switching from dameon to client mode fails.
579 - track how long it takes for ctdbd and the recovery daemon to perform the rec-lock fcntl() lock attemt and show this in the ctdb statistics output.
580 * Thu May 14 2009 : Version 1.0.82
581 - Update the "ctdb lvsmaster" command to return -1 on error.
582 - Add a -Y flag to "ctdb lvsmaster"
583 - RHEL5 apache leaks semaphores when crashing. Add semaphore cleanup to the
584 41.httpd eventscript and try to restart apache when it has crashed.
585 - Fixes to some tests
586 - Add a -o option to "onnode" which will redirect all stdout to a file for
588 - Add a natgw and a lvs node specifier to onnode so that we can use
590 - Assign the natgw address to lo instead of the private network so it can also
591 be used where private and public networks are the same.
592 - Add GPL boilerplates to two missing scripts.
593 - Change the natgw prefix NATGW_ to CTDB_NATGW_
594 * Fri May 8 2009 : Version 1.0.81
595 - use smbstatus -np instead of smbstatus -n in the 50.samba eventscript
596 since this avoids performing an expensive traverse on the locking and brlock
598 - make ctdb automatically terminate all traverse child processes clusterwide
599 associated to a client application that terminates before the traversal is
601 - From Sumit Bose : fixes to AC_INIT handling.
602 - From Michael Adam, add Tridge's "ping_pong" tool the the ctdb distro since
603 this is very useful for testing the backend filesystem.
604 - From Sumit bose, add support for additional 64 bit platforms.
605 - Add a link from the webpage to Michael Adams SambaXP paper on CTDB.
606 * Fri May 1 2009 : Version 1.0.80
607 - change init shutdown level to 01 for ctdb so it stops before any of the other services
608 - if we can not pull a database from a remote node during recovery, mark that node as a culprit so it becomes banned
609 - increase the loglevel when we volunteer to drop all ip addresses after beeing in recovery mode for too long. Make this timeout tuneable with "RecoveryDropAllIPs" and have it default to 60 seconds
610 - Add a new flag TDB_NO_NESTING to the tdb layer to prevent nested transactions which ctdb does not use and does not expect. Have ctdb set this flag to prevent nested transactions from occuring.
611 - dont unconditionally kill off ctdb and restrat it on "service ctdb start". Fail "service ctdb start" with an error if ctdb is already running.
612 - Add a new tunable "VerifyRecoveryLock" that can be set to 0 to prevent the main ctdb daemon to verify that the recovery master has locked the reclock file correctly before allowing it to set the recovery mode to active.
613 - fix a cosmetic bug with ctdb statistics where certain counters could become negative.
614 * Wed Apr 8 2009 : Version 1.0.79
615 - From Mathieu Parent: add a ctdb pkgconfig file
617 - add a funciton remove_ip to safely remove an ip from an interface, taking care to workaround an issue with linux alias interfaces.
618 - Update the natgw eventscript to use the safe remove_ip() function
619 - fix a bug in the eventscript child process that would cause the socket to be removed.
620 - dont verify nodemap on banned nodes during cluster monitoring
621 - Update the dodgy SeqnumInterval to have ms resolution
622 * Tue Mar 31 2009 : Version 1.0.78
623 - Add a notify mechanism so we can send snmptraps/email to external management systems when the node becomes unhealthy
624 - include 11.natgw eventscript in thew install so that the NATGW feature works
625 * Tue Mar 31 2009 : Version 1.0.77
626 - Update the 99.routing eventscript to also try to add the routes (back) during a releaseip event. Similar to the reasons why we must add addresses back during releaseip in 10.interfaces
627 * Wed Mar 24 2009 : Version 1.0.76
628 - Add a debugging command "xpnn" which can print the pnn of the node even when ctdbd is not running.
629 - Redo the NATGW implementation to allow multiple disjoing NATGW groups in the same cluster.
630 * Tue Mar 24 2009 : Version 1.0.75
631 - Various updates to LVS
632 - Fix a bug in the killtcp control where we did not set the port correctly
633 - add a new "ctdb scriptstatus" command that shows the status of the eventrscripts.
634 * Mon Mar 16 2009 : Version 1.0.74
635 - Fixes to AIX from C Cowan.
636 - Fixes to ctdb_diagnostics so we collect correct GPFS data
637 - Fixes to the net conf list command in ctdb_diagnostics
638 - Check the static-routes file IFF it exists in ctdb_diagnostics
639 * Wed Mar 4 2009 : Version 1.0.73
640 - Add possibility to disable the check of shares for NFS and Samba
641 - From Sumit Bose, fix dependencies so make -j works
642 * Wed Feb 18 2009 : Version 1.0.72
643 - Updates to test scripts by martin s
644 - Adding a COPYING file
645 - Use netstat to check for services and ports and fallback to netcat
646 only if netstat is unavailable.
647 * Mon Feb 2 2009 : Version 1.0.71
648 - Additional ipv6 fixes from Michael Adams
649 * Fri Jan 16 2009 : Version 1.0.70
650 - IPv6 support is completed. this is backward compatible with ipv4-only
651 systems. To use IPv6 with samba and ctdb you need current GIT of samba 3.3
652 or michael adams samba-ctdeb branch.
653 - Many enhancements to the build system and scripts to make it more SUSE
654 friendly by Michael Adams.
655 - Change of how the naming of the package is structured. We are now
656 using "1.0.70" as a release and "-1" as the revision instead of as
657 previously using "1.0" as release and ".70" as the revision.
659 * Thu Dec 18 2008 : Version 1.0.69
660 - Various fixes to scripts by M Adam
661 - Dont call ctdb_fatal() when the transport is down during shutdown
662 * Fri Dec 12 2008 : Version 1.0.68
663 - Fixes for monitoring of interfaces status from Michael Adam.
664 - Use -q instead of >/dev/null for grep to enhance readability of the
665 scripts from Michael Adam.
666 - Update to the "ctdb recover" command. This command now block until the
667 has completed. This makes it much easier to use in scripts and avoids
668 the common workaround :
670 ... loop while waiting for recovery completes ...
672 - Add a CTDB_TIMEOUT variable. If set, this variable provides an automatic
673 timeout for "ctdb <command>", similar to using -T <timeout>
674 - Set a unique errorcode for "ctdb <command>" when it terminates due to a
675 timeout so that scripts can distinguish between a hung command and what was
677 - Update "ctdb ban/unban" so that if the cluster is in recovery these commands
678 blocks and waits until after recovery is complete before the perform the
679 ban/unban operation. This is necessary since the recovery process can cause
680 nodes to become automatically unbanned.
681 - Update "ctdb ban/unban" to block until the recovery that will follow shortly
682 after this command has completed.
683 This makes it much easier to use in scripts and avoids the common
686 ... loop while waiting for recovery completes ...
688 - Bugfix for the new flags handling in 1.0.67. Abort and restart monitoring
689 if we failed to get proper nodemaps from a remote node instead of
690 dereferencing a null pointer.
691 - If ctdbd was explicitely started with the '--socket' argument, make
692 ctdbd automatically set CTDB_SOCKET to the specified argument.
693 This ensures that eventscripts spawned by the ctdb daemon will default to
694 using the same socket and talk to the correct daemon.
695 This primarily affects running multiple daemons on the same host and where
696 you want each instance of ctdb daemons have their eventscripts talk to the
698 - Update "ctdb ping" to return an error code if the ping fail so that it
699 can be used in scripts.
700 - Update to how to synchronize management of node flags across the cluster.
701 * Thu Dec 3 2008 : Version 1.0.67
702 - Add a document describing the recovery process.
703 - Fix a bug in "ctdb setdebug" where it would refuse to set a negative
705 - Print the list of literals for debug names if an invalid one was given
707 - Redesign how "ctdb reloadnodes" works and reduce the amont of tcp teardowns
708 used during this event.
709 - Make it possible to delete a public ip from all nodes at once using
711 * Mon Nov 24 2008 : Version 1.0.66
712 - Allow to change the recmaster even when we are not frozen.
713 - Remove two redundant SAMBA_CHECK variables from the sysconf example
714 - After a node failure it can take very long before some lock operations
715 ctdb needs to perform are allowed/works with gpfs again. Workaround this
716 by treating a hang/timeout as success.
717 - Dont override CTDB_BASE is fet in the shell already
718 - Always send keepalive packets regardless of whether the link is idle or not.
719 - Rewrite the disable/enable flag update logic to prevent a race between
720 "ctdb disable/enable" and the recovery daemon when updating the flags to
722 * Thu Nov 13 2008 : Version 1.0.65
723 - Update the sysconfig example: The default debug level is 2 (NOTICE) and not
725 - Add support for a CTDB_SOCKET environment variable for the ctdb command
726 line tool. If set, this overrides the default socket the ctdb tool will
728 - Add logging of high latency operations.
729 * Mon Oct 22 2008 : Version 1.0.64
730 - Add a context and a timed event so that once we have been in recovery for
731 too long we drop all public addresses.
732 * Mon Oct 20 2008 : Version 1.0.63
733 - Remove logging of "periodic cleanup ..." in 50.samba
734 - When we reload a nodes file, we must detect this and reload the file also
735 in the recovery daemon before we try to dereference somethoung beyond the end
737 * Thu Oct 16 2008 : Version 1.0.62
738 - Allow multiple eventscritps using the same prefix number.
739 It is undefined which order scripts with the same prefix will execute in.
740 * Wed Oct 15 2008 : Version 1.0.61
741 - Use "route add -net" instead of "ip route add" when adding routes in 99.routing
742 - lower the loglevel os several debug statements
743 - check the status returned from ctdb_ctrl_get_tickles() before we try to print them out to the screen.
744 - install a new eventscript 20.multipathd whoich can be used to monitor that multipath devices are healthy
745 * Wed Oct 15 2008 : Version 1.0.60
746 - Verify that nodes we try to ban/unban are reachable and print an error othervise.
747 - Update the client and server sides of TAKEIP/RELEASEIP/GETPUBLICIPS and GETNODEMAP to fall back to the old style ipv4-only controls if the new ipv4/ipv6 controls fail. This allows an ipv4/v6 enabled ctdb daemon to interoperate with earlier ipv4-only versions of the daemons.
748 - From Mathieu Parent : log debian systems log the package versions in ctdb diagnostics
749 - From Mathieu Parent : specify logdir location for debian (this patch was later reversed)
750 - From Michael Adams : allow # comments in nodes/public_addresses files
751 * Tue Oct 7 2008 : Version 1.0.59
752 - Updated "reloadnodes" logic. Instead of bouncing the entire tcp layer it is sufficient to just close and reopen all outgoing tcp connections.
753 - New eventscript 99.routing which can be used to re-attach routes to public interfaces after a takeip event. (routes may be deleted by the kernel when we release an ip)
754 - IDR tree fix from Jim Houston
755 - Better handling of critical events if the local clock is suddenly changed forward by a lot.
756 - Fix three slow memory leaks in the recovery daemon
757 - New ctdb command : ctdb recmaster which prints the pnn of the recmaster
758 - Onnode enhancements from Martin S : "healthy" and "connected" are now possible nodespecifiers
759 - From Martin S : doc fixes
760 - lowering some debug levels for some nonvital informational messages
761 - Make the daemon daemon monitoring stronger and allow ctdbd to detect a hung
763 - From C Cowan : patches to compile ipv6 under AIX
764 - zero out some structs to keep valgrind happy
765 * Wed Aug 27 2008 : Version 1.0.58
766 - revert the name change tcp_tcp_client back to tcp_control_tcp so
768 - Updates to the init script from Abhijith Das <adas@redhat.com>
769 * Mon Aug 25 2008 : Version 1.0.57
770 - initial support for IPv6
771 * Mon Aug 11 2008 : Version 1.0.56
772 - fix a memory leak in the recovery daemon.
