1 %define initdir %{_sysconfdir}/init.d
5 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
16 Prereq: /bin/mktemp /usr/bin/killall
17 Prereq: fileutils sed /etc/init.d
19 Provides: ctdb = %{version}
22 BuildRoot: %{_tmppath}/%{name}-%{version}-root
25 ctdb is the clustered database used by samba
28 #######################################################################
32 # setup the init script and sysconfig file
33 %setup -T -D -n ctdb-%{version} -q
38 if ccache -h >/dev/null 2>&1 ; then
46 ## always run autogen.sh
49 CFLAGS="$RPM_OPT_FLAGS $EXTRA -O0 -D_GNU_SOURCE -DCTDB_VERS=\"%{version}-%{release}\"" ./configure \
51 --sysconfdir=%{_sysconfdir} \
53 --localstatedir="/var"
55 make docdir=%{_docdir} showflags
56 make docdir=%{_docdir}
59 # Clean up in case there is trash left from a previous build
60 rm -rf $RPM_BUILD_ROOT
62 # Create the target build directory hierarchy
63 mkdir -p $RPM_BUILD_ROOT%{_sysconfdir}/sysconfig
64 mkdir -p $RPM_BUILD_ROOT%{_sysconfdir}/init.d
66 make DESTDIR=$RPM_BUILD_ROOT docdir=%{_docdir} install
68 install -m644 config/ctdb.sysconfig $RPM_BUILD_ROOT%{_sysconfdir}/sysconfig/ctdb
69 install -m755 config/ctdb.init $RPM_BUILD_ROOT%{initdir}/ctdb
71 mkdir -p $RPM_BUILD_ROOT%{_docdir}/ctdb/tests/bin
72 install -m755 tests/bin/ctdb_transaction $RPM_BUILD_ROOT%{_docdir}/ctdb/tests/bin
74 # Remove "*.old" files
75 find $RPM_BUILD_ROOT -name "*.old" -exec rm -f {} \;
78 rm -rf $RPM_BUILD_ROOT
81 #######################################################################
83 #######################################################################
88 %config(noreplace) %{_sysconfdir}/sysconfig/ctdb
89 %config(noreplace) %{_sysconfdir}/ctdb/notify.sh
90 %config(noreplace) %{_sysconfdir}/ctdb/ctdb-crash-cleanup.sh
91 %config(noreplace) %{_sysconfdir}/ctdb/functions
92 %attr(755,root,root) %{initdir}/ctdb
94 %{_docdir}/ctdb/README.eventscripts
95 %{_docdir}/ctdb/recovery-process.txt
96 %{_sysconfdir}/ctdb/events.d/00.ctdb
97 %{_sysconfdir}/ctdb/events.d/01.reclock
98 %{_sysconfdir}/ctdb/events.d/10.interface
99 %{_sysconfdir}/ctdb/events.d/11.natgw
100 %{_sysconfdir}/ctdb/events.d/11.routing
101 %{_sysconfdir}/ctdb/events.d/20.multipathd
102 %{_sysconfdir}/ctdb/events.d/31.clamd
103 %{_sysconfdir}/ctdb/events.d/40.vsftpd
104 %{_sysconfdir}/ctdb/events.d/41.httpd
105 %{_sysconfdir}/ctdb/events.d/50.samba
106 %{_sysconfdir}/ctdb/events.d/60.nfs
107 %{_sysconfdir}/ctdb/events.d/61.nfstickle
108 %{_sysconfdir}/ctdb/events.d/62.cnfs
109 %{_sysconfdir}/ctdb/events.d/70.iscsi
110 %{_sysconfdir}/ctdb/events.d/91.lvs
111 %{_sysconfdir}/ctdb/statd-callout
116 %{_bindir}/ctdb_diagnostics
118 %{_mandir}/man1/ctdb.1.gz
119 %{_mandir}/man1/ctdbd.1.gz
120 %{_mandir}/man1/onnode.1.gz
121 %{_includedir}/ctdb.h
122 %{_includedir}/ctdb_private.h
123 %{_libdir}/pkgconfig/ctdb.pc
125 %{_docdir}/ctdb/tests/bin/ctdb_transaction
128 * Wed Jul 21 2010 : Version 1.0.112-28
129 - Turn back on deterministic IPs by default
130 (needs more analysis)
132 * Wed Jul 21 2010 : Version 1.0.112-27
133 - Turn off deterministic IPs by default
135 - vaccuum abort during freeze
136 CQ:S1018154 & S1018349
137 - logging enhancement for ctdbd children
139 * Thu Jul 15 2010 : Version 1.0.112-26
140 - iptables parallel invocation fix
143 * Tue Jul 6 2010 : Version 1.0.112-25
147 * Mon Jul 5 2010 : Version 1.0.112-24
148 - Extra logging on tdb_chainunlock failures.
149 - Extra sanity check in ctdb_become_dmaster
151 - More robustness against IDR wrap
153 - Recovery failure under stress fix
156 * Tue Jun 8 2010 : Version 1.0.112-23
157 - Fix a SEGV that can be triggered by "ctdb delip"
159 - Add iptables filters to stop clients from connecting to the NATGW
162 - Add timestamps to the ctdb statistics output
163 - Change "ctdb addip" to block until the address is active
165 - Add additional log messages when tdbs can no longer be locked/chain
168 * Mon May 24 2010 : Version 1.0.112-22
169 - Fix bug in 62.cnfs to allow exports that are quoted.
170 - Add monitoring og Quorum for the 62.cnfs script
171 - Allow restoredb to restore a backup to to a different database
172 * Fri May 21 2010 : Version 1.0.112-21
173 - Fix a bug where we would fail to remove the natgw configuration
174 and would leave a ip address for natgw still remaining on loopback.
176 - In the ctdb command line tool, we do not wait long enough for the ipreallocate
177 command to finish on the recovery daemon. which sometimes would timeout if the
178 recovery daemon was busy, or we had rolling recoveries.
180 - During failed recoveries we could get in a state where recovery attempts stopped
181 while database priority level 2 or 3 remain frozen.
183 * Tue May 11 2010 : Version 1.0.112-20
184 - Add number of recoveries done to the ctdb statistics output.
185 - Use the fuill range for IDR values.
187 - When performing a recovery, make sure all nodes agree with recmaster
190 - Lower the loglevel for some debug messages
192 * Tue May 4 2010 : Version 1.0.112-19
193 - Add an eventscript 62.cnfs to allow better integration with GPFS.
195 * Mon May 3 2010 : Version 1.0.112-18
196 - Make the NATGW eventscript check for when the system is misconfigured
197 to use the same address for both natgw as well as a public address
199 - A recent change to monitor the ip address assignment broke "ctdb moveip".
200 Fix ctdb moveip so it works again. BZ62782
201 * Fri Apr 23 2010 : Version 1.0.112-17
202 - In ctdb-crash-cleanup.sh also check for and remove the NATGW address
204 - Add monitoring and warning of low memory condition
205 CTDB_MONITOR_FREE_MEMORY_WARN
206 * Thu Apr 22 2010 : Version 1.0.112-16
207 - Fix for a syntax error bug in 10.interfaces
208 - Add a ctdb-crash-cleanup-sh script that can be used
209 from a cronjob to release ip addresses if ctdb has stopped.
210 * Wed Apr 21 2010 : Version 1.0.112-15
211 - Change how we add/remove iptable rules during recovery to make
212 rules leaks less likey.
213 - change a debug message loglevel from ERR to NOTICE
215 - In the recovery daemon, track pulbic ip assignment
216 across the cluster and verify consistency.
217 - From Martins: change the 10.interface script to handle
218 virtio interfaces correctly for virtual clusters.
219 - Make the recovery master verify the reclock setting across the cluster
220 and ban nodes with inconsistencies.
222 * Mon Mar 29 2010 : Version 1.0.112-14
223 - Wipe the new serverid database before starting winbind
224 * Wed Mar 25 2010 : Version 1.0.112-13
225 - Try to restart NFS if "service nfs start" failed.
227 * Wed Mar 11 2010 : Version 1.0.112-12
228 - From christian ambach : drop the loglevel of a vacuuming message
229 - From Wolfgang Mueller-Friedt : fix bug in ctdb_setstatus
230 - Use "service nfs status" instead of rpcinfo when probing nfs
231 - make sure to always create the tickles directory
232 * Thu Feb 25 2010 : Version 1.0.112-11
233 - Increase default script timeout to 90 seconds.
235 - Add a new tunable EventScriptLogTimeout which will log a message everytime
236 a monitor script takes longer than this to finish.
238 * Tue Feb 23 2010 : Version 1.0.112-10
239 - revert the change in 10.0.0.112-9 and make a new attempt to make the scripts behave.
240 - make writing the ticklelist in 61.nfstickle a background task to avoid
241 having a long cluster fs pause cause a node to become unhealthy
242 - critical bugfix. during an error path in the "end recovery" code
243 we could release a memory block before we had finished referencing it
244 which could lead to a segv. bz 61068
245 - make sure we tear down the natgw configuration when a node become stopped
246 or else we might end up with a duplicate ip address when a different node
247 takes over the natgw role. bz 61036
248 * Tue Feb 16 2010 : Version 1.0.112-9
249 - treat all scripts that hang (except for startup) as success even if they hung
250 * Mon Feb 15 2010 : Version 1.0.112-8
251 - dont increase the tdb seqnum when we just do a dmsater migration
252 this prevents trashing for notify.tdb and registry.tdb since
253 smbd uses tdb seqnums for these databases.
254 - try to restart rpc.rquotad if it is not running
255 * Thu Feb 11 2010 : Version 1.0.112-7
256 - Add logging of process if we detect we have started to swap from
257 the 00.ctdb eventscript.
258 - Reduce the log level of various messages to stop them spamming the log file
259 - Add a variable to disable checking if kNFSd has hung/died or not
260 * Tue Feb 9 2010 : Version 1.0.112-6
261 - Reduce the log level for debug messages related to tracking filedescriptors
263 - Update the linked list macros to provide efficient O(1) behaviour for
264 adding new entries to the tail of the list. This allows much larger lists.
265 - Increase the packet queue list to host 1000000 entries instead of 1000.
266 - when a smbd has hung and the packet queue has filled up. Close that client
267 instead of starting to discard packets. Discarding packets could discard
268 traversal related packets and could cause smbd to block indefinitely.
269 - fix the printout of high latency in the log.
270 * Tue Feb 2 2010 : Version 1.0.112-5
271 - Patch from Volker to fix a TDB transaction rolling back issue.
272 * Wed Jan 21 2010 : Version 1.0.112-4
273 - Update onnode with more flexible ways to define the path to the nodes file
274 * Wed Jan 20 2010 : Version 1.0.112-3
275 - Explicitely source the nfs sysconfig file from the 61.nfstickles script.
276 * Wed Jan 13 2010 : Version 1.0.112-2
277 - Incorrect use of dup2() could cause ctdb to spin eating 100% cpu.
278 * Tue Jan 12 2010 : Version 1.0.112
279 - Revert the use of wbinfo --ping-dc as it is proving too unreliable.
280 - Minor testsuite changes.
281 * Fri Dec 18 2009 : Version 1.0.111
282 - Fix a logging bug when an eventscript is aborted that could cause a crash.
283 - Add back cb_status that was lost in a previous commit.
284 * Fri Dec 18 2009 : Version 1.0.110
285 - Metxe: fix for filedescriptor leak in the new eventscript code.
286 - Rusty: fix for a crash bug in the eventscript code.
287 * Thu Dec 17 2009 : Version 1.0.109
288 - Massive eventscript updates. (bz58828)
289 - Nice the daemon instead of using realtime scheduler, also use mlockall() to
290 reduce the risk of blockign due to paging.
291 - Workarounds for valgrind when forking once for each script. Valgrind consumes
292 massive cpu when terminating the scripts on virtual systems.
293 - Sync the tdb library with upstream, and use the new TDB_DISALLOW_NESTING flag.
294 - Add new command "ctdb dumpdbbackup"
295 - Start using the new tdb check framework to validate tdb files upon startup.
296 - A new framework where we can control health for individual tdb databases.
297 - Fix a crash bug in the logging code.
298 - New transaction code for persistent databases.
299 - Various other smaller fixes.
300 * Mon Dec 7 2009 : Version 1.0.108
301 - Transaction updates from Michael Adam.
302 - Use the new wbinfo --ping-dc instead of -p in the eventscript for samba
303 the check if winbindd is ok.
304 - Add a better "process-exist" for samba so it will automatically
305 reap smbd's on stopped and banned nodes to reclaim subrecords.
306 This will be done a bit differently in the next release.
307 - Use a statically allocated buffer for the 'first-time' capture buffer
308 to reduce the pressure on malloc/free.
309 * Wed Dec 2 2009 : Version 1.0.107
310 - fix for rusty to solve a double-free that can happen when there are
311 multiple packets queued and the connection is destroyed before
312 all packets are processed.
313 * Tue Dec 1 2009 : Version 1.0.106
314 - Buildscript changes from Michael Adam
315 - Dont do a full recovery when there is a mismatch detected for ip addresses,
316 just do a less disruptive ip-reallocation
317 - When starting ctdbd, wait until all initial recoveries have finished
318 before we issue the "startup" event.
