github/repmgr
1
0
Fork 0
mirror of https://github.com/EnterpriseDB/repmgr.git synced 2026-03-26 16:46:28 +00:00
Code Issues Packages Projects Releases 1 Wiki Activity

Commit Graph

  • 855ca8fe1a Support separately-compiled pg_rewind for "repmgr standby switchover" in 9.3/9.4 Ian Barwick 2016-01-20 14:21:02 +09:00
  • daa79d1a0f Remove any recovery.done file copied in by pg_rewind Ian Barwick 2016-01-20 13:18:52 +09:00
  • 211768d911 Support pg_rewind when executing "repmgr standby switchover" Ian Barwick 2016-01-20 13:05:47 +09:00
  • f982708b35 Add function test_db_connection() Ian Barwick 2016-01-20 07:56:03 +09:00
  • 995083d66c Change event record type for repmgr standby follow to 'standby_follow' for consistency Ian Barwick 2016-01-19 14:53:16 +09:00
  • be58d6af96 Update TODO Ian Barwick 2016-01-19 13:57:46 +09:00
  • a52e97e622 Update FAQ Ian Barwick 2016-01-19 13:53:18 +09:00
  • cc1ea00333 Ensure event logging doesn't generate an error when cloning from a standby Ian Barwick 2016-01-19 13:51:49 +09:00
  • 7cc7a278c2 Fix comment typo REL3_0_STABLE Ian Barwick 2016-01-19 11:39:41 +09:00
  • ec3596521f Fix comment typo Ian Barwick 2016-01-19 11:39:41 +09:00
  • 66245ccc03 Add warning if -r used with -c Ian Barwick 2016-01-14 23:12:41 +09:00
  • c7542063be Only display -c/--fast-checkpoint hint when using pg_basebackup Ian Barwick 2016-01-14 22:51:02 +09:00
  • 2633d994ef Add FAQ item Ian Barwick 2016-01-14 21:00:30 +09:00
  • 5359d45463 Remove deprecated -l/--local-port from --help output Ian Barwick 2016-01-14 20:49:27 +09:00
  • 20d9f978ab Update HISTORY Ian Barwick 2016-01-14 11:51:51 +09:00
  • 70a2797b9a Improve "archive_mode" configuration check Ian Barwick 2016-01-14 09:33:08 +09:00
  • efa60d142c Update HISTORY Ian Barwick 2016-01-14 11:49:55 +09:00
  • f3d0ab9ab9 Improve "archive_mode" configuration check Ian Barwick 2016-01-14 09:33:08 +09:00
  • 7e6bac1be6 Display a couple of repetitive log messages in verbose mode only Ian Barwick 2016-01-08 11:10:34 +09:00
  • b72058dba8 Update copyright notice to 2016 Ian Barwick 2016-01-05 15:57:25 +09:00
  • 8f62b4c9e6 Update copyright notice to 2016 Ian Barwick 2016-01-05 15:57:25 +09:00
  • 79d1332f9c Update HISTORY Ian Barwick 2016-01-05 13:22:16 +09:00
  • 310f3f31f9 Update HISTORY v3.0.3 Ian Barwick 2016-01-05 13:23:03 +09:00
  • 4f849de95e Suppress warning about -w/--wal-keep-segments unless actually specified Ian Barwick 2016-01-05 13:16:38 +09:00
  • 0de4260664 repmgr: -r/--rsync-only does not require a parameter Ian Barwick 2016-01-05 11:03:30 +09:00
  • cde721e3fc repmgr: -r/--rsync-only does not require a parameter Ian Barwick 2016-01-05 11:03:30 +09:00
  • fc75084e42 repmgrd: -v/--verbose option does not require a parameter Ian Barwick 2016-01-05 10:45:47 +09:00
  • 7b2439b824 repmgrd: -v/--verbose option does not require a parameter Ian Barwick 2016-01-05 10:45:47 +09:00
  • 787cd94142 Changing the "Maintainer" field in the Debian control file to a generic value. Who ever wants to build a deb from the tar-ball can modify the control file and add his contact information. Martin 2016-01-04 13:33:25 -03:00
  • 056e64f635 Bug in the control file: Version should not have a 'v' in front Martin 2016-01-04 13:28:51 -03:00
  • 6b5a609d30 Merge pull request #72 from Bengrunt/patch-1 Martín Marqués 2016-01-04 13:26:59 -03:00
