Commit Graph

2609 Commits

Author SHA1 Message Date
Evan Tschannen 72c86e909e fix: tracking of the number of unhealthy servers was incorrect
fix: locality equality was only checking zoneId
2018-08-31 17:40:27 -07:00
Evan Tschannen 90bf277206 require key value store memory to recover cleanly when recovering the txnStateStore, since all of the data it is recovering has been fsync’ed 2018-08-31 13:07:48 -07:00
Evan Tschannen 1e2ce75ce4 fix: if usable_regions=1 extraTlogEligibleMachines was calculated incorrectly 2018-08-31 13:04:00 -07:00
Evan Tschannen b0d94597d4 Add additional metrics to track fetch key duration on the storage servers 2018-08-31 13:01:36 -07:00
Evan Tschannen d8659a5822 fix: bytesWritten would overflow and go negative 2018-08-31 12:46:57 -07:00
Evan Tschannen 6496a6d9c8 fix: start move keys will only move destination servers to become source servers if less than destination servers are healthy and the total number of sources is less than 2x the number of destinations 2018-08-31 12:43:14 -07:00
Evan Tschannen e60c668853 The cluster controller will increase its failure monitoring delay after there have been many unfinishedRecoveries 2018-08-31 10:51:55 -07:00
Evan Tschannen 04d998978c
Merge pull request #723 from etschannen/release-6.0
update versions.target for the next release
2018-08-22 00:23:34 -07:00
Evan Tschannen b615f36aee update versions.target for the next release 2018-08-22 00:21:41 -07:00
Evan Tschannen 4d0c4869d2
Merge pull request #722 from etschannen/release-6.0
prepare for 6.0.8
2018-08-21 22:54:03 -07:00
Evan Tschannen c43d07745f prepare for 6.0.8 2018-08-21 22:53:25 -07:00
A.J. Beamon 0b8d5cb287
Merge pull request #721 from etschannen/feature-dd-fixes
Fixed issues with data distribution when configuring from usable_regions=2 to usable_regions=1
2018-08-21 22:48:27 -07:00
Evan Tschannen 84e1f7b2b5 added overhead bytes durable to complement overhead bytes input 2018-08-21 22:35:04 -07:00
Evan Tschannen 74f7412975 added separate logging for overhead bytes 2018-08-21 22:18:38 -07:00
Evan Tschannen ffde1a0e28 renamed onlySystem to mustContainSystemMutations, to accurately represent what setting the key does 2018-08-21 22:15:45 -07:00
Evan Tschannen 8870e6a211 updated release notes for 6.0.8 2018-08-21 21:33:37 -07:00
Evan Tschannen d7c01f0419 added a separate knob for tlog’s recoverMemoryLimit 2018-08-21 21:11:23 -07:00
Evan Tschannen cb60002944 Added the ability to disable all commits which do not modify the system keys by setting \xff/onlySystem = 1 in the database 2018-08-21 21:09:50 -07:00
Evan Tschannen a694364a39 fix: teams larger than the storageTeamSize can never become healthy, so we do not need to track them in our data structures. After configuring from usable_regions=2 to usable_regions=1 we will have a lot of these types of teams, leading to performance issues 2018-08-21 21:08:15 -07:00
Alex Miller 0081add1ef Preparing branch for a future 6.0.8 2018-08-16 14:22:36 -07:00
Alex Miller b97b764514 Bump versions.target to 6.0.7 2018-08-15 20:12:10 -07:00
Alex Miller f363385676 Update versions on downloads page for 6.0.7 2018-08-15 19:59:38 -07:00
Alex Miller 1934750b3a Bump MSI GUID. 2018-08-15 19:59:37 -07:00
Alex Miller e366df9752
Merge pull request #711 from etschannen/release-6.0
Fixed inefficient method of serializing status
2018-08-15 19:47:00 -07:00
Evan Tschannen c950b4780a updated release notes for 6.0.7 2018-08-15 19:42:34 -07:00
Evan Tschannen e770629229 fix: json_spirit::write_string is very CPU intensive, especially for large JSON documents. The cluster controller would call this function for each status reply it needed to send, resulting in a slow task. 2018-08-15 19:39:06 -07:00
Evan Tschannen b7e08377b4
Merge pull request #707 from ajbeamon/remove-non-network-thread-suppressions
Remove suppressions happening off the network thread
2018-08-14 11:03:16 -07:00
A.J. Beamon 168dce94cb Remove some trace event suppressions that were happening off the network thread. Downgrade some trace events related to trace logging problems from SