ctdb-vl.spec 52 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760761762763764765766767768769770771772773774775776777778779780781782783784785786787788789790791792793794795796797798799800801802803804805806807808809810811812813814815816817818819820821822823824825826827828829830831832833834835836837838839840841842843844845846847848849850851852853854855856857858859860861862863864865866867868869870871872873874875876877878879880881882883884885886887888889890891892893894895896897898899900901902903904905906907908909910911912913914915916917918919920921922923924925926927928929930931932933934935936937938939940941942943944945946947948949950951952953954955956957958959960961962963964965966967968969970971972973974975976977978979980981982983984985986987988989990991992993994995996997998999100010011002100310041005100610071008100910101011101210131014101510161017101810191020102110221023102410251026102710281029103010311032103310341035103610371038103910401041104210431044104510461047104810491050105110521053105410551056105710581059106010611062106310641065106610671068106910701071107210731074107510761077107810791080108110821083108410851086108710881089109010911092109310941095109610971098109911001101110211031104110511061107110811091110111111121113111411151116111711181119112011211122112311241125112611271128112911301131113211331134113511361137113811391140114111421143114411451146114711481149115011511152115311541155115611571158115911601161116211631164116511661167116811691170117111721173117411751176117711781179118011811182118311841185118611871188118911901191119211931194119511961197
  1. %define initdir %{_sysconfdir}/rc.d/init.d
  2. Summary: A Clustered Database based on Samba's Trivial Database (TDB)
  3. Name: ctdb
  4. Version: 2.5.5
  5. Release: 1%{?_dist_release}
  6. License: GPLv3+
  7. Group: System Environment/Daemons
  8. URL: http://ctdb.samba.org/
  9. Vendor: Project Vine
  10. Distribution: Vine Linux
  11. Packager: tomop
  12. # Tarfile created using git
  13. # git clone git://git.samba.org/sahlberg/ctdb.git ctdb
  14. # cd ctdb
  15. # git-archive --format=tar --prefix=%{name}-%{version}/ %{name}-%{version} | bzip2 > %{name}-%{version}.tar.bz2
  16. Source0: %{name}-%{version}.tar.gz
  17. # Fedora specific patch, ctdb should not be enabled by default in the runlevels
  18. #Patch1: ctdb-no_default_runlevel.patch
  19. Requires: chkconfig coreutils psmisc
  20. Requires: fileutils sed
  21. Requires: tdb-tools
  22. Requires(preun): chkconfig initscripts
  23. Requires(post): chkconfig
  24. Requires(postun): initscripts
  25. BuildRoot: %(mktemp -ud %{_tmppath}/%{name}-%{version}-%{release}-XXXXXX)
  26. BuildRequires: autoconf net-tools popt-devel
  27. %description
  28. CTDB is a cluster implementation of the TDB database used by Samba and other
  29. projects to store temporary data. If an application is already using TDB for
  30. temporary data it is very easy to convert that application to be cluster aware
  31. and use CTDB instead.
  32. %package devel
  33. Group: Development/Libraries
  34. Summary: CTDB clustered database development package
  35. Requires: ctdb = %{version}-%{release}
  36. Provides: ctdb-static = %{version}-%{release}
  37. %description devel
  38. Libraries, include files, etc you can use to develop CTDB applications.
  39. CTDB is a cluster implementation of the TDB database used by Samba and other
  40. projects to store temporary data. If an application is already using TDB for
  41. temporary data it is very easy to convert that application to be cluster aware
  42. and use CTDB instead.
  43. #######################################################################
  44. %prep
  45. %setup -q
  46. # setup the init script and sysconfig file
  47. %setup -T -D -n ctdb-%{version} -q
  48. #%patch1 -p1
  49. %build
  50. CC="gcc"
  51. ## always run autogen.sh
  52. ./autogen.sh
  53. CFLAGS="$(echo '%{optflags}') $EXTRA -D_GNU_SOURCE -DCTDB_VERS=\"%{version}-%{release}\"" %configure
  54. perl -pi -e 's|rpcgen|rpcgen -Y /usr/bin|g' Makefile
  55. make showflags
  56. make %{_smp_mflags}
  57. %install
  58. # Clean up in case there is trash left from a previous build
  59. rm -rf %{buildroot}
  60. # Create the target build directory hierarchy
  61. mkdir -p %{buildroot}%{_sysconfdir}/sysconfig
  62. mkdir -p %{buildroot}%{initdir}
  63. make DESTDIR=%{buildroot} install
  64. install -m644 config/ctdb.sysconfig %{buildroot}%{_sysconfdir}/sysconfig/ctdb
  65. install -m755 config/ctdb.init %{buildroot}%{initdir}/ctdb
  66. mkdir -p %{buildroot}%{_docdir}/ctdb/tests/bin
  67. install -m755 tests/bin/ctdb_transaction %{buildroot}%{_docdir}/ctdb/tests/bin
  68. # Remove "*.old" files
  69. find %{buildroot} -name "*.old" -exec rm -f {} \;
  70. # fix doc path
  71. mv %{buildroot}%{_docdir}/ctdb %{buildroot}%{_docdir}/ctdb-%{version}
  72. cp -r COPYING web %{buildroot}%{_docdir}/ctdb-%{version}
  73. %clean
  74. rm -rf %{buildroot}
  75. %post
  76. /sbin/chkconfig --add ctdb
  77. %preun
  78. if [ "$1" -eq "0" ] ; then
  79. /sbin/service ctdb stop > /dev/null 2>&1
  80. /sbin/chkconfig --del ctdb
  81. fi
  82. %postun
  83. if [ "$1" -ge "1" ]; then
  84. /sbin/service ctdb condrestart >/dev/null 2>&1 || true
  85. fi
  86. # Files section
  87. %files
  88. %defattr(-,root,root,-)
  89. %config(noreplace) %{_sysconfdir}/sudoers.d/ctdb
  90. %config(noreplace) %{_sysconfdir}/sysconfig/ctdb
  91. %config(noreplace) %{_sysconfdir}/ctdb/notify.sh
  92. %config(noreplace) %{_sysconfdir}/ctdb/ctdb-crash-cleanup.sh
  93. %config(noreplace) %{_sysconfdir}/ctdb/debug_locks.sh
  94. %config(noreplace) %{_sysconfdir}/ctdb/functions
  95. %config(noreplace) %{_sysconfdir}/ctdb/nfs-rpc-checks.d/*
  96. %attr(755,root,root) %{initdir}/ctdb
  97. %{_docdir}/ctdb-%{version}
  98. %dir %{_sysconfdir}/ctdb
  99. %dir %{_sysconfdir}/ctdb/nfs-rpc-checks.d
  100. %{_sysconfdir}/ctdb/statd-callout
  101. %{_sysconfdir}/ctdb/events.d/
  102. %{_sysconfdir}/ctdb/debug-hung-script.sh
  103. %{_sysconfdir}/ctdb/gcore_trace.sh
  104. %{_sbindir}/ctdbd
  105. %{_sbindir}/ctdbd_wrapper
  106. %{_bindir}/ctdb
  107. %{_bindir}/ctdb_event_helper
  108. %{_bindir}/ctdb_lock_helper
  109. %{_bindir}/smnotify
  110. %{_bindir}/ping_pong
  111. %{_bindir}/ctdb_diagnostics
  112. %{_bindir}/onnode
  113. %{_bindir}/ltdbtool
  114. %{_mandir}/man1/ctdb.1*
  115. %{_mandir}/man1/ctdbd.1*
  116. %{_mandir}/man1/ctdbd_wrapper.1*
  117. %{_mandir}/man1/onnode.1*
  118. %{_mandir}/man1/ltdbtool.1*
  119. %{_mandir}/man1/ping_pong.1*
  120. %{_mandir}/man5/ctdbd.conf.5*
  121. %{_mandir}/man7/ctdb-statistics.7*
  122. %{_mandir}/man7/ctdb-tunables.7*
  123. %{_mandir}/man7/ctdb.7*
  124. %files devel
  125. %defattr(-,root,root,-)
  126. %{_includedir}/ctdb.h
  127. %{_includedir}/ctdb_client.h
  128. %{_includedir}/ctdb_protocol.h
  129. %{_includedir}/ctdb_private.h
  130. %{_includedir}/ctdb_typesafe_cb.h
  131. #%{_libdir}/libctdb.a
  132. %{_libdir}/pkgconfig/ctdb.pc
  133. %changelog
  134. * Sat Jun 13 2015 Tomohiro "Tomo-p" KATO <tomop@teamgedoh.net> 2.5.5-1
  135. - new upstream release.
  136. * Wed Dec 3 2014 Tomohiro "Tomo-p" KATO <tomop@teamgedoh.net> 2.5.4-1
  137. - new upstream release.
  138. * Tue Aug 19 2014 Tomohiro "Tomo-p" KATO <tomop@teamgedoh.net> 2.5.3-1
  139. - new upstream release.
  140. * Wed Jan 08 2014 Tomohiro "Tomo-p" KATO <tomop@teamgedoh.net> 2.5-1
  141. - new upstream release.
  142. * Tue Oct 08 2013 Tomohiro "Tomo-p" KATO <tomop@teamgedoh.net> 2.4-1
  143. - new upstream release.
  144. * Mon Dec 17 2012 Tomohiro "Tomo-p" KATO <tomop@teamgedoh.net> 2.0-1
  145. - new upstream release.
  146. - initial build for Vine Linux.
  147. * Mon Jun 27 2011 Michael Schwendt <mschwendt@fedoraproject.org> - 1.2.28-2
  148. - Provide virtual -static package to meet guidelines (#700029).
  149. * Mon Apr 18 2011 Sumit Bose <sbose@redhat.com> - 1.2.28-1
  150. - Update to ctdb version 1.2.28
  151. * Tue Feb 08 2011 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 1.0.114-2
  152. - Rebuilt for https://fedoraproject.org/wiki/Fedora_15_Mass_Rebuild
  153. * Tue Feb 08 2011 Sumit Bose <sbose@redhat.com> - 1.0.114-1
  154. - Changed $RPM_BUILD_ROOT to %{buildroot}
  155. - Update to ctdb version 1.0.114
  156. - Added patch to fix configure issue
  157. - Added assorted backport patches recommended by upstream developer
  158. * Thu Jan 14 2010 Sumit Bose <sbose@redhat.com> - 1.0.113-1
  159. - Update to ctdb version 1.0.113
  160. * Wed Jan 13 2010 : Version 1.0.113
  161. - Incorrect use of dup2() could cause ctdb to spin eating 100% cpu.
  162. * Tue Jan 12 2010 : Version 1.0.112
  163. - Revert the use of wbinfo --ping-dc as it is proving too unreliable.
  164. - Minor testsuite changes.
  165. * Fri Dec 18 2009 : Version 1.0.111
  166. - Fix a logging bug when an eventscript is aborted that could cause a crash.
  167. - Add back cb_status that was lost in a previous commit.
  168. * Fri Dec 18 2009 : Version 1.0.110
  169. - Metxe: fix for filedescriptor leak in the new eventscript code.
  170. - Rusty: fix for a crash bug in the eventscript code.
  171. * Thu Dec 17 2009 : Version 1.0.109
  172. - Massive eventscript updates. (bz58828)
  173. - Nice the daemon instead of using realtime scheduler, also use mlockall() to
  174. reduce the risk of blockign due to paging.
  175. - Workarounds for valgrind when forking once for each script. Valgrind
  176. consumes massive cpu when terminating the scripts on virtual systems.
  177. - Sync the tdb library with upstream, and use the new TDB_DISALLOW_NESTING
  178. flag.
  179. - Add new command "ctdb dumpdbbackup"
  180. - Start using the new tdb check framework to validate tdb files upon startup.
  181. - A new framework where we can control health for individual tdb databases.
  182. - Fix a crash bug in the logging code.
  183. - New transaction code for persistent databases.
  184. - Various other smaller fixes.
  185. * Tue Dec 8 2009 Sumit Bose <sbose@redhat.com> - 1.0.108-1
  186. - Update to ctdb version 1.0.108
  187. - added fix for bz537223
  188. - added tdb-tools to Requires, fixes bz526479
  189. * Wed Dec 2 2009 Sumit Bose <sbose@redhat.com> - 1.0.107-1
  190. - Update to ctdb version 1.0.107
  191. * Wed Dec 2 2009 : Version 1.0.107
  192. - fix for rusty to solve a double-free that can happen when there are
  193. multiple packets queued and the connection is destroyed before
  194. all packets are processed.
  195. * Tue Dec 1 2009 : Version 1.0.106
  196. - Buildscript changes from Michael Adam
  197. - Dont do a full recovery when there is a mismatch detected for ip addresses,
  198. just do a less disruptive ip-reallocation
  199. - When starting ctdbd, wait until all initial recoveries have finished
  200. before we issue the "startup" event.
  201. So dont start services or monitoring until the cluster has
  202. stabilized.
  203. - Major eventscript overhaul by Ronnie, Rusty and Martins and fixes of a few
  204. bugs found.
  205. * Thu Nov 19 2009 : Version 1.0.105
  206. - Fix a bug where we could SEGV if multiple concurrent "ctdb eventscript ..."
  207. are used and some of them block.
  208. - Monitor the daemon from the syslog child process so we shutdown cleanly when
  209. the main daemon terminates.
  210. - Add a 500k line ringbuffer in memory where all log messages are stored.
  211. - Add a "ctdb getlog <level>" command to pull log messages from the in memory
  212. ringbuffer.
  213. - From martin : fixes to cifs and nfs autotests
  214. - from michael a : fix a bashism in 11.natgw
  215. * Fri Nov 6 2009 : Version 1.0.104
  216. - Suggestion from Metze, we can now use killtcp to kill local connections
  217. for nfs so change the killtcp script to kill both directions of an NFS
  218. connection.
  219. We used to deliberately only kill one direction in these cases due to
  220. limitations.
  221. - Suggestion from christian Ambach, when using natgw, try to avoid using a
  222. UNHEALTHY node as the natgw master.
  223. - From Michael Adam: Fix a SEGV bug in the recent change to the eventscripts
  224. to allow the timeout to apply to each individual script.
  225. - fix a talloc bug in teh vacuuming code that produced nasty valgrind
  226. warnings.
  227. - From Rusty: Set up ulimit to create core files for ctdb, and spawned
  228. processes by default. This is useful for debugging and testing but can be
  229. disabled by setting CTDB_SUPRESS_COREFILE=yes in the sysconfig file.
  230. - Remove the wbinfo -t check from the startup check that winbindd is happy.
  231. - Enhance the test for bond devices so we also check if the sysadmin have
  232. disabled all slave devices using "ifdown".
  233. * Tue Nov 3 2009 : Version 1.0.103
  234. - Dont use vacuuming on persistent databases
  235. - Michael A : transaction updates to persistent databases
  236. - Dont activate service automatically when installing the RPM. Leave this to
  237. the admin.
  238. - Create a child process to send all log messages to, to prevent a hung/slow
  239. syslogd from blocking the main daemon. In this case, discard log messages
  240. instead and let the child process block.
  241. - Michael A: updates to log messages
  242. * Thu Oct 29 2009 : Version 1.0.102
  243. - Wolfgang: fix for the vacuuming code
  244. - Wolfgang: stronger tests for persistent database filename tests
  245. - Improve the log message when we refuse to startup since wbinfo -t fails
  246. to make it easier to spot in the log.
  247. - Update the uptime command output and the man page to indicate that
  248. "time since last ..." if from either the last recovery OR the last failover
  249. - Michael A: transaction updates
  250. * Wed Oct 28 2009 : Version 1.0.101
  251. - create a separate context for non-monitoring events so they dont interfere
  252. with the monitor event
  253. - make sure to return status 0 in teh callback when we abort an event
  254. * Wed Oct 28 2009 : Version 1.0.100
  255. - Change eventscript handling to allow EventScriptTimeout for each individual
  256. script instead of for all scripts as a whole.
  257. - Enhanced logging from the eventscripts, log the name and the duration for
  258. each script as it finishes.
  259. - Add a check to use wbinfo -t for the startup event of samba
  260. - TEMP: allow clients to attach to databases even when teh node is in recovery
  261. mode
  262. - dont run the monitor event as frequently after an event has failed
  263. - DEBUG: in the eventloops, check the local time and warn if the time changes
  264. backward or rapidly forward
  265. - From Metze, fix a bug where recovery master becoming unhealthy did not
  266. trigger an ip failover.
  267. - Disable the multipath script by default
  268. - Automatically re-activate the reclock checking if the reclock file is
  269. specified at runtime. Update manpage to reflect this.
  270. - Add a mechanism where samba can register a SRVID and if samba unexpectedly
  271. disconnects, a message will be broadcasted to all other samba daemons.
  272. - Log the pstree on hung scripts to a file in /tmp isntead of
  273. /var/log/messages
  274. - change ban count before unhealthy/banned to 10
  275. * Thu Oct 22 2009 : Version 1.0.99
  276. - Fix a SEGV in the new db priority code.
  277. - From Wolfgang : eliminate a ctdb_fatal() if there is a dmaster violation
  278. detected.
  279. - During testing we often add/delete eventscripts at runtime. This could cause
  280. an eventscript to fail and mark the node unhealthy if an eventscript was
  281. deleted while we were listing the names. Handle the errorcode and make sure
  282. the node does not becomne unhealthy in this case.
  283. - Lower the debuglevel for the messages when ctdb creates a filedescruiptor so
  284. we dont spam the logs with these messages.
  285. - Dont have the RPM automatically restart ctdb
  286. - Volker : add a missing transaction_cancel() in the handling of persistent
  287. databases
  288. - Treat interfaces with the anme ethX* as bond devices in 10.interfaces so we
  289. do the correct test for if they are up or not.
  290. * Tue Oct 20 2009 : Version 1.0.98
  291. - Fix for the vacuuming database from Wolfgang M
  292. - Create a directory where the test framework can put temporary overrides
  293. to variables and functions.
  294. - Wait a lot longer before shutting down the node when the reclock file
  295. is incorrectly configured, and log where it is configured.
  296. - Try to avoid running the "monitor" event when databases are frozen.
  297. - Add logging for every time we create a filedescriptor so we can trap
  298. fd leaks.
  299. * Wed Oct 14 2009 : Version 1.0.97
  300. - From martins : update onnode.
  301. Update onnode to allow specifying an alternative nodes file from
  302. the command line and also to be able to specify hostnames on the
  303. list of targets :
  304. onnode host1,host2,...
  305. * Wed Oct 14 2009 Sumit Bose <sbose@redhat.com> - 1.0.96-1
  306. - Update to ctdb version 1.0.96
  307. * Tue Oct 13 2009 : Version 1.0.96
  308. - Add more debugging output when eventscripts have trouble. Print a
  309. "pstree -p" to the log when scripts have hung.
  310. - Update the initscript, only print the "No reclock file used" warning
  311. when we do "service ctdb start", dont also print them for all other
  312. actions.
  313. - When changing between unhealthy/healthy state, push a request to the
  314. recovery master to perform an ip reallocation instead of waiting for the
  315. recovery master to pull and check the state change.
  316. - Fix a bug in the new db-priority handling where a pre-.95 recovery master
  317. could no longer lock the databases on a post-.95 daemon.
  318. - Always create the nfs state directories during the "monitor" event.
  319. This makes it easier to configure and enable nfs at runtime.
  320. - From Volker, forward-port a simper deadlock avoiding patch from the 1.0.82
  321. branch. This is a simpler versionof the "db priority lock order" patch
  322. that went into 1.0.95, and will be kept for a few versions until samba
  323. has been updated to use the functionality from 1.0.95.
  324. * Mon Oct 12 2009 : Version 1.0.95
  325. - Add database priorities. Allow samba to set the priority of databases
  326. and lock the databases in priority order during recovery
  327. to avoid a deadlock when samba locks one database then blocks indefinitely
  328. while waiting for the second databaso to become locked.
  329. - Be aggressive and ban nodes where the recovery transaction start call
  330. fails.
  331. * Sat Oct 10 2009 : Version 1.0.94
  332. - Be very aggressive and quickly ban nodes that can not freeze their databases
  333. * Thu Oct 8 2009 : Version 1.0.93
  334. - When adding an ip, make sure to update this assignment on all nodes
  335. so it wont show up as -1 on other nodes.
  336. - When adding an ip and immediately deleting it, it was possible that
  337. the daemon would crash accessing already freed memory.
  338. Readjust the memory hierarchy so the destructors are called in the right
  339. order.
  340. - Add a handshake to the recovery daemon to eliminate some rare cases where
  341. addip/delip might cause a recovery to occur.
  342. - updated onnode documenation from Martin S
  343. - Updates to the natgw eventscript to allow disabling natgw at runtime
  344. * Fri Oct 2 2009 : Version 1.0.92
  345. - Test updates and merge from martin
  346. - Add notification for "startup"
  347. - Add documentation for notification
  348. - from martin, a fix for restarting vsftpd in the eventscript
  349. * Tue Sep 29 2009 Sumit Bose <sbose@redhat.com> - 1.0.91-1
  350. - Update to ctdb version 1.0.91
  351. * Tue Sep 29 2009 : Version 1.0.91
  352. - New vacuum and repack design from Wolgang Mueller.
  353. - Add a new eventscript 01.reclock that will first mark a node unhealthy and
  354. later ban the node if the reclock file can not be accessed.
  355. - Add machinereadable output to the ctdb getreclock command
  356. - merge transaction updates from Michael Adam
  357. - In the new banning code, reset the culprit count to 0 for all nodes that
  358. could successfully compelte a full recovery.
  359. - dont mark the recovery master as a ban culprit because a node in the cluster
  360. needs a recovery. this happens naturally when using ctdb recover command so
  361. dont make this cause a node to be banned.
  362. * Wed Sep 23 2009 Sumit Bose <sbose@redhat.com> - 1.0.90-1
  363. - Update to ctdb version 1.0.90
  364. * Sat Sep 12 2009 : Version 1.0.90
  365. - Be more forgiving for eventscripts that hang during startup
  366. - Fix for a banning bug in the new banning logic
  367. * Thu Sep 3 2009 : Version 1.0.89
  368. - Make it possible to manage winbind independently of samba.
  369. - Add new prototype banning code
  370. - Overwrite the vsftpd state file instead of appending. This eliminates
  371. annoying errors in the log.
  372. - Redirect some iptables commands to dev null
  373. - From Michael A, explicitely set the broadcast when we takeover a public ip
  374. - Remove a reclock file check we no longer need
  375. - Skip any persistent database files ending in .bak
  376. * Mon Aug 17 2009 Sumit Bose <sbose@redhat.com> - 1.0.88-1
  377. - Update to ctdb version 1.0.88
  378. * Mon Aug 17 2009 : Version 1.0.88
  379. - Add a new state for eventscripts : DISABLED.
  380. Add two new commands "ctdb enablescript/disablescript" to enable/disable
  381. eventscripts at runtime.