319 So dont start services or monitoring until the cluster has
321 - Major eventscript overhaul by Ronnie, Rusty and Martins and fixes of a few
323 * Thu Nov 19 2009 : Version 1.0.105
324 - Fix a bug where we could SEGV if multiple concurrent "ctdb eventscript ..."
325 are used and some of them block.
326 - Monitor the daemon from the syslog child process so we shutdown cleanly when
327 the main daemon terminates.
328 - Add a 500k line ringbuffer in memory where all log messages are stored.
329 - Add a "ctdb getlog <level>" command to pull log messages from the in memory
331 - From martin : fixes to cifs and nfs autotests
332 - from michael a : fix a bashism in 11.natgw
333 * Fri Nov 6 2009 : Version 1.0.104
334 - Suggestion from Metze, we can now use killtcp to kill local connections
335 for nfs so change the killtcp script to kill both directions of an NFS
337 We used to deliberately only kill one direction in these cases due to
339 - Suggestion from christian Ambach, when using natgw, try to avoid using a
340 UNHEALTHY node as the natgw master.
341 - From Michael Adam: Fix a SEGV bug in the recent change to the eventscripts
342 to allow the timeout to apply to each individual script.
343 - fix a talloc bug in teh vacuuming code that produced nasty valgrind
345 - From Rusty: Set up ulimit to create core files for ctdb, and spawned
346 processes by default. This is useful for debugging and testing but can be
347 disabled by setting CTDB_SUPRESS_COREFILE=yes in the sysconfig file.
348 - Remove the wbinfo -t check from the startup check that winbindd is happy.
349 - Enhance the test for bond devices so we also check if the sysadmin have
350 disabled all slave devices using "ifdown".
351 * Tue Nov 3 2009 : Version 1.0.103
352 - Dont use vacuuming on persistent databases
353 - Michael A : transaction updates to persistent databases
354 - Dont activate service automatically when installing the RPM. Leave this to the admin.
355 - Create a child process to send all log messages to, to prevent a hung/slow syslogd
356 from blocking the main daemon. In this case, discard log messages instead and let the child
358 - Michael A: updates to log messages
359 * Thu Oct 29 2009 : Version 1.0.102
360 - Wolfgang: fix for the vacuuming code
361 - Wolfgang: stronger tests for persistent database filename tests
362 - Improve the log message when we refuse to startup since wbinfo -t fails
363 to make it easier to spot in the log.
364 - Update the uptime command output and the man page to indicate that
365 "time since last ..." if from either the last recovery OR the last failover
366 - Michael A: transaction updates
367 * Wed Oct 28 2009 : Version 1.0.101
368 - create a separate context for non-monitoring events so they dont interfere with the monitor event
369 - make sure to return status 0 in teh callback when we abort an event
370 * Wed Oct 28 2009 : Version 1.0.100
371 - Change eventscript handling to allow EventScriptTimeout for each individual script instead of for all scripts as a whole.
372 - Enhanced logging from the eventscripts, log the name and the duration for each script as it finishes.
373 - Add a check to use wbinfo -t for the startup event of samba
374 - TEMP: allow clients to attach to databases even when teh node is in recovery mode
375 - dont run the monitor event as frequently after an event has failed
376 - DEBUG: in the eventloops, check the local time and warn if the time changes backward or rapidly forward
377 - From Metze, fix a bug where recovery master becoming unhealthy did not trigger an ip failover.
378 - Disable the multipath script by default
379 - Automatically re-activate the reclock checking if the reclock file is specified at runtime. Update manpage to reflect this.
380 - Add a mechanism where samba can register a SRVID and if samba unexpectedly disconnects, a message will be broadcasted to all other samba daemons.
381 - Log the pstree on hung scripts to a file in /tmp isntead of /var/log/messages
382 - change ban count before unhealthy/banned to 10
383 * Thu Oct 22 2009 : Version 1.0.99
384 - Fix a SEGV in the new db priority code.
385 - From Wolfgang : eliminate a ctdb_fatal() if there is a dmaster violation detected.
386 - 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.
387 - Lower the debuglevel for the messages when ctdb creates a filedescruiptor so we dont spam the logs with these messages.
388 - Dont have the RPM automatically restart ctdb
389 - Volker : add a missing transaction_cancel() in the handling of persistent databases
390 - 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.
391 * Tue Oct 20 2009 : Version 1.0.98
392 - Fix for the vacuuming database from Wolfgang M
393 - Create a directory where the test framework can put temporary overrides
394 to variables and functions.
395 - Wait a lot longer before shutting down the node when the reclock file
396 is incorrectly configured, and log where it is configured.
397 - Try to avoid running the "monitor" event when databases are frozen.
398 - Add logging for every time we create a filedescriptor so we can trap
400 * Thu Oct 14 2009 : Version 1.0.97
401 - From martins : update onnode.
402 Update onnode to allow specifying an alternative nodes file from
403 the command line and also to be able to specify hostnames on the
405 onnode host1,host2,...
406 * Tue Oct 13 2009 : Version 1.0.96
407 - Add more debugging output when eventscripts have trouble. Print a
408 "pstree -p" to the log when scripts have hung.
409 - Update the initscript, only print the "No reclock file used" warning
410 when we do "service ctdb start", dont also print them for all other
412 - When changing between unhealthy/healthy state, push a request to the
413 recovery master to perform an ip reallocation instead of waiting for the
414 recovery master to pull and check the state change.
415 - Fix a bug in the new db-priority handling where a pre-.95 recovery master
416 could no longer lock the databases on a post-.95 daemon.
417 - Always create the nfs state directories during the "monitor" event.
418 This makes it easier to configure and enable nfs at runtime.
419 - From Volker, forward-port a simper deadlock avoiding patch from the 1.0.82
420 branch. This is a simpler versionof the "db priority lock order" patch
421 that went into 1.0.95, and will be kept for a few versions until samba
422 has been updated to use the functionality from 1.0.95.
423 * Mon Oct 12 2009 : Version 1.0.95
424 - Add database priorities. Allow samba to set the priority of databases
425 and lock the databases in priority order during recovery
426 to avoid a deadlock when samba locks one database then blocks indefinitely
427 while waiting for the second databaso to become locked.
428 - Be aggressive and ban nodes where the recovery transaction start call
430 * Thu Oct 10 2009 : Version 1.0.94
431 - Be very aggressive and quickly ban nodes that can not freeze their databases
432 * Tue Oct 8 2009 : Version 1.0.93
433 - When adding an ip, make sure to update this assignment on all nodes
434 so it wont show up as -1 on other nodes.
435 - When adding an ip and immediately deleting it, it was possible that
436 the daemon would crash accessing already freed memory.