  • cfbc9dd3c6 Backport update_node_record_status() Ian Barwick 2016-01-04 15:27:47 +09:00
  • 94579b5f2e Clean up whitespace and comments Ian Barwick 2016-01-04 11:58:33 +09:00
  • e9a25c367a Prevent invalid replication_lag values being written to the monitoring table Ian Barwick 2016-01-04 13:31:50 +09:00
  • 3088096318 No need to manually create repmgr schema. Ian Barwick 2015-12-29 11:56:39 +09:00
  • 3bbd32c73c Add note about why 'hot_standby=on' is currently required Ian Barwick 2015-12-22 10:47:44 +09:00
  • ac17033d61 This doesn't really mean the standby s following a new master, so we are removing it. Basically, on startup the standby will start receiving again from the begining of the WAL and so received will be lower then applied. Martin 2015-12-17 12:17:03 -03:00
  • 711ad0a76c Change where we activate back the standby node that was failed. Martín Marqués 2015-12-11 09:36:48 -03:00
  • ad988dccce Fix bug discovered last week which prevents recovered standby from being used in the cluster. Main issue was that if the local repmgrd was not able to connect locally, it would set the local node as failed (active = false). This is fine, because we actually don't know if the node is active (actually, it's not active ATM) so it's best to keep it out of the cluster. The problem is that if the postgres service comes back up, and is able to recover by it self, then we should ack that fact and set it as active. There was another issue related with repmgrd being terminated if the postgres service was downs. This is not the correct thing to do: we should keep trying to connect to the local standby. Martín Marqués 2015-12-07 16:14:19 -03:00
  • 53fe3c7e5a Fix bug discovered last week which prevents recovered standby from being used in the cluster. Main issue was that if the local repmgrd was not able to connect locally, it would set the local node as failed (active = false). This is fine, because we actually don't know if the node is active (actually, it's not active ATM) so it's best to keep it out of the cluster. The problem is that if the postgres service comes back up, and is able to recover by it self, then we should ack that fact and set it as active. There was another issue related with repmgrd being terminated if the postgres service was downs. This is not the correct thing to do: we should keep trying to connect to the local standby. Martín Marqués 2015-12-07 15:59:28 -03:00
  • 7a439c90d0 Debian init script repmgrd process stop fix József Kószó 2015-09-21 03:13:19 +02:00
  • 7a4d84379c Prevent invalid replication_lag values being written to the monitoring table Ian Barwick 2016-01-04 13:31:50 +09:00
  • 490e12b1af Clean up whitespace and comments Ian Barwick 2016-01-04 11:58:33 +09:00
  • 7b9df3ac8f Merge pull request #133 from martinmarques/fix-standby-follows-other-node-repmgrd-fails Martín Marqués 2015-12-29 13:25:09 -03:00
  • d6bf870316 Merge pull request #131 from martinmarques/fix-failed-standby Martín Marqués 2015-12-29 13:24:08 -03:00
  • b15e8debe1 No need to manually create repmgr schema. Ian Barwick 2015-12-29 11:56:39 +09:00
  • 310faf1bd9 Merge pull request #137 from martinmarques/create_view_repl_show_nodes Martín Marqués 2015-12-22 13:29:47 -03:00
  • 35caeaa66a We need to update the repmgr.sql file with the new view and create a repmgr3.0_repmgr3.1.sql for the upgrade process when passing to 3.1 Martin 2015-12-22 13:27:42 -03:00