  382. - Bugfixes for TDB from rusty.
  383. - Merge/Port changes from upstream TDB library by rusty.
  384. - Additional new tests from MartinS. Tests for stop/continue.
  385. - Initial patch to rework vacuuming/repacking process from Wolfgang Mueller.
  386. - Updates from Michael Adam for persistent writes.
  387. - Updates from MartinS to handle the new STOPPED bit in the test framework.
  388. - Make it possible to enable/disable the RECMASTER and LMASTER roles
  389. at runtime. Add two new commands
  390. "ctdb setlmasterrole/setrecmasterrole on/off"
  391. - Make it possible to enable/disable the natgw feature at runtime. Add
  392. the command "ctdb setnatgwstate on/off"
  393. * Fri Jul 24 2009 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 1.0.87-2
  394. - Rebuilt for https://fedoraproject.org/wiki/Fedora_12_Mass_Rebuild
  395. * Fri Jul 17 2009 Sumit Bose <sbose@redhat.com> - 1.0.87-1
  396. - Update to ctdb version 1.0.87
  397. * Fri Jul 17 2009 : Version 1.0.87
  398. - Add a new event "stopped" that is called when a node is stopped.
  399. - Documentation of the STOPPED flag and the stop/continue commands
  400. - Make it possible to start a node in STOPPED mode.
  401. - Add a new node flag : STOPPED and commands "ctdb stop" "ctdb continue"
  402. These commands are similar to "diasble/enable" but will also remove the node
  403. from the vnnmap, while disable only fails all ip addresses over.
  404. - tests for NFS , CIFS by martins
  405. - major updates to the init script by martins
  406. - Send gratious arps with a 1.1 second stride instead of a 1 second stride to
  407. workaround interesting "features" of common linux stacks.
  408. - Various test enhancements from martins:
  409. - additional other tests
  410. - add tests for grat arp generation, ping during failover, ssh and failover
  411. - New/updated tcp tickle tests and supprot functions
  412. - provide better debugging when a test fails
  413. - make ctdbd restarts more reliable in the tests
  414. - update the "wait bar" to make the wait progress in tests more obvious
  415. - various cleanups
  416. - when dispatching a message to a handler, make the message a real talloc
  417. object so that we can reparent the object in the tallic hierarchy.
  418. - document the ipreallocate command
  419. - Updates to enable/disable to use the ipreallocate command to block until the
  420. following ipreallocation has completed.
  421. - Update the main daemon and the tools to allow debug level to be a string
  422. instead of an integer.
  423. - Update the sysconfig file to show using string literals instead of numeric
  424. values for the debuglevels used.
  425. - If no debuglevel is specific, make "ctdb setdebug" show the available
  426. options.
  427. - When trying to allocate network packets, add explicit checks if the network
  428. transport has been shutdown before trying and failing, to make log messages
  429. easier to read. Add this extra check and logging to every plave packets are
  430. allocated.
  431. * Wed Jul 1 2009 Sumit Bose <sbose@redhat.com> - 1.0.86-1
  432. - Update to ctdb version 1.0.86
  433. * Tue Jun 30 2009 : Version 1.0.86
  434. - Do not access the reclock at all if VerifyRecoveryLock is zero, not even try
  435. to probe it.
  436. - Allow setting the reclock file as "", which means that no reclock file at
  437. all should be used.
  438. - Document that a reclock file is no longer required, but that it is
  439. dangerous.
  440. - Add a control that can be used to set/clear/change the reclock file in the
  441. daemon during runtime.
  442. - Update the recovery daemon to poll whether a reclock file should be sued and
  443. if so which file at runtime in each monitoring cycle.
  444. - Automatically disable VerifyRecoveryLock everytime a user changes the
  445. location of the reclock file.
  446. - do not allow the VerifyRecoveryLock to be set using ctdb setvar if there is
  447. no recovery lock file specified.
  448. - Add two commands "ctdb getreclock" and "ctdb setreclock" to modify the
  449. reclock file.
  450. * Tue Jun 23 2009 : Version 1.0.85
  451. - From William Jojo : Dont use getopt on AIX
  452. - Make it possible to use "ctdb listnodes" also when the daemon is not running
  453. - Provide machinereadable output to "ctdb listnodes"
  454. - Dont list DELETED nodes in the ctdb listnodes output
  455. - Try to avoid causing a recovery for the average case when
  456. adding/deleting/moving an ip
  457. - When banning a node, drop the IPs on that node only and not all nodes.
  458. - Add tests for NFS and CIFS tickles
  459. - Rename 99.routing to 11.routing so it executes before NFS and LVS scripts
  460. - Increase the default timeout before we deem an unresponsive recovery daemon
  461. hung and shutdown
  462. - Reduce the reclock timout to 5 seconds
  463. - Spawn a child process in the recovery daemon ot check the reclock file to
  464. avoid blocking the process if the underlying filesystem is unresponsive
  465. - fix for filedescriptor leak when a child process timesout
  466. - Dont log errors if waitpid() returns -1
  467. - Onnode updates by Martins
  468. - Test and initscript cleanups from Martin S
  469. * Fri Jun 5 2009 Sumit Bose <sbose@redhat.com> - 1.0.84-1
  470. - Update to ctdb version 1.0.84
  471. * Tue Jun 2 2009 : Version 1.0.84
  472. - Fix a bug in onnode that could not handle dead nodes
  473. * Tue Jun 2 2009 : Version 1.0.83
  474. - Document how to remove a ndoe from a running cluster.
  475. - Hide all deleted nodes from ctdb output.
  476. - Lower the loglevel on some eventscript related items
  477. - Dont queue packets to deleted nodes
  478. - When building initial vnnmap, ignode any nonexisting nodes
  479. - Add a new nodestate : DELETED that is used when deleting a node from an
  480. existing cluster.
  481. - dont remove the ctdb socket when shutting down. This prevents a race in the
  482. initscripts when restarting ctdb quickly after stopping it.
  483. - TDB nesting reworked.
  484. - Remove obsolete ipmux
  485. - From Flavio Carmo Junior: Add eventscript and documentation for ClamAV
  486. antivirus engine
  487. - From Sumit Bose: fix the regex in the test to handle the new ctdb
  488. statistics output that was recently added.
  489. - change the socket type we use for grauitious arps from the obsolete
  490. AF_INET/SOCK_PACKET to instead use PF_PACKET/SOCK_RAW.
  491. - Check return codes for some functions, from Sumit Bose, based on codereview
  492. by Jim Meyering.
  493. - Sumit Bose: Remove structure memeber node_list_file that is no longer used.
  494. - Sumit Bose: fix configure warning for netfilter.h
  495. - Updates to the webpages by Volker.
  496. - Remove error messages about missing /var/log/log.ctdb file from
  497. ctdb_diagnostics.sh from christian Ambach
  498. - Additional error logs if hte eventscript switching from dameon to client
  499. mode fails.
  500. - track how long it takes for ctdbd and the recovery daemon to perform the
  501. rec-lock fcntl() lock attemt and show this in the ctdb statistics output.
  502. * Thu May 14 2009 Sumit Bose <sbose@redhat.com> - 1.0.82-1
  503. - Update to ctdb version 1.0.82
  504. * Thu May 14 2009 : Version 1.0.82
  505. - Update the "ctdb lvsmaster" command to return -1 on error.
  506. - Add a -Y flag to "ctdb lvsmaster"
  507. - RHEL5 apache leaks semaphores when crashing. Add semaphore cleanup to the
  508. 41.httpd eventscript and try to restart apache when it has crashed.
  509. - Fixes to some tests
  510. - Add a -o option to "onnode" which will redirect all stdout to a file for
  511. each of the nodes.
  512. - Add a natgw and a lvs node specifier to onnode so that we can use
  513. "onnode natgw ..."