773 * Mon Aug 11 2008 : Version 1.0.55
774 - Fix the releaseip message we seond to samba.
775 * Fri Aug 8 2008 : Version 1.0.54
776 - fix a looping error in the transaction code
777 - provide a more detailed error code for persistent store errors
778 so clients can make more intelligent choices on how to try to recover
779 * Thu Aug 7 2008 : Version 1.0.53
780 - Remove the reclock.pnn file it can cause gpfs to fail to umount
781 - New transaction code
782 * Mon Aug 4 2008 : Version 1.0.52
783 - Send an explicit gratious arp when starting sending the tcp tickles.
784 - When doing failover, issue a killtcp to non-NFS/non-CIFS clients
785 so that they fail quickly. NFS and CIFS already fail and recover
787 - Update the test scripts to handle CTRL-C to kill off the test.
788 * Mon Jul 28 2008 : Version 1.0.51
789 - Strip off the vlan tag from bond devices before we check in /proc
790 if the interface is up or not.
791 - Use testparm in the background in the scripts to allow probing
792 that the shares do exist.
793 - Fix a bug in the logging code to handle multiline entries better
794 - Rename private elements from private to private_data
795 * Fri Jul 18 2008 : Version 1.0.50
796 - Dont assume that just because we can establish a TCP connection
797 that we are actually talking to a functioning ctdb daemon.
798 So dont mark the node as CONNECTED just because the tcp handshake
800 - Dont try to set the recmaster to ourself during elections for those
801 cases we know this will fail. To remove some annoying benign but scary
802 looking entries from the log.
803 - Bugfix for eventsystem for signal handling that could cause a node to
805 * Thu Jul 17 2008 : Version 1.0.49
806 - Update the safe persistent update fix to work with unpatched samba
808 * Thu Jul 17 2008 : Version 1.0.48
809 - Update the spec file.
810 - Do not start new user-triggered eventscripts if we are already
811 inside recovery mode.
812 - Add two new controls to start/cancel a persistent update.
813 A client such as samba can use these to tell ctdbd that it will soon
814 be writing directly to the persistent database tdb file. So if
815 samba is -9ed before it has eitehr done the persistent_store or
816 canceled the operation, ctdb knows that the persistent databases
817 'may' be out of sync and therefore a full blown recovery is called for.
818 - Add two new options :
819 CTDB_SAMBA_SKIP_CONF_CHECK and CTDB_SAMBA_CHECK_PORTS that can be used
820 to override what checks to do when monitoring samba health.
821 We can no longer use the smbstatus, net or testparm commands to check
822 if samba or its config is healthy since these commands may block
823 indefinitely and thus can not be used in scripts.
824 * Fri Jul 11 2008 : Version 1.0.47
825 - Fix a double free bug where if a user striggered (ctdb eventscript)
826 hung and while the timeout handler was being processed a new user
827 triggered eventscript was started we would free state twice.
828 - Rewrite of onnode and associated documentation.
829 * Thu Jul 10 2008 : Version 1.0.46
830 - Document both the LVS:cingle-ip-address and the REMOTE-NODE:wan-accelerator
832 - Add commands "ctdb pnn", "ctdb lvs", "ctdb lvsmaster".
833 - LVS improvements. LVS is the single-ip-address mode for a ctdb cluster.
834 - Fixes to supress rpmlint warnings
836 - Change \s to [[:space:]] in some scripts. Not all RHEL5 packages come
837 with a egrep that handles \s even same version but different arch.
838 - Revert the change to NFS restart. CTDB should NOT attempt to restart
840 - Rewrite of the waitpid() patch to use the eventsystem for handling
842 * Tue Jul 8 2008 : Version 1.0.45
843 - Try to restart the nfs service if it has failed to respond 3 times in a row.
844 - waitpid() can block if the child does not respond promptly to SIGTERM.
845 ignore all SIGCHILD signals by setting SIGCHLD to SIG_DEF.