437 Readjust the memory hierarchy so the destructors are called in the right order.
438 - Add a handshake to the recovery daemon to eliminate some rare cases where
439 addip/delip might cause a recovery to occur.
440 - updated onnode documenation from Martin S
441 - Updates to the natgw eventscript to allow disabling natgw at runtime
442 * Fri Oct 2 2009 : Version 1.0.92
443 - Test updates and merge from martin
444 - Add notification for "startup"
445 - Add documentation for notification
446 - from martin, a fix for restarting vsftpd in the eventscript
447 * Tue Sep 29 2009 : Version 1.0.91
448 - New vacuum and repack design from Wolgang Mueller.
449 - 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.
450 - Add machinereadable output to the ctdb getreclock command
451 - merge transaction updates from Michael Adam
452 - In the new banning code, reset the culprit count to 0 for all nodes that could successfully compelte a full recovery.
453 - 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.
454 * Sat Sep 12 2009 : Version 1.0.90
455 - Be more forgiving for eventscripts that hang during startup
456 - Fix for a banning bug in the new banning logic
457 * Thu Sep 3 2009 : Version 1.0.89
458 - Make it possible to manage winbind independently of samba.
459 - Add new prototype banning code
460 - Overwrite the vsftpd state file instead of appending. This eliminates
461 annoying errors in the log.
462 - Redirect some iptables commands to dev null
463 - From Michael A, explicitely set the broadcast when we takeover a public ip
464 - Remove a reclock file check we no longer need
465 - Skip any persistent database files ending in .bak
466 * Mon Aug 17 2009 : Version 1.0.88
467 - Add a new state for eventscripts : DISABLED.
468 Add two new commands "ctdb enablescript/disablescript" to enable/disable
469 eventscripts at runtime.
470 - Bugfixes for TDB from rusty.
471 - Merge/Port changes from upstream TDB library by rusty.
472 - Additional new tests from MartinS. Tests for stop/continue.
473 - Initial patch to rework vacuuming/repacking process from Wolfgang Mueller.
474 - Updates from Michael Adam for persistent writes.
475 - Updates from MartinS to handle the new STOPPED bit in the test framework.
476 - Make it possible to enable/disable the RECMASTER and LMASTER roles
477 at runtime. Add two new commands
478 "ctdb setlmasterrole/setrecmasterrole on/off"
479 - Make it possible to enable/disable the natgw feature at runtime. Add
480 the command "ctdb setnatgwstate on/off"
481 * Fri Jul 17 2009 : Version 1.0.87
482 - Add a new event "stopped" that is called when a node is stopped.
483 - Documentation of the STOPPED flag and the stop/continue commands
484 - Make it possible to start a node in STOPPED mode.
485 - Add a new node flag : STOPPED and commands "ctdb stop" "ctdb continue"
486 These commands are similar to "diasble/enable" but will also remove the node from the vnnmap, while disable only fails all ip addresses over.
487 - tests for NFS , CIFS by martins
488 - major updates to the init script by martins
489 - Send gratious arps with a 1.1 second stride instead of a 1 second stride to workaround interesting "features" of common linux stacks.
490 - Various test enhancements from martins:
491 - additional other tests
492 - add tests for grat arp generation, ping during failover, ssh and failover
493 - New/updated tcp tickle tests and supprot functions
494 - provide better debugging when a test fails
495 - make ctdbd restarts more reliable in the tests
496 - update the "wait bar" to make the wait progress in tests more obvious
498 - 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.
499 - document the ipreallocate command
500 - Updates to enable/disable to use the ipreallocate command to block until the following ipreallocation has completed.
501 - Update the main daemon and the tools to allow debug level to be a string instead of an integer.
502 - Update the sysconfig file to show using string literals instead of numeric values for the debuglevels used.
503 - If no debuglevel is specific, make "ctdb setdebug" show the available options.
504 - 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.
505 * Tue Jun 30 2009 : Version 1.0.86
506 - Do not access the reclock at all if VerifyRecoveryLock is zero, not even try to probe it.
507 - Allow setting the reclock file as "", which means that no reclock file at all should be used.
508 - Document that a reclock file is no longer required, but that it is dangerous.
509 - Add a control that can be used to set/clear/change the reclock file in the daemon during runtime.
510 - Update the recovery daemon to poll whether a reclock file should be sued and if so which file at runtime in each monitoring cycle.
511 - Automatically disable VerifyRecoveryLock everytime a user changes the location of the reclock file.
512 - do not allow the VerifyRecoveryLock to be set using ctdb setvar if there is no recovery lock file specified.
513 - Add two commands "ctdb getreclock" and "ctdb setreclock" to modify the reclock file.
514 * Tue Jun 23 2009 : Version 1.0.85
515 - From William Jojo : Dont use getopt on AIX
516 - Make it possible to use "ctdb listnodes" also when the daemon is not running
517 - Provide machinereadable output to "ctdb listnodes"
518 - Dont list DELETED nodes in the ctdb listnodes output
519 - Try to avoid causing a recovery for the average case when adding/deleting/moving an ip
520 - When banning a node, drop the IPs on that node only and not all nodes.
521 - Add tests for NFS and CIFS tickles
522 - Rename 99.routing to 11.routing so it executes before NFS and LVS scripts
523 - Increase the default timeout before we deem an unresponsive recovery daemon hung and shutdown
524 - Reduce the reclock timout to 5 seconds
525 - Spawn a child process in the recovery daemon ot check the reclock file to
526 avoid blocking the process if the underlying filesystem is unresponsive
527 - fix for filedescriptor leak when a child process timesout
528 - Dont log errors if waitpid() returns -1
529 - Onnode updates by Martins
530 - Test and initscript cleanups from Martin S
531 * Tue Jun 2 2009 : Version 1.0.84
532 - Fix a bug in onnode that could not handle dead nodes
533 * Tue Jun 2 2009 : Version 1.0.83
534 - Document how to remove a ndoe from a running cluster.
535 - Hide all deleted nodes from ctdb output.
536 - Lower the loglevel on some eventscript related items
537 - Dont queue packets to deleted nodes
538 - When building initial vnnmap, ignode any nonexisting nodes
539 - Add a new nodestate : DELETED that is used when deleting a node from an
541 - dont remove the ctdb socket when shutting down. This prevents a race in the
542 initscripts when restarting ctdb quickly after stopping it.
543 - TDB nesting reworked.
544 - Remove obsolete ipmux
545 - From Flavio Carmo Junior: Add eventscript and documentation for ClamAV antivirus engine
546 - From Sumit Bose: fix the regex in the test to handle the new ctdb
547 statistics output that was recently added.
548 - change the socket type we use for grauitious arps from the obsolete
549 AF_INET/SOCK_PACKET to instead use PF_PACKET/SOCK_RAW.
550 - Check return codes for some functions, from Sumit Bose, based on codereview by Jim Meyering.