  514. - Assign the natgw address to lo instead of the private network so it can also
  515. be used where private and public networks are the same.
  516. - Add GPL boilerplates to two missing scripts.
  517. - Change the natgw prefix NATGW_ to CTDB_NATGW_
  518. * Fri May 8 2009 Sumit Bose <sbose@redhat.com> - 1.0.81-1
  519. - Update to ctdb version 1.0.81
  520. * Fri May 8 2009 : Version 1.0.81
  521. - use smbstatus -np instead of smbstatus -n in the 50.samba eventscript
  522. since this avoids performing an expensive traverse on the locking and brlock
  523. databases.
  524. - make ctdb automatically terminate all traverse child processes clusterwide
  525. associated to a client application that terminates before the traversal is
  526. completed.
  527. - From Sumit Bose : fixes to AC_INIT handling.
  528. - From Michael Adam, add Tridge's "ping_pong" tool the the ctdb distro since
  529. this is very useful for testing the backend filesystem.
  530. - From Sumit bose, add support for additional 64 bit platforms.
  531. - Add a link from the webpage to Michael Adams SambaXP paper on CTDB.
  532. * Fri May 1 2009 : Version 1.0.80
  533. - change init shutdown level to 01 for ctdb so it stops before any of the
  534. other services
  535. - if we can not pull a database from a remote node during recovery, mark that
  536. node as a culprit so it becomes banned
  537. - increase the loglevel when we volunteer to drop all ip addresses after
  538. beeing in recovery mode for too long. Make this timeout tuneable with
  539. "RecoveryDropAllIPs" and have it default to 60 seconds
  540. - Add a new flag TDB_NO_NESTING to the tdb layer to prevent nested
  541. transactions which ctdb does not use and does not expect. Have ctdb set this
  542. flag to prevent nested transactions from occuring.
  543. - dont unconditionally kill off ctdb and restrat it on "service ctdb start".
  544. Fail "service ctdb start" with an error if ctdb is already running.
  545. - Add a new tunable "VerifyRecoveryLock" that can be set to 0 to prevent the
  546. main ctdb daemon to verify that the recovery master has locked the reclock
  547. file correctly before allowing it to set the recovery mode to active.
  548. - fix a cosmetic bug with ctdb statistics where certain counters could become
  549. negative.
  550. * Thu Apr 30 2009 Sumit Bose <sbose@redhat.com> - 1.0.79-2
  551. - fixed a ppc64 build issue
  552. * Wed Apr 29 2009 Sumit Bose <sbose@redhat.com> - 1.0.79-1
  553. - Update to ctdb version 1.0.79
  554. * Wed Apr 8 2009 : Version 1.0.79
  555. - From Mathieu Parent: add a ctdb pkgconfig file
  556. - Fix bug 6250
  557. - add a funciton remove_ip to safely remove an ip from an interface, taking
  558. care to workaround an issue with linux alias interfaces.
  559. - Update the natgw eventscript to use the safe remove_ip() function
  560. - fix a bug in the eventscript child process that would cause the socket to be
  561. removed.
  562. - dont verify nodemap on banned nodes during cluster monitoring
  563. - Update the dodgy SeqnumInterval to have ms resolution
  564. * Tue Mar 31 2009 : Version 1.0.78
  565. - Add a notify mechanism so we can send snmptraps/email to external management
  566. systems when the node becomes unhealthy
  567. - include 11.natgw eventscript in thew install so that the NATGW feature works
  568. * Tue Mar 31 2009 : Version 1.0.77
  569. - Update the 99.routing eventscript to also try to add the routes (back)
  570. during a releaseip event. Similar to the reasons why we must add addresses
  571. back during releaseip in 10.interfaces
  572. * Tue Mar 24 2009 : Version 1.0.76
  573. - Add a debugging command "xpnn" which can print the pnn of the node even when
  574. ctdbd is not running.
  575. - Redo the NATGW implementation to allow multiple disjoing NATGW groups in the
  576. same cluster.
  577. * Tue Mar 24 2009 : Version 1.0.75
  578. - Various updates to LVS
  579. - Fix a bug in the killtcp control where we did not set the port correctly
  580. - add a new "ctdb scriptstatus" command that shows the status of the
  581. eventrscripts.
  582. * Mon Mar 16 2009 : Version 1.0.74
  583. - Fixes to AIX from C Cowan.
  584. - Fixes to ctdb_diagnostics so we collect correct GPFS data
  585. - Fixes to the net conf list command in ctdb_diagnostics
  586. - Check the static-routes file IFF it exists in ctdb_diagnostics
  587. * Thu Mar 05 2009 Sumit Bose <sbose@redhat.com> - 1.0.73-1
  588. - Update to ctdb version 1.0.73
  589. * Wed Mar 4 2009 : Version 1.0.73
  590. - Add possibility to disable the check of shares for NFS and Samba
  591. - From Sumit Bose, fix dependencies so make -j works
  592. * Tue Feb 24 2009 Sumit Bose <sbose@redhat.com> - 1.0.72-3
  593. - fix a make -j dependency problem
  594. * Tue Feb 24 2009 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 1.0.72-2
  595. - Rebuilt for https://fedoraproject.org/wiki/Fedora_11_Mass_Rebuild
  596. * Wed Feb 18 2009 Sumit Bose <sbose@redhat.com> - 1.0.72-1
  597. - Update to ctdb version 1.0.72
  598. * Wed Feb 18 2009 : Version 1.0.72
  599. - Updates to test scripts by martin s
  600. - Adding a COPYING file
  601. - Use netstat to check for services and ports and fallback to netcat
  602. only if netstat is unavailable.
  603. * Tue Feb 17 2009 Sumit Bose <sbose@redhat.com> - 1.0.71-5
  604. - more fixed according to https://bugzilla.redhat.com/show_bug.cgi?id=459444
  605. * Sun Feb 8 2009 Sumit Bose <sbose@redhat.com> - 1.0.71-4
  606. - added upstream patch with license file
  607. * Fri Feb 6 2009 Sumit Bose <sbose@redhat.com> - 1.0.71-3
  608. - fixed package according to https://bugzilla.redhat.com/show_bug.cgi?id=459444
  609. * Thu Feb 5 2009 Guenther Deschner <gdeschner@redhat.com> - 1.0.71-2
  610. - Update to ctdb version 1.0.71
  611. * Sun Feb 01 2009 : Version 1.0.71
  612. - Additional ipv6 fixes from Michael Adams
  613. * Thu Jan 15 2009 : Version 1.0.70
  614. - IPv6 support is completed. this is backward compatible with ipv4-only
  615. systems. To use IPv6 with samba and ctdb you need current GIT of samba 3.3
  616. or michael adams samba-ctdeb branch.
  617. - Many enhancements to the build system and scripts to make it more SUSE
  618. friendly by Michael Adams.
  619. - Change of how the naming of the package is structured. We are now
  620. using "1.0.70" as a release and "-1" as the revision instead of as
  621. previously using "1.0" as release and ".70" as the revision.
  622. By Michael Adams.
  623. * Wed Dec 17 2008 : Version 1.0.69
  624. - Various fixes to scripts by M Adam
  625. - Dont call ctdb_fatal() when the transport is down during shutdown
  626. * Thu Dec 11 2008 : Version 1.0.68
  627. - Fixes for monitoring of interfaces status from Michael Adam.
  628. - Use -q instead of >/dev/null for grep to enhance readability of the
  629. scripts from Michael Adam.
  630. - Update to the "ctdb recover" command. This command now block until the
  631. has completed. This makes it much easier to use in scripts and avoids
  632. the common workaround :
  633. ctdb recover
  634. ... loop while waiting for recovery completes ...
  635. continue ...