846 get rid of all calls to waitpid().
847 - make handling of eventscripts hanging more liberal.
848 only consider the script to have failed and making the node unhealthy
849 IF the eventscript terminated wiht an error
850 OR the eventscript hung 5 or more times in a row
851 * Mon Jul 7 2008 : Version 1.0.44
852 - Add a CTDB_VALGRIND option to /etc/sysconfig/ctdb to make it start
853 ctdb under valgrind. Logs go to /var/log/ctdb_valgrind.PID
854 - Add a hack to show the control opcode that caused uninitialized data
855 in the valgrind output by encoding the opcode as the line number.
856 - Initialize structures and allocated memory in various places in
857 ctdb to make it valgrind-clean and remove all valgrind errors/warnings.
858 - If/when we destroy a lockwait child, also make sure we cancel any pending transactions
859 - If a transaction_commit fails, delete/cancel any pending transactions and
860 return an error instead of calling ctdb_fatal()
861 - When running ctdb under valgrind, make sure we run it with --nosetsched and also
862 ensure that we do not use mem-mapped i/o when accessing the tdb's.
863 - zero out ctdb->freeze_handle when we free/destroy a freeze-child.
864 This prevent a heap corruption/ctdb crash bug that could trigger
865 if the freeze child times out.
866 - we dont need to explicitely thaw the databases from the recovery daemon
867 since this is done implicitely when we restore the recovery mode back to normal.
868 - track when we start and stop a recovery. Add the 'time it took to complete the
869 recovery' to the 'ctdb uptime' output.
870 Ensure by tracking the start/stop recovery timestamps that we do not
871 check that the ip allocation is consistend from inside the recovery daemon
872 while a different node (recovery master) is performing a recovery.
873 This prevent a race that could cause a full recovery to trigger if the
874 'ctdb disable/enable' commands took very long.
875 - The freeze child indicates to the master daemon that all databases are locked
876 by writing data to the pipe shared with the master daemon.
877 This write sometimes fail and thus the master daemon never notices that the databases
878 are locked cvausing long timeouts and extra recoveries.
879 Check that the write is successful and try the write again if it failed.
880 - In each node, verify that the recmaster have the right node flags for us
881 and force a push of our flags to the recmaster if wrong.
882 * Tue Jul 1 2008 : Version 1.0.43
883 - Updates and bugfixes to the specfile to keep rpmlint happy
884 - Force a global flags update after each recovery event.
885 - Verify that the recmaster agrees with our node flags and update the
887 - When writing back to the parent from a freeze-child across the pipe,
888 loop over the write in case the write failed with an error othervise
889 the parent will never be notified tha the child has completed the operation.
890 - Automatically thaw all databases when recmaster marks us as being in normal
891 mode instead of recovery mode.
892 * Fri Jun 13 2008 : Version 1.0.42
893 - When event scripts have hung/timedout more than EventScriptBanCount times
894 in a row the node will ban itself.
895 - Many updates to persistent write tests and the test scripts.
896 * Wed May 28 2008 : Version 1.0.41
897 - Reactivate the safe writes to persistent databases and solve the
898 locking issues. Locking issues are solved the only possible way,
899 by using a child process to do the writes. Expensive and slow but... .
900 * Tue May 27 2008 : Version 1.0.40
901 - Read the samba sysconfig file from the 50.samba eventscript
902 - Fix some emmory hierarchical bugs in the persistent write handling
903 * Thu May 22 2008 : Version 1.0.39
904 - Moved a CTDB_MANAGES_NFS, CTDB_MANAGES_ISCSI and CTDB_MANAGES_CSFTPD
905 into /etc/sysconfig/ctdb
906 - Lowered some debug messages to not fill the logfile with entries
907 that normally occur in the default configuration.
908 * Fri May 16 2008 : Version 1.0.38
909 - Add machine readable output support to "ctdb getmonmode"
910 - Lots of tweaks and enhancements if the event scripts are "slow"
911 - Merge from tridge: an attempt to break the chicken-and-egg deadlock that
912 net conf introduces if used from an eventscript.