551 - Sumit Bose: Remove structure memeber node_list_file that is no longer used.
552 - Sumit Bose: fix configure warning for netfilter.h
553 - Updates to the webpages by Volker.
554 - Remove error messages about missing /var/log/log.ctdb file from ctdb_diagnostics.sh from christian Ambach
555 - Additional error logs if hte eventscript switching from dameon to client mode fails.
556 - 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.
557 * Thu May 14 2009 : Version 1.0.82
558 - Update the "ctdb lvsmaster" command to return -1 on error.
559 - Add a -Y flag to "ctdb lvsmaster"
560 - RHEL5 apache leaks semaphores when crashing. Add semaphore cleanup to the
561 41.httpd eventscript and try to restart apache when it has crashed.
562 - Fixes to some tests
563 - Add a -o option to "onnode" which will redirect all stdout to a file for
565 - Add a natgw and a lvs node specifier to onnode so that we can use
567 - Assign the natgw address to lo instead of the private network so it can also
568 be used where private and public networks are the same.
569 - Add GPL boilerplates to two missing scripts.
570 - Change the natgw prefix NATGW_ to CTDB_NATGW_
571 * Fri May 8 2009 : Version 1.0.81
572 - use smbstatus -np instead of smbstatus -n in the 50.samba eventscript
573 since this avoids performing an expensive traverse on the locking and brlock
575 - make ctdb automatically terminate all traverse child processes clusterwide
576 associated to a client application that terminates before the traversal is
578 - From Sumit Bose : fixes to AC_INIT handling.
579 - From Michael Adam, add Tridge's "ping_pong" tool the the ctdb distro since
580 this is very useful for testing the backend filesystem.
581 - From Sumit bose, add support for additional 64 bit platforms.
582 - Add a link from the webpage to Michael Adams SambaXP paper on CTDB.
583 * Fri May 1 2009 : Version 1.0.80
584 - change init shutdown level to 01 for ctdb so it stops before any of the other services
585 - if we can not pull a database from a remote node during recovery, mark that node as a culprit so it becomes banned
586 - 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
587 - 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.
588 - dont unconditionally kill off ctdb and restrat it on "service ctdb start". Fail "service ctdb start" with an error if ctdb is already running.
589 - 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.
590 - fix a cosmetic bug with ctdb statistics where certain counters could become negative.
591 * Wed Apr 8 2009 : Version 1.0.79
592 - From Mathieu Parent: add a ctdb pkgconfig file
594 - add a funciton remove_ip to safely remove an ip from an interface, taking care to workaround an issue with linux alias interfaces.
595 - Update the natgw eventscript to use the safe remove_ip() function
596 - fix a bug in the eventscript child process that would cause the socket to be removed.
597 - dont verify nodemap on banned nodes during cluster monitoring
598 - Update the dodgy SeqnumInterval to have ms resolution
599 * Tue Mar 31 2009 : Version 1.0.78
600 - Add a notify mechanism so we can send snmptraps/email to external management systems when the node becomes unhealthy
601 - include 11.natgw eventscript in thew install so that the NATGW feature works
602 * Tue Mar 31 2009 : Version 1.0.77
603 - 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
604 * Wed Mar 24 2009 : Version 1.0.76
605 - Add a debugging command "xpnn" which can print the pnn of the node even when ctdbd is not running.
606 - Redo the NATGW implementation to allow multiple disjoing NATGW groups in the same cluster.
607 * Tue Mar 24 2009 : Version 1.0.75
608 - Various updates to LVS
609 - Fix a bug in the killtcp control where we did not set the port correctly
610 - add a new "ctdb scriptstatus" command that shows the status of the eventrscripts.
611 * Mon Mar 16 2009 : Version 1.0.74
612 - Fixes to AIX from C Cowan.
613 - Fixes to ctdb_diagnostics so we collect correct GPFS data
614 - Fixes to the net conf list command in ctdb_diagnostics
615 - Check the static-routes file IFF it exists in ctdb_diagnostics
616 * Wed Mar 4 2009 : Version 1.0.73
617 - Add possibility to disable the check of shares for NFS and Samba
618 - From Sumit Bose, fix dependencies so make -j works
619 * Wed Feb 18 2009 : Version 1.0.72
620 - Updates to test scripts by martin s
621 - Adding a COPYING file
622 - Use netstat to check for services and ports and fallback to netcat
623 only if netstat is unavailable.
624 * Mon Feb 2 2009 : Version 1.0.71
625 - Additional ipv6 fixes from Michael Adams
626 * Fri Jan 16 2009 : Version 1.0.70
627 - IPv6 support is completed. this is backward compatible with ipv4-only
628 systems. To use IPv6 with samba and ctdb you need current GIT of samba 3.3
629 or michael adams samba-ctdeb branch.
630 - Many enhancements to the build system and scripts to make it more SUSE
631 friendly by Michael Adams.
632 - Change of how the naming of the package is structured. We are now
633 using "1.0.70" as a release and "-1" as the revision instead of as
634 previously using "1.0" as release and ".70" as the revision.
636 * Thu Dec 18 2008 : Version 1.0.69
637 - Various fixes to scripts by M Adam
638 - Dont call ctdb_fatal() when the transport is down during shutdown
639 * Fri Dec 12 2008 : Version 1.0.68
640 - Fixes for monitoring of interfaces status from Michael Adam.
641 - Use -q instead of >/dev/null for grep to enhance readability of the
642 scripts from Michael Adam.
643 - Update to the "ctdb recover" command. This command now block until the
644 has completed. This makes it much easier to use in scripts and avoids
645 the common workaround :
647 ... loop while waiting for recovery completes ...
649 - Add a CTDB_TIMEOUT variable. If set, this variable provides an automatic
650 timeout for "ctdb <command>", similar to using -T <timeout>
651 - Set a unique errorcode for "ctdb <command>" when it terminates due to a
652 timeout so that scripts can distinguish between a hung command and what was
654 - Update "ctdb ban/unban" so that if the cluster is in recovery these commands
655 blocks and waits until after recovery is complete before the perform the
656 ban/unban operation. This is necessary since the recovery process can cause
657 nodes to become automatically unbanned.
658 - Update "ctdb ban/unban" to block until the recovery that will follow shortly
659 after this command has completed.
660 This makes it much easier to use in scripts and avoids the common
663 ... loop while waiting for recovery completes ...
665 - Bugfix for the new flags handling in 1.0.67. Abort and restart monitoring
666 if we failed to get proper nodemaps from a remote node instead of
667 dereferencing a null pointer.