  636. - Add a CTDB_TIMEOUT variable. If set, this variable provides an automatic
  637. timeout for "ctdb <command>", similar to using -T <timeout>
  638. - Set a unique errorcode for "ctdb <command>" when it terminates due to a
  639. timeout so that scripts can distinguish between a hung command and what was
  640. just a failure.
  641. - Update "ctdb ban/unban" so that if the cluster is in recovery these commands
  642. blocks and waits until after recovery is complete before the perform the
  643. ban/unban operation. This is necessary since the recovery process can cause
  644. nodes to become automatically unbanned.
  645. - Update "ctdb ban/unban" to block until the recovery that will follow shortly
  646. after this command has completed.
  647. This makes it much easier to use in scripts and avoids the common
  648. workaround :
  649. ctdb ban/unban
  650. ... loop while waiting for recovery completes ...
  651. continue ...
  652. - Bugfix for the new flags handling in 1.0.67. Abort and restart monitoring
  653. if we failed to get proper nodemaps from a remote node instead of
  654. dereferencing a null pointer.
  655. - If ctdbd was explicitely started with the '--socket' argument, make
  656. ctdbd automatically set CTDB_SOCKET to the specified argument.
  657. This ensures that eventscripts spawned by the ctdb daemon will default to
  658. using the same socket and talk to the correct daemon.
  659. This primarily affects running multiple daemons on the same host and where
  660. you want each instance of ctdb daemons have their eventscripts talk to the
  661. "correct" daemon.
  662. - Update "ctdb ping" to return an error code if the ping fail so that it
  663. can be used in scripts.
  664. - Update to how to synchronize management of node flags across the cluster.
  665. * Tue Dec 02 2008 : Version 1.0.67
  666. - Add a document describing the recovery process.
  667. - Fix a bug in "ctdb setdebug" where it would refuse to set a negative
  668. debug level.
  669. - Print the list of literals for debug names if an invalid one was given
  670. to "ctdb setdebug"
  671. - Redesign how "ctdb reloadnodes" works and reduce the amont of tcp teardowns
  672. used during this event.
  673. - Make it possible to delete a public ip from all nodes at once using
  674. "ctdb delip -n all"
  675. * Sun Nov 23 2008 : Version 1.0.66
  676. - Allow to change the recmaster even when we are not frozen.
  677. - Remove two redundant SAMBA_CHECK variables from the sysconf example
  678. - After a node failure it can take very long before some lock operations
  679. ctdb needs to perform are allowed/works with gpfs again. Workaround this
  680. by treating a hang/timeout as success.
  681. - Dont override CTDB_BASE is fet in the shell already
  682. - Always send keepalive packets regardless of whether the link is idle or not.
  683. - Rewrite the disable/enable flag update logic to prevent a race between
  684. "ctdb disable/enable" and the recovery daemon when updating the flags to
  685. all nodes.
  686. * Wed Nov 12 2008 : Version 1.0.65
  687. - Update the sysconfig example: The default debug level is 2 (NOTICE) and not
  688. 0 (ERROR)
  689. - Add support for a CTDB_SOCKET environment variable for the ctdb command
  690. line tool. If set, this overrides the default socket the ctdb tool will
  691. use.
  692. - Add logging of high latency operations.
  693. * Tue Oct 21 2008 : Version 1.0.64
  694. - Add a context and a timed event so that once we have been in recovery for
  695. too long we drop all public addresses.
  696. * Sun Oct 19 2008 : Version 1.0.63
  697. - Remove logging of "periodic cleanup ..." in 50.samba
  698. - When we reload a nodes file, we must detect this and reload the file also
  699. in the recovery daemon before we try to dereference somethoung beyond the end
  700. of the nodes array.
  701. * Wed Oct 15 2008 : Version 1.0.62
  702. - Allow multiple eventscritps using the same prefix number.
  703. It is undefined which order scripts with the same prefix will execute in.
  704. * Tue Oct 14 2008 : Version 1.0.61
  705. - Use "route add -net" instead of "ip route add" when adding routes in 99.routing
  706. - lower the loglevel os several debug statements
  707. - check the status returned from ctdb_ctrl_get_tickles() before we try to
  708. print them out to the screen.
  709. - install a new eventscript 20.multipathd whoich can be used to monitor that
  710. multipath devices are healthy
  711. * Tue Oct 14 2008 : Version 1.0.60
  712. - Verify that nodes we try to ban/unban are reachable and print an error othervise.
  713. - Update the client and server sides of TAKEIP/RELEASEIP/GETPUBLICIPS and
  714. GETNODEMAP to fall back to the old style ipv4-only controls if the new
  715. ipv4/ipv6 controls fail. This allows an ipv4/v6 enabled ctdb daemon to
  716. interoperate with earlier ipv4-only versions of the daemons.
  717. - From Mathieu Parent : log debian systems log the package versions in ctdb
  718. diagnostics
  719. - From Mathieu Parent : specify logdir location for debian (this patch was
  720. later reversed)
  721. - From Michael Adams : allow # comments in nodes/public_addresses files
  722. * Mon Oct 06 2008 : Version 1.0.59
  723. - Updated "reloadnodes" logic. Instead of bouncing the entire tcp layer it is
  724. sufficient to just close and reopen all outgoing tcp connections.
  725. - New eventscript 99.routing which can be used to re-attach routes to public
  726. interfaces after a takeip event. (routes may be deleted by the kernel when we
  727. release an ip)
  728. - IDR tree fix from Jim Houston
  729. - Better handling of critical events if the local clock is suddenly changed
  730. forward by a lot.
  731. - Fix three slow memory leaks in the recovery daemon
  732. - New ctdb command : ctdb recmaster which prints the pnn of the recmaster
  733. - Onnode enhancements from Martin S : "healthy" and "connected" are now
  734. possible nodespecifiers
  735. - From Martin S : doc fixes
  736. - lowering some debug levels for some nonvital informational messages
  737. - Make the daemon daemon monitoring stronger and allow ctdbd to detect a hung
  738. recovery daemon.
  739. - From C Cowan : patches to compile ipv6 under AIX
  740. - zero out some structs to keep valgrind happy
  741. * Mon Sep 8 2008 Abhijith Das <adas@redhat.com> - 1.0.58-1
  742. - This release repackages upstream's version 1.0.58 for fedora
  743. * Wed Aug 27 2008 : Version 1.0.58
  744. - revert the name change tcp_tcp_client back to tcp_control_tcp so
  745. samba can build.
  746. - Updates to the init script from Abhijith Das <adas@redhat.com>
  747. * Mon Aug 25 2008 : Version 1.0.57
  748. - initial support for IPv6
  749. * Mon Aug 11 2008 : Version 1.0.56
  750. - fix a memory leak in the recovery daemon.
  751. * Mon Aug 11 2008 : Version 1.0.55
  752. - Fix the releaseip message we seond to samba.
  753. * Fri Aug 8 2008 : Version 1.0.54
  754. - fix a looping error in the transaction code
  755. - provide a more detailed error code for persistent store errors
  756. so clients can make more intelligent choices on how to try to recover
  757. * Thu Aug 7 2008 : Version 1.0.53
  758. - Remove the reclock.pnn file it can cause gpfs to fail to umount
  759. - New transaction code
  760. * Mon Aug 4 2008 : Version 1.0.52
  761. - Send an explicit gratious arp when starting sending the tcp tickles.
  762. - When doing failover, issue a killtcp to non-NFS/non-CIFS clients
  763. so that they fail quickly. NFS and CIFS already fail and recover
  764. quickly.
  765. - Update the test scripts to handle CTRL-C to kill off the test.
  766. * Mon Jul 28 2008 : Version 1.0.51
  767. - Strip off the vlan tag from bond devices before we check in /proc
  768. if the interface is up or not.
  769. - Use testparm in the background in the scripts to allow probing
  770. that the shares do exist.