913 - Enhance tickles so we can tickle an ipv6 connection.
914 - Start adding ipv6 support : create a new container to replace sockaddr_in.
915 - Add a checksum routine for ipv6/tcp
916 - When starting up ctdb, let the init script do a tdbdump on all
917 persistent databases and verify that they are good (i.e. not corrupted).
918 - Try to use "safe transactions" when writing to a persistent database
919 that was opened with the TDB_NOSYNC flag. If we can get the transaction
920 thats great, if we cant we have to write anyway since we cant block here.
921 * Mon May 12 2008 : Version 1.0.37
922 - When we shutdown ctdb we close the transport down before we run the
923 "shutdown" eventscripts. If ctdb decides to send a packet to a remote node
924 after we have shutdown the transport but before we have shutdown ctdbd
925 itself this could lead to a SEGV instead of a clean shutdown. Fix.
926 - When using the "exportfs" command to extract which NFS export directories
927 to monitor, exportfs violates the "principle of least surprise" and
928 sometimes report a single export line as two lines of text output
929 causing the monitoring to fail.
930 * Fri May 9 2008 : Version 1.0.36
931 - fix a memory corruption bug that could cause the recovery daemon to crash.
932 - fix a bug with distributing public ip addresses during recovery.
933 If the node that is the recovery master did NOT use public addresses,
934 then it assumed that no other node in the cluster used them either and
935 thus skipped the entire step of reallocating public addresses.
936 * Wed May 7 2008 : Version 1.0.35
937 - During recovery, when we define the new set of lmasters (vnnmap)
938 only consider those nodes that have the can-be-lmaster capability
939 when we create the vnnmap. unless there are no nodes available which
940 supports this capability in which case we allow the recmaster to
941 become lmaster capable (temporarily).
942 - Extend the async framework so that we can use paralell async calls
943 to controls that return data.
944 - If we do not have the "can be recmaster" capability, make sure we will
945 lose any recmaster elections, unless there are no nodes available that
946 have the capability, in which case we "take/win" the election anyway.
947 - Close and reopen the reclock pnn file at regular intervals.
948 Make it a non-fatal event if we occasionally fail to open/read/write
950 - Monitor that the recovery daemon is still running from the main ctdb
951 daemon and shutdown the main daemon when recovery daemon has terminated.
952 - Add a "ctdb getcapabilities" command to read the capabilities off a node.
953 - Define two new capabilities : can be recmaster and can be lmaster
954 and default both capabilities to YES.
955 - Log denied tcp connection attempts with DEBUG_ERR and not DEBUG_WARNING
956 * Thu Apr 24 2008 : Version 1.0.34
957 - When deleting a public ip from a node, try to migrate the ip to a different
959 - Change catdb to produce output similar to tdbdump
960 - When adding a new public ip address, if this ip does not exist yet in
961 the cluster, then grab the ip on the local node and activate it.
962 - When a node disagrees with the recmaster on WHO is the recmaster, then
963 mark that node as a recovery culprit so it will eventually become
965 - Make ctdb eventscript support the -n all argument.
966 * Thu Apr 10 2008 : Version 1.0.33
967 - Add facilities to include site local adaptations to the eventscript
968 by /etc/ctdb/rc.local which will be read by all eventscripts.
969 - Add a "ctdb version" command.
970 - Secure the domain socket with proper permissions from Chris Cowan
971 - Bugfixes for AIX from Chris Cowan
972 * Wed Apr 02 2008 : Version 1.0.32
973 - Add a control to have a node execute the eventscripts with arbitrary
974 command line arguments.
975 - Add a control "rddumpmemory" that will dump the talloc memory allocations
976 for the recovery daemon.
977 - Decorate the talloc memdump to produce better and easier memory leak
979 - Update the RHEL5 iscsi tgtd scripts to allow one iscsi target for each
981 - Add two new controls "addip/delip" that can be used to add/remove public
982 addresses to a node at runtime. After using these controls a "ctdb recover"
983 ir required to make the changes take.