  • ba300c58f7 Make "cluster show" output dynamic Ian Barwick 2015-12-22 15:03:04 +09:00
  • f2370de2fa Minor fixes related to changes in 56b9ca79 Ian Barwick 2015-12-22 13:35:52 +09:00
  • 3920deb803 Merge branch 'create_view_repl_show_nodes' of https://github.com/martinmarques/repmgr Ian Barwick 2015-12-22 13:24:42 +09:00
  • 87e5257cb8 Short option -c does not take a value Ian Barwick 2015-12-22 12:35:42 +09:00
  • e452bf6601 Short option -c does not take a value Ian Barwick 2015-12-22 12:35:42 +09:00
  • 167b4efbb3 Add note about why 'hot_standby=on' is currently required Ian Barwick 2015-12-22 10:47:44 +09:00
  • 56b9ca7992 Re-write part of commit from #116 so we have the query as a view to use from the database. Use the view instead of the query in cluster_show() Martin 2015-12-18 20:53:52 -03:00
  • 9c002c7e38 Merge pull request #116 from renard/follow Martín Marqués 2015-12-18 18:31:07 -03:00
  • cfec04d19f Modify log output to hint Ian Barwick 2015-12-18 17:24:04 +09:00
  • 4f1c67a1bf This doesn't really mean the standby s following a new master, so we are removing it. Basically, on startup the standby will start receiving again from the begining of the WAL and so received will be lower then applied. Martin 2015-12-17 12:17:03 -03:00
  • 2f4fd2b7fa Merge pull request #110 from kjoe/master Martín Marqués 2015-12-12 08:46:56 -03:00
  • aca2b9547f Change where we activate back the standby node that was failed. Martín Marqués 2015-12-11 09:36:48 -03:00
  • c9db7f57d2 Fix bug discovered last week which prevents recovered standby from being used in the cluster. Main issue was that if the local repmgrd was not able to connect locally, it would set the local node as failed (active = false). This is fine, because we actually don't know if the node is active (actually, it's not active ATM) so it's best to keep it out of the cluster. The problem is that if the postgres service comes back up, and is able to recover by it self, then we should ack that fact and set it as active. There was another issue related with repmgrd being terminated if the postgres service was downs. This is not the correct thing to do: we should keep trying to connect to the local standby. Martín Marqués 2015-12-07 16:14:19 -03:00
  • 96ac39ba0f Fix bug discovered last week which prevents recovered standby from being used in the cluster. Main issue was that if the local repmgrd was not able to connect locally, it would set the local node as failed (active = false). This is fine, because we actually don't know if the node is active (actually, it's not active ATM) so it's best to keep it out of the cluster. The problem is that if the postgres service comes back up, and is able to recover by it self, then we should ack that fact and set it as active. There was another issue related with repmgrd being terminated if the postgres service was downs. This is not the correct thing to do: we should keep trying to connect to the local standby. Martín Marqués 2015-12-07 15:59:28 -03:00
  • 1f240ff9b3 Update HISTORY Ian Barwick 2015-11-30 16:58:13 +09:00
  • 88a3378203 Update HISTORY Ian Barwick 2015-11-30 16:56:34 +09:00
  • 9d6cff0d40 Bump version to 3.0.3 Ian Barwick 2015-11-30 16:30:50 +09:00
  • f86e251430 Backport drop_replication_slot() from HEAD Ian Barwick 2015-11-30 16:30:29 +09:00