  771. - Fix a bug in the logging code to handle multiline entries better
  772. - Rename private elements from private to private_data
  773. * Fri Jul 18 2008 : Version 1.0.50
  774. - Dont assume that just because we can establish a TCP connection
  775. that we are actually talking to a functioning ctdb daemon.
  776. So dont mark the node as CONNECTED just because the tcp handshake
  777. was successful.
  778. - Dont try to set the recmaster to ourself during elections for those
  779. cases we know this will fail. To remove some annoying benign but scary
  780. looking entries from the log.
  781. - Bugfix for eventsystem for signal handling that could cause a node to
  782. hang.
  783. * Thu Jul 17 2008 : Version 1.0.49
  784. - Update the safe persistent update fix to work with unpatched samba
  785. servers.
  786. * Thu Jul 17 2008 : Version 1.0.48
  787. - Update the spec file.
  788. - Do not start new user-triggered eventscripts if we are already
  789. inside recovery mode.
  790. - Add two new controls to start/cancel a persistent update.
  791. A client such as samba can use these to tell ctdbd that it will soon
  792. be writing directly to the persistent database tdb file. So if
  793. samba is -9ed before it has eitehr done the persistent_store or
  794. canceled the operation, ctdb knows that the persistent databases
  795. 'may' be out of sync and therefore a full blown recovery is called for.
  796. - Add two new options :
  797. CTDB_SAMBA_SKIP_CONF_CHECK and CTDB_SAMBA_CHECK_PORTS that can be used
  798. to override what checks to do when monitoring samba health.
  799. We can no longer use the smbstatus, net or testparm commands to check
  800. if samba or its config is healthy since these commands may block
  801. indefinitely and thus can not be used in scripts.
  802. * Fri Jul 11 2008 : Version 1.0.47
  803. - Fix a double free bug where if a user striggered (ctdb eventscript)
  804. hung and while the timeout handler was being processed a new user
  805. triggered eventscript was started we would free state twice.
  806. - Rewrite of onnode and associated documentation.
  807. * Thu Jul 10 2008 : Version 1.0.46
  808. - Document both the LVS:cingle-ip-address and the REMOTE-NODE:wan-accelerator
  809. capabilities.
  810. - Add commands "ctdb pnn", "ctdb lvs", "ctdb lvsmaster".
  811. - LVS improvements. LVS is the single-ip-address mode for a ctdb cluster.
  812. - Fixes to supress rpmlint warnings
  813. - AXI compile fixes.
  814. - Change \s to [[:space:]] in some scripts. Not all RHEL5 packages come
  815. with a egrep that handles \s even same version but different arch.
  816. - Revert the change to NFS restart. CTDB should NOT attempt to restart
  817. failed services.
  818. - Rewrite of the waitpid() patch to use the eventsystem for handling
  819. signals.
  820. * Tue Jul 8 2008 : Version 1.0.45
  821. - Try to restart the nfs service if it has failed to respond 3 times in a row.
  822. - waitpid() can block if the child does not respond promptly to SIGTERM.
  823. ignore all SIGCHILD signals by setting SIGCHLD to SIG_DEF.
  824. get rid of all calls to waitpid().
  825. - make handling of eventscripts hanging more liberal.
  826. only consider the script to have failed and making the node unhealthy
  827. IF the eventscript terminated wiht an error
  828. OR the eventscript hung 5 or more times in a row
  829. * Mon Jul 7 2008 : Version 1.0.44
  830. - Add a CTDB_VALGRIND option to /etc/sysconfig/ctdb to make it start
  831. ctdb under valgrind. Logs go to /var/log/ctdb_valgrind.PID
  832. - Add a hack to show the control opcode that caused uninitialized data
  833. in the valgrind output by encoding the opcode as the line number.
  834. - Initialize structures and allocated memory in various places in
  835. ctdb to make it valgrind-clean and remove all valgrind errors/warnings.
  836. - If/when we destroy a lockwait child, also make sure we cancel any pending transactions
  837. - If a transaction_commit fails, delete/cancel any pending transactions and
  838. return an error instead of calling ctdb_fatal()
  839. - When running ctdb under valgrind, make sure we run it with --nosetsched and also
  840. ensure that we do not use mem-mapped i/o when accessing the tdb's.
  841. - zero out ctdb->freeze_handle when we free/destroy a freeze-child.
  842. This prevent a heap corruption/ctdb crash bug that could trigger
  843. if the freeze child times out.
  844. - we dont need to explicitely thaw the databases from the recovery daemon
  845. since this is done implicitely when we restore the recovery mode back to normal.
  846. - track when we start and stop a recovery. Add the 'time it took to complete the
  847. recovery' to the 'ctdb uptime' output.
  848. Ensure by tracking the start/stop recovery timestamps that we do not
  849. check that the ip allocation is consistend from inside the recovery daemon
  850. while a different node (recovery master) is performing a recovery.
  851. This prevent a race that could cause a full recovery to trigger if the
  852. 'ctdb disable/enable' commands took very long.
  853. - The freeze child indicates to the master daemon that all databases are locked
  854. by writing data to the pipe shared with the master daemon.
  855. This write sometimes fail and thus the master daemon never notices that the databases
  856. are locked cvausing long timeouts and extra recoveries.
  857. Check that the write is successful and try the write again if it failed.
  858. - In each node, verify that the recmaster have the right node flags for us
  859. and force a push of our flags to the recmaster if wrong.
  860. * Tue Jul 1 2008 : Version 1.0.43
  861. - Updates and bugfixes to the specfile to keep rpmlint happy
  862. - Force a global flags update after each recovery event.
  863. - Verify that the recmaster agrees with our node flags and update the
  864. recmaster othervise.
  865. - When writing back to the parent from a freeze-child across the pipe,
  866. loop over the write in case the write failed with an error othervise
  867. the parent will never be notified tha the child has completed the operation.
  868. - Automatically thaw all databases when recmaster marks us as being in normal
  869. mode instead of recovery mode.
  870. * Fri Jun 13 2008 : Version 1.0.42
  871. - When event scripts have hung/timedout more than EventScriptBanCount times
  872. in a row the node will ban itself.
  873. - Many updates to persistent write tests and the test scripts.
  874. * Wed May 28 2008 : Version 1.0.41
  875. - Reactivate the safe writes to persistent databases and solve the
  876. locking issues. Locking issues are solved the only possible way,
  877. by using a child process to do the writes. Expensive and slow but... .
  878. * Tue May 27 2008 : Version 1.0.40
  879. - Read the samba sysconfig file from the 50.samba eventscript
  880. - Fix some emmory hierarchical bugs in the persistent write handling
  881. * Thu May 22 2008 : Version 1.0.39
  882. - Moved a CTDB_MANAGES_NFS, CTDB_MANAGES_ISCSI and CTDB_MANAGES_CSFTPD
  883. into /etc/sysconfig/ctdb
  884. - Lowered some debug messages to not fill the logfile with entries
  885. that normally occur in the default configuration.
  886. * Fri May 16 2008 : Version 1.0.38
  887. - Add machine readable output support to "ctdb getmonmode"
  888. - Lots of tweaks and enhancements if the event scripts are "slow"
  889. - Merge from tridge: an attempt to break the chicken-and-egg deadlock that
  890. net conf introduces if used from an eventscript.
  891. - Enhance tickles so we can tickle an ipv6 connection.
  892. - Start adding ipv6 support : create a new container to replace sockaddr_in.
  893. - Add a checksum routine for ipv6/tcp
  894. - When starting up ctdb, let the init script do a tdbdump on all
  895. persistent databases and verify that they are good (i.e. not corrupted).
  896. - Try to use "safe transactions" when writing to a persistent database
  897. that was opened with the TDB_NOSYNC flag. If we can get the transaction
  898. thats great, if we cant we have to write anyway since we cant block here.