668 - If ctdbd was explicitely started with the '--socket' argument, make
669 ctdbd automatically set CTDB_SOCKET to the specified argument.
670 This ensures that eventscripts spawned by the ctdb daemon will default to
671 using the same socket and talk to the correct daemon.
672 This primarily affects running multiple daemons on the same host and where
673 you want each instance of ctdb daemons have their eventscripts talk to the
675 - Update "ctdb ping" to return an error code if the ping fail so that it
676 can be used in scripts.
677 - Update to how to synchronize management of node flags across the cluster.
678 * Thu Dec 3 2008 : Version 1.0.67
679 - Add a document describing the recovery process.
680 - Fix a bug in "ctdb setdebug" where it would refuse to set a negative
682 - Print the list of literals for debug names if an invalid one was given
684 - Redesign how "ctdb reloadnodes" works and reduce the amont of tcp teardowns
685 used during this event.
686 - Make it possible to delete a public ip from all nodes at once using
688 * Mon Nov 24 2008 : Version 1.0.66
689 - Allow to change the recmaster even when we are not frozen.
690 - Remove two redundant SAMBA_CHECK variables from the sysconf example
691 - After a node failure it can take very long before some lock operations
692 ctdb needs to perform are allowed/works with gpfs again. Workaround this
693 by treating a hang/timeout as success.
694 - Dont override CTDB_BASE is fet in the shell already
695 - Always send keepalive packets regardless of whether the link is idle or not.
696 - Rewrite the disable/enable flag update logic to prevent a race between
697 "ctdb disable/enable" and the recovery daemon when updating the flags to
699 * Thu Nov 13 2008 : Version 1.0.65
700 - Update the sysconfig example: The default debug level is 2 (NOTICE) and not
702 - Add support for a CTDB_SOCKET environment variable for the ctdb command
703 line tool. If set, this overrides the default socket the ctdb tool will
705 - Add logging of high latency operations.
706 * Mon Oct 22 2008 : Version 1.0.64
707 - Add a context and a timed event so that once we have been in recovery for
708 too long we drop all public addresses.
709 * Mon Oct 20 2008 : Version 1.0.63
710 - Remove logging of "periodic cleanup ..." in 50.samba
711 - When we reload a nodes file, we must detect this and reload the file also
712 in the recovery daemon before we try to dereference somethoung beyond the end
714 * Thu Oct 16 2008 : Version 1.0.62
715 - Allow multiple eventscritps using the same prefix number.
716 It is undefined which order scripts with the same prefix will execute in.
717 * Wed Oct 15 2008 : Version 1.0.61
718 - Use "route add -net" instead of "ip route add" when adding routes in 99.routing
719 - lower the loglevel os several debug statements
720 - check the status returned from ctdb_ctrl_get_tickles() before we try to print them out to the screen.
721 - install a new eventscript 20.multipathd whoich can be used to monitor that multipath devices are healthy
722 * Wed Oct 15 2008 : Version 1.0.60
723 - Verify that nodes we try to ban/unban are reachable and print an error othervise.
724 - 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.
725 - From Mathieu Parent : log debian systems log the package versions in ctdb diagnostics
726 - From Mathieu Parent : specify logdir location for debian (this patch was later reversed)
727 - From Michael Adams : allow # comments in nodes/public_addresses files
728 * Tue Oct 7 2008 : Version 1.0.59
729 - Updated "reloadnodes" logic. Instead of bouncing the entire tcp layer it is sufficient to just close and reopen all outgoing tcp connections.
730 - 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)
731 - IDR tree fix from Jim Houston
732 - Better handling of critical events if the local clock is suddenly changed forward by a lot.
733 - Fix three slow memory leaks in the recovery daemon
734 - New ctdb command : ctdb recmaster which prints the pnn of the recmaster
735 - Onnode enhancements from Martin S : "healthy" and "connected" are now possible nodespecifiers
736 - From Martin S : doc fixes
737 - lowering some debug levels for some nonvital informational messages
738 - Make the daemon daemon monitoring stronger and allow ctdbd to detect a hung
740 - From C Cowan : patches to compile ipv6 under AIX
741 - zero out some structs to keep valgrind happy
742 * Wed Aug 27 2008 : Version 1.0.58
743 - revert the name change tcp_tcp_client back to tcp_control_tcp so
745 - Updates to the init script from Abhijith Das <adas@redhat.com>
746 * Mon Aug 25 2008 : Version 1.0.57
747 - initial support for IPv6
748 * Mon Aug 11 2008 : Version 1.0.56
749 - fix a memory leak in the recovery daemon.
750 * Mon Aug 11 2008 : Version 1.0.55
751 - Fix the releaseip message we seond to samba.
752 * Fri Aug 8 2008 : Version 1.0.54
753 - fix a looping error in the transaction code
754 - provide a more detailed error code for persistent store errors
755 so clients can make more intelligent choices on how to try to recover
756 * Thu Aug 7 2008 : Version 1.0.53
757 - Remove the reclock.pnn file it can cause gpfs to fail to umount
758 - New transaction code
759 * Mon Aug 4 2008 : Version 1.0.52
760 - Send an explicit gratious arp when starting sending the tcp tickles.
761 - When doing failover, issue a killtcp to non-NFS/non-CIFS clients
762 so that they fail quickly. NFS and CIFS already fail and recover
764 - Update the test scripts to handle CTRL-C to kill off the test.
765 * Mon Jul 28 2008 : Version 1.0.51
766 - Strip off the vlan tag from bond devices before we check in /proc
767 if the interface is up or not.
768 - Use testparm in the background in the scripts to allow probing
769 that the shares do exist.
770 - Fix a bug in the logging code to handle multiline entries better
771 - Rename private elements from private to private_data
772 * Fri Jul 18 2008 : Version 1.0.50
773 - Dont assume that just because we can establish a TCP connection
774 that we are actually talking to a functioning ctdb daemon.
775 So dont mark the node as CONNECTED just because the tcp handshake
777 - Dont try to set the recmaster to ourself during elections for those
778 cases we know this will fail. To remove some annoying benign but scary
779 looking entries from the log.
780 - Bugfix for eventsystem for signal handling that could cause a node to
782 * Thu Jul 17 2008 : Version 1.0.49
783 - Update the safe persistent update fix to work with unpatched samba
785 * Thu Jul 17 2008 : Version 1.0.48
786 - Update the spec file.
787 - Do not start new user-triggered eventscripts if we are already
788 inside recovery mode.
789 - Add two new controls to start/cancel a persistent update.
790 A client such as samba can use these to tell ctdbd that it will soon
791 be writing directly to the persistent database tdb file. So if
792 samba is -9ed before it has eitehr done the persistent_store or
793 canceled the operation, ctdb knows that the persistent databases
794 'may' be out of sync and therefore a full blown recovery is called for.