  • 085b7cb8b4 pg_replslot will only exist in 9.4 and later Ian Barwick 2015-11-30 15:40:43 +09:00
  • 5ccf89ad9b Ensure pg_replslot directory is cleaned up after "standby clone" with rsync Ian Barwick 2015-11-30 15:31:43 +09:00
  • 6ae5401df0 Update TODO Ian Barwick 2015-11-30 14:26:00 +09:00
  • 4bd8190d02 Drop a previously created replication slot if base backup fails for any reason Ian Barwick 2015-11-30 12:38:51 +09:00
  • efdc2355a7 Ensure all failures encountered during a base backup jump to the stop_backup label Ian Barwick 2015-11-30 12:11:42 +09:00
  • 61b1f72a0e Put "starting backup" notice after any slot creation Ian Barwick 2015-11-30 11:58:01 +09:00
  • 882bfd9d8e If we're using replication slots, we need to create them earlier Abhijit Menon-Sen 2015-11-28 21:28:50 +05:30
  • c93f717305 Ensure 'master register --force' can't create more than one active primary node record Ian Barwick 2015-11-26 10:53:59 +09:00
  • 85be96a0be Remove unusable setting Ian Barwick 2015-11-26 10:20:54 +09:00
  • ce2d4fb86f Make t_node_info generally available Ian Barwick 2015-11-25 11:16:45 +09:00
  • 40354e1d62 Add item about hash indexes. Ian Barwick 2015-11-24 13:26:36 +09:00
  • 3e1655f241 Remove hint about hash indexes entirely. Ian Barwick 2015-11-24 13:16:46 +09:00
  • 8387e7f65e Add missing 'break' Ian Barwick 2015-11-23 20:16:01 +09:00
  • aa4dd155b2 Remove unused variable Ian Barwick 2015-11-23 16:42:53 +09:00
  • a171a501ab Shift some common but not terribly informative log messages to verbose mode only Ian Barwick 2015-11-23 14:23:40 +09:00
  • f42f771ff4 Logging fixes Ian Barwick 2015-11-23 14:07:14 +09:00
  • 88cfcf358e Update TODO Ian Barwick 2015-11-23 10:53:59 +09:00
  • ce3594d52d Add /etc/repmgr.conf as a default configuration file location Ian Barwick 2015-11-19 15:16:18 +09:00
  • f64c42a514 Simplify logger_init() parameters Ian Barwick 2015-11-19 14:03:14 +09:00
  • 3072139d06 Update code comments Ian Barwick 2015-11-19 12:55:13 +09:00
  • 3b7185fd39 Update TODO Ian Barwick 2015-11-19 11:38:44 +09:00
  • 819f980e76 Don't display warnings about unused command line parameters in --terse mode Ian Barwick 2015-11-19 10:32:13 +09:00
  • 49316fb8fb repmgr: don't error out on superfluous command line options Ian Barwick 2015-11-19 10:19:15 +09:00
  • fa4ff73b87 Remove implemented TODO item Ian Barwick 2015-11-19 09:38:20 +09:00
  • 29842f0e0d Metadata update also handled by repmgr Ian Barwick 2015-11-18 13:17:51 +09:00
  • 25db1ba737 When following a new primary, have repmgr (not repmgrd) create the new slot Ian Barwick 2015-11-18 13:05:29 +09:00
  • 7b9f6f5352 Minor log message fixes Ian Barwick 2015-11-18 09:10:22 +09:00
  • 53b8f99217 Add a TODO item Ian Barwick 2015-11-17 17:13:26 +09:00
  • 95cdaac91d Update TODO Ian Barwick 2015-11-17 15:42:36 +09:00
  • e7dd0f690c Remove implemented items from TODO list Ian Barwick 2015-11-17 15:35:13 +09:00
  • e0c5bb8d31 Refactor get_master_connection() and update description Ian Barwick 2015-11-17 13:56:14 +09:00
  • df3e55fa35 get_master_connection(): order node list by node type and priority Ian Barwick 2015-11-17 13:10:23 +09:00
  • 0ee2a1e6ba Code formatting Ian Barwick 2015-11-17 11:01:53 +09:00
  • df05214970 Fix variable argument handling with log_hint()/log_verbose() Ian Barwick 2015-11-17 07:34:20 +09:00