  899. * Mon May 12 2008 : Version 1.0.37
  900. - When we shutdown ctdb we close the transport down before we run the
  901. "shutdown" eventscripts. If ctdb decides to send a packet to a remote node
  902. after we have shutdown the transport but before we have shutdown ctdbd
  903. itself this could lead to a SEGV instead of a clean shutdown. Fix.
  904. - When using the "exportfs" command to extract which NFS export directories
  905. to monitor, exportfs violates the "principle of least surprise" and
  906. sometimes report a single export line as two lines of text output
  907. causing the monitoring to fail.
  908. * Fri May 9 2008 : Version 1.0.36
  909. - fix a memory corruption bug that could cause the recovery daemon to crash.
  910. - fix a bug with distributing public ip addresses during recovery.
  911. If the node that is the recovery master did NOT use public addresses,
  912. then it assumed that no other node in the cluster used them either and
  913. thus skipped the entire step of reallocating public addresses.
  914. * Wed May 7 2008 : Version 1.0.35
  915. - During recovery, when we define the new set of lmasters (vnnmap)
  916. only consider those nodes that have the can-be-lmaster capability
  917. when we create the vnnmap. unless there are no nodes available which
  918. supports this capability in which case we allow the recmaster to
  919. become lmaster capable (temporarily).
  920. - Extend the async framework so that we can use paralell async calls
  921. to controls that return data.
  922. - If we do not have the "can be recmaster" capability, make sure we will
  923. lose any recmaster elections, unless there are no nodes available that
  924. have the capability, in which case we "take/win" the election anyway.
  925. - Close and reopen the reclock pnn file at regular intervals.
  926. Make it a non-fatal event if we occasionally fail to open/read/write
  927. to this file.
  928. - Monitor that the recovery daemon is still running from the main ctdb
  929. daemon and shutdown the main daemon when recovery daemon has terminated.
  930. - Add a "ctdb getcapabilities" command to read the capabilities off a node.
  931. - Define two new capabilities : can be recmaster and can be lmaster
  932. and default both capabilities to YES.
  933. - Log denied tcp connection attempts with DEBUG_ERR and not DEBUG_WARNING
  934. * Thu Apr 24 2008 : Version 1.0.34
  935. - When deleting a public ip from a node, try to migrate the ip to a different
  936. node first.
  937. - Change catdb to produce output similar to tdbdump
  938. - When adding a new public ip address, if this ip does not exist yet in
  939. the cluster, then grab the ip on the local node and activate it.
  940. - When a node disagrees with the recmaster on WHO is the recmaster, then
  941. mark that node as a recovery culprit so it will eventually become
  942. banned.
  943. - Make ctdb eventscript support the -n all argument.
  944. * Thu Apr 10 2008 : Version 1.0.33
  945. - Add facilities to include site local adaptations to the eventscript
  946. by /etc/ctdb/rc.local which will be read by all eventscripts.
  947. - Add a "ctdb version" command.
  948. - Secure the domain socket with proper permissions from Chris Cowan
  949. - Bugfixes for AIX from Chris Cowan
  950. * Wed Apr 02 2008 : Version 1.0.32
  951. - Add a control to have a node execute the eventscripts with arbitrary
  952. command line arguments.
  953. - Add a control "rddumpmemory" that will dump the talloc memory allocations
  954. for the recovery daemon.
  955. - Decorate the talloc memdump to produce better and easier memory leak
  956. tracking.
  957. - Update the RHEL5 iscsi tgtd scripts to allow one iscsi target for each
  958. public address.
  959. - Add two new controls "addip/delip" that can be used to add/remove public
  960. addresses to a node at runtime. After using these controls a "ctdb recover"
  961. ir required to make the changes take.
  962. - Fix a couple of slow memory leaks.
  963. * Tue Mar 25 2008 : Version 1.0.31
  964. - Add back controls to disable/enable monitoring on a node.
  965. - Fix a memory leak where we used to attach CALL data to the ctdb structure
  966. when performing a local call. Memory which would be lost if the call was
  967. aborted.
  968. - Reduce the loglevel for the log output when someone connects to a non
  969. public ip address for samba.
  970. - Redo and optimize the vacuuming process to send only one control to each
  971. other node containing all records to be vacuumed instead of one
  972. control per node per record.
  973. * Tue Mar 04 2008 : Version 1.0.30
  974. - Update documentation cor new commands and tuneables
  975. - Add machinereadable output to the ip,uptime and getdebug commands
  976. - Add a moveip command to manually failover/failback public ips
  977. - Add NoIPFallback tuneable that prevents ip address failback
  978. - Use file locking inside the CFS as alternative to verify when other nodes
  979. Are connected/disconnected to be able to recover from split network
  980. - Add DisableWhenUnhealthy tunable
  981. - Add CTDB_START_AS_DISABLED sysconfig param
  982. - Add --start-as-disabled flag to ctdb
  983. - Add ability to monitor for OOM condition
  984. * Thu Feb 21 2008 : Version 1.0.29
  985. - Add a new command to make expansion of an existing cluster easier
  986. - Fix bug with references to freed objects in the ctdb structure
  987. - Propagate debuglevel changes to the recovery daemon
  988. - Merge patches to event scripts from Mathieu Parent :
  989. - MP: Simulate "service" on systems which do not provide this tool
  990. - MP: Set correct permissions for events.d/README
  991. - Add nice helper functions to start/stop nfs from the event scripts
  992. * Fri Feb 08 2008 : Version 1.0.28
  993. - Fix a problem where we tried to use ethtool on non-ethernet interfaces
  994. - Warn if the ipvsadm packege is missing when LVS is used
  995. - Dont use absolute pathnames in some of the event scripts
  996. - Fix for persistent tdbs growing inifinitely.
  997. * Wed Feb 06 2008 : Version 1.0.27
  998. - Add eventscript for iscsi
  999. * Thu Jan 31 2008 : Version 1.0.26
  1000. - Fix crashbug in tdb transaction code
  1001. * Tue Jan 29 2008 : Version 1.0.25
  1002. - added async recovery code
  1003. - make event scripts more portable
  1004. - fixed ctdb dumpmemory
  1005. - more efficient tdb allocation code
  1006. - improved machine readable ctdb status output
  1007. - added ctdb uptime
  1008. * Wed Jan 16 2008 : Version 1.0.24
  1009. - added syslog support
  1010. - documentation updates
  1011. * Wed Jan 16 2008 : Version 1.0.23
  1012. - fixed a memory leak in the recoveryd
  1013. - fixed a corruption bug in the new transaction code
  1014. - fixed a case where an packet for a disconnected client could be processed
  1015. - added http event script
  1016. - updated documentation
  1017. * Thu Jan 10 2008 : Version 1.0.22
  1018. - auto-run vacuum and repack ops
  1019. * Wed Jan 09 2008 : Version 1.0.21
  1020. - added ctdb vacuum and ctdb repack code
  1021. * Sun Jan 06 2008 : Version 1.0.20
  1022. - new transaction based recovery code
  1023. * Sat Jan 05 2008 : Version 1.0.19
  1024. - fixed non-master bug
  1025. - big speedup in recovery for large databases
  1026. - lots of changes to improve tdb and ctdb for high churn databases
  1027. * Thu Dec 27 2007 : Version 1.0.18
  1028. - fixed crash bug in monitor_handler
  1029. * Tue Dec 04 2007 : Version 1.0.17
  1030. - fixed bugs related to ban/unban of nodes
  1031. - fixed a race condition that could lead to monitoring being permanently disabled,
  1032. which would lead to long recovery times
  1033. - make deterministic IPs the default
  1034. - fixed a bug related to continuous recovery
  1035. - added a debugging option --node-ip