795 - Add two new options :
796 CTDB_SAMBA_SKIP_CONF_CHECK and CTDB_SAMBA_CHECK_PORTS that can be used
797 to override what checks to do when monitoring samba health.
798 We can no longer use the smbstatus, net or testparm commands to check
799 if samba or its config is healthy since these commands may block
800 indefinitely and thus can not be used in scripts.
801 * Fri Jul 11 2008 : Version 1.0.47
802 - Fix a double free bug where if a user striggered (ctdb eventscript)
803 hung and while the timeout handler was being processed a new user
804 triggered eventscript was started we would free state twice.
805 - Rewrite of onnode and associated documentation.
806 * Thu Jul 10 2008 : Version 1.0.46
807 - Document both the LVS:cingle-ip-address and the REMOTE-NODE:wan-accelerator
809 - Add commands "ctdb pnn", "ctdb lvs", "ctdb lvsmaster".
810 - LVS improvements. LVS is the single-ip-address mode for a ctdb cluster.
811 - Fixes to supress rpmlint warnings
813 - Change \s to [[:space:]] in some scripts. Not all RHEL5 packages come
814 with a egrep that handles \s even same version but different arch.
815 - Revert the change to NFS restart. CTDB should NOT attempt to restart
817 - Rewrite of the waitpid() patch to use the eventsystem for handling
819 * Tue Jul 8 2008 : Version 1.0.45
820 - Try to restart the nfs service if it has failed to respond 3 times in a row.
821 - waitpid() can block if the child does not respond promptly to SIGTERM.
822 ignore all SIGCHILD signals by setting SIGCHLD to SIG_DEF.
823 get rid of all calls to waitpid().
824 - make handling of eventscripts hanging more liberal.
825 only consider the script to have failed and making the node unhealthy
826 IF the eventscript terminated wiht an error
827 OR the eventscript hung 5 or more times in a row
828 * Mon Jul 7 2008 : Version 1.0.44
829 - Add a CTDB_VALGRIND option to /etc/sysconfig/ctdb to make it start
830 ctdb under valgrind. Logs go to /var/log/ctdb_valgrind.PID
831 - Add a hack to show the control opcode that caused uninitialized data
832 in the valgrind output by encoding the opcode as the line number.
833 - Initialize structures and allocated memory in various places in
834 ctdb to make it valgrind-clean and remove all valgrind errors/warnings.
835 - If/when we destroy a lockwait child, also make sure we cancel any pending transactions
836 - If a transaction_commit fails, delete/cancel any pending transactions and
837 return an error instead of calling ctdb_fatal()
838 - When running ctdb under valgrind, make sure we run it with --nosetsched and also
839 ensure that we do not use mem-mapped i/o when accessing the tdb's.
840 - zero out ctdb->freeze_handle when we free/destroy a freeze-child.
841 This prevent a heap corruption/ctdb crash bug that could trigger
842 if the freeze child times out.
843 - we dont need to explicitely thaw the databases from the recovery daemon
844 since this is done implicitely when we restore the recovery mode back to normal.
845 - track when we start and stop a recovery. Add the 'time it took to complete the
846 recovery' to the 'ctdb uptime' output.
847 Ensure by tracking the start/stop recovery timestamps that we do not
848 check that the ip allocation is consistend from inside the recovery daemon
849 while a different node (recovery master) is performing a recovery.
850 This prevent a race that could cause a full recovery to trigger if the
851 'ctdb disable/enable' commands took very long.
852 - The freeze child indicates to the master daemon that all databases are locked
853 by writing data to the pipe shared with the master daemon.
854 This write sometimes fail and thus the master daemon never notices that the databases
855 are locked cvausing long timeouts and extra recoveries.
856 Check that the write is successful and try the write again if it failed.
857 - In each node, verify that the recmaster have the right node flags for us
858 and force a push of our flags to the recmaster if wrong.
859 * Tue Jul 1 2008 : Version 1.0.43
860 - Updates and bugfixes to the specfile to keep rpmlint happy
861 - Force a global flags update after each recovery event.
862 - Verify that the recmaster agrees with our node flags and update the
864 - When writing back to the parent from a freeze-child across the pipe,
865 loop over the write in case the write failed with an error othervise
866 the parent will never be notified tha the child has completed the operation.
867 - Automatically thaw all databases when recmaster marks us as being in normal
868 mode instead of recovery mode.
869 * Fri Jun 13 2008 : Version 1.0.42
870 - When event scripts have hung/timedout more than EventScriptBanCount times
871 in a row the node will ban itself.
872 - Many updates to persistent write tests and the test scripts.
873 * Wed May 28 2008 : Version 1.0.41
874 - Reactivate the safe writes to persistent databases and solve the
875 locking issues. Locking issues are solved the only possible way,
876 by using a child process to do the writes. Expensive and slow but... .
877 * Tue May 27 2008 : Version 1.0.40
878 - Read the samba sysconfig file from the 50.samba eventscript
879 - Fix some emmory hierarchical bugs in the persistent write handling
880 * Thu May 22 2008 : Version 1.0.39
881 - Moved a CTDB_MANAGES_NFS, CTDB_MANAGES_ISCSI and CTDB_MANAGES_CSFTPD
882 into /etc/sysconfig/ctdb
883 - Lowered some debug messages to not fill the logfile with entries
884 that normally occur in the default configuration.
885 * Fri May 16 2008 : Version 1.0.38
886 - Add machine readable output support to "ctdb getmonmode"
887 - Lots of tweaks and enhancements if the event scripts are "slow"
888 - Merge from tridge: an attempt to break the chicken-and-egg deadlock that
889 net conf introduces if used from an eventscript.
890 - Enhance tickles so we can tickle an ipv6 connection.
891 - Start adding ipv6 support : create a new container to replace sockaddr_in.
892 - Add a checksum routine for ipv6/tcp
893 - When starting up ctdb, let the init script do a tdbdump on all
894 persistent databases and verify that they are good (i.e. not corrupted).
895 - Try to use "safe transactions" when writing to a persistent database
896 that was opened with the TDB_NOSYNC flag. If we can get the transaction
897 thats great, if we cant we have to write anyway since we cant block here.
898 * Mon May 12 2008 : Version 1.0.37
899 - When we shutdown ctdb we close the transport down before we run the
900 "shutdown" eventscripts. If ctdb decides to send a packet to a remote node
901 after we have shutdown the transport but before we have shutdown ctdbd
902 itself this could lead to a SEGV instead of a clean shutdown. Fix.
903 - When using the "exportfs" command to extract which NFS export directories
904 to monitor, exportfs violates the "principle of least surprise" and
905 sometimes report a single export line as two lines of text output
906 causing the monitoring to fail.