984 - Fix a couple of slow memory leaks.
985 * Tue Mar 25 2008 : Version 1.0.31
986 - Add back controls to disable/enable monitoring on a node.
987 - Fix a memory leak where we used to attach CALL data to the ctdb structure
988 when performing a local call. Memory which would be lost if the call was
990 - Reduce the loglevel for the log output when someone connects to a non
991 public ip address for samba.
992 - Redo and optimize the vacuuming process to send only one control to each
993 other node containing all records to be vacuumed instead of one
994 control per node per record.
995 * Tue Mar 04 2008 : Version 1.0.30
996 - Update documentation cor new commands and tuneables
997 - Add machinereadable output to the ip,uptime and getdebug commands
998 - Add a moveip command to manually failover/failback public ips
999 - Add NoIPFallback tuneable that prevents ip address failback
1000 - Use file locking inside the CFS as alternative to verify when other nodes
1001 Are connected/disconnected to be able to recover from split network
1002 - Add DisableWhenUnhealthy tunable
1003 - Add CTDB_START_AS_DISABLED sysconfig param
1004 - Add --start-as-disabled flag to ctdb
1005 - Add ability to monitor for OOM condition
1006 * Thu Feb 21 2008 : Version 1.0.29
1007 - Add a new command to make expansion of an existing cluster easier
1008 - Fix bug with references to freed objects in the ctdb structure
1009 - Propagate debuglevel changes to the recovery daemon
1010 - Merge patches to event scripts from Mathieu Parent :
1011 - MP: Simulate "service" on systems which do not provide this tool
1012 - MP: Set correct permissions for events.d/README
1013 - Add nice helper functions to start/stop nfs from the event scripts
1014 * Fri Feb 08 2008 : Version 1.0.28
1015 - Fix a problem where we tried to use ethtool on non-ethernet interfaces
1016 - Warn if the ipvsadm packege is missing when LVS is used
1017 - Dont use absolute pathnames in some of the event scripts
1018 - Fix for persistent tdbs growing inifinitely.
1019 * Wed Feb 06 2008 : Version 1.0.27
1020 - Add eventscript for iscsi
1021 * Thu Jan 31 2008 : Version 1.0.26
1022 - Fix crashbug in tdb transaction code
1023 * Tue Jan 29 2008 : Version 1.0.25
1024 - added async recovery code
1025 - make event scripts more portable
1026 - fixed ctdb dumpmemory
1027 - more efficient tdb allocation code
1028 - improved machine readable ctdb status output
1030 * Wed Jan 16 2008 : Version 1.0.24
1031 - added syslog support
1032 - documentation updates
1033 * Wed Jan 16 2008 : Version 1.0.23
1034 - fixed a memory leak in the recoveryd
1035 - fixed a corruption bug in the new transaction code
1036 - fixed a case where an packet for a disconnected client could be processed
1037 - added http event script
1038 - updated documentation
1039 * Thu Jan 10 2008 : Version 1.0.22
1040 - auto-run vacuum and repack ops
1041 * Wed Jan 09 2008 : Version 1.0.21
1042 - added ctdb vacuum and ctdb repack code
1043 * Sun Jan 06 2008 : Version 1.0.20
1044 - new transaction based recovery code
1045 * Sat Jan 05 2008 : Version 1.0.19
1046 - fixed non-master bug
1047 - big speedup in recovery for large databases
1048 - lots of changes to improve tdb and ctdb for high churn databases
1049 * Thu Dec 27 2007 : Version 1.0.18
1050 - fixed crash bug in monitor_handler
1051 * Tue Dec 04 2007 : Version 1.0.17
1052 - fixed bugs related to ban/unban of nodes
1053 - fixed a race condition that could lead to monitoring being permanently disabled,
1054 which would lead to long recovery times
1055 - make deterministic IPs the default
1056 - fixed a bug related to continuous recovery
1057 - added a debugging option --node-ip