907 * Fri May 9 2008 : Version 1.0.36
908 - fix a memory corruption bug that could cause the recovery daemon to crash.
909 - fix a bug with distributing public ip addresses during recovery.
910 If the node that is the recovery master did NOT use public addresses,
911 then it assumed that no other node in the cluster used them either and
912 thus skipped the entire step of reallocating public addresses.
913 * Wed May 7 2008 : Version 1.0.35
914 - During recovery, when we define the new set of lmasters (vnnmap)
915 only consider those nodes that have the can-be-lmaster capability
916 when we create the vnnmap. unless there are no nodes available which
917 supports this capability in which case we allow the recmaster to
918 become lmaster capable (temporarily).
919 - Extend the async framework so that we can use paralell async calls
920 to controls that return data.
921 - If we do not have the "can be recmaster" capability, make sure we will
922 lose any recmaster elections, unless there are no nodes available that
923 have the capability, in which case we "take/win" the election anyway.
924 - Close and reopen the reclock pnn file at regular intervals.
925 Make it a non-fatal event if we occasionally fail to open/read/write
927 - Monitor that the recovery daemon is still running from the main ctdb
928 daemon and shutdown the main daemon when recovery daemon has terminated.
929 - Add a "ctdb getcapabilities" command to read the capabilities off a node.
930 - Define two new capabilities : can be recmaster and can be lmaster
931 and default both capabilities to YES.
932 - Log denied tcp connection attempts with DEBUG_ERR and not DEBUG_WARNING
933 * Thu Apr 24 2008 : Version 1.0.34
934 - When deleting a public ip from a node, try to migrate the ip to a different
936 - Change catdb to produce output similar to tdbdump
937 - When adding a new public ip address, if this ip does not exist yet in
938 the cluster, then grab the ip on the local node and activate it.
939 - When a node disagrees with the recmaster on WHO is the recmaster, then
940 mark that node as a recovery culprit so it will eventually become
942 - Make ctdb eventscript support the -n all argument.
943 * Thu Apr 10 2008 : Version 1.0.33
944 - Add facilities to include site local adaptations to the eventscript
945 by /etc/ctdb/rc.local which will be read by all eventscripts.
946 - Add a "ctdb version" command.
947 - Secure the domain socket with proper permissions from Chris Cowan
948 - Bugfixes for AIX from Chris Cowan
949 * Wed Apr 02 2008 : Version 1.0.32
950 - Add a control to have a node execute the eventscripts with arbitrary
951 command line arguments.
952 - Add a control "rddumpmemory" that will dump the talloc memory allocations
953 for the recovery daemon.
954 - Decorate the talloc memdump to produce better and easier memory leak
956 - Update the RHEL5 iscsi tgtd scripts to allow one iscsi target for each
958 - Add two new controls "addip/delip" that can be used to add/remove public
959 addresses to a node at runtime. After using these controls a "ctdb recover"
960 ir required to make the changes take.
961 - Fix a couple of slow memory leaks.
962 * Tue Mar 25 2008 : Version 1.0.31
963 - Add back controls to disable/enable monitoring on a node.
964 - Fix a memory leak where we used to attach CALL data to the ctdb structure
965 when performing a local call. Memory which would be lost if the call was
967 - Reduce the loglevel for the log output when someone connects to a non
968 public ip address for samba.
969 - Redo and optimize the vacuuming process to send only one control to each
970 other node containing all records to be vacuumed instead of one
971 control per node per record.
972 * Tue Mar 04 2008 : Version 1.0.30
973 - Update documentation cor new commands and tuneables
974 - Add machinereadable output to the ip,uptime and getdebug commands
975 - Add a moveip command to manually failover/failback public ips
976 - Add NoIPFallback tuneable that prevents ip address failback
977 - Use file locking inside the CFS as alternative to verify when other nodes
978 Are connected/disconnected to be able to recover from split network
979 - Add DisableWhenUnhealthy tunable
980 - Add CTDB_START_AS_DISABLED sysconfig param
981 - Add --start-as-disabled flag to ctdb
982 - Add ability to monitor for OOM condition
983 * Thu Feb 21 2008 : Version 1.0.29
984 - Add a new command to make expansion of an existing cluster easier
985 - Fix bug with references to freed objects in the ctdb structure
986 - Propagate debuglevel changes to the recovery daemon
987 - Merge patches to event scripts from Mathieu Parent :
988 - MP: Simulate "service" on systems which do not provide this tool
989 - MP: Set correct permissions for events.d/README
990 - Add nice helper functions to start/stop nfs from the event scripts
991 * Fri Feb 08 2008 : Version 1.0.28
992 - Fix a problem where we tried to use ethtool on non-ethernet interfaces
993 - Warn if the ipvsadm packege is missing when LVS is used
994 - Dont use absolute pathnames in some of the event scripts
995 - Fix for persistent tdbs growing inifinitely.
996 * Wed Feb 06 2008 : Version 1.0.27
997 - Add eventscript for iscsi
998 * Thu Jan 31 2008 : Version 1.0.26
999 - Fix crashbug in tdb transaction code
1000 * Tue Jan 29 2008 : Version 1.0.25
1001 - added async recovery code
1002 - make event scripts more portable
1003 - fixed ctdb dumpmemory
1004 - more efficient tdb allocation code
1005 - improved machine readable ctdb status output
1007 * Wed Jan 16 2008 : Version 1.0.24
1008 - added syslog support
1009 - documentation updates
1010 * Wed Jan 16 2008 : Version 1.0.23
1011 - fixed a memory leak in the recoveryd
1012 - fixed a corruption bug in the new transaction code
1013 - fixed a case where an packet for a disconnected client could be processed
1014 - added http event script
1015 - updated documentation
1016 * Thu Jan 10 2008 : Version 1.0.22
1017 - auto-run vacuum and repack ops
1018 * Wed Jan 09 2008 : Version 1.0.21
1019 - added ctdb vacuum and ctdb repack code
1020 * Sun Jan 06 2008 : Version 1.0.20
1021 - new transaction based recovery code
1022 * Sat Jan 05 2008 : Version 1.0.19
1023 - fixed non-master bug
1024 - big speedup in recovery for large databases
1025 - lots of changes to improve tdb and ctdb for high churn databases
1026 * Thu Dec 27 2007 : Version 1.0.18
1027 - fixed crash bug in monitor_handler
1028 * Tue Dec 04 2007 : Version 1.0.17
1029 - fixed bugs related to ban/unban of nodes
1030 - fixed a race condition that could lead to monitoring being permanently disabled,
1031 which would lead to long recovery times
1032 - make deterministic IPs the default
1033 - fixed a bug related to continuous recovery
1034 - added a debugging option --node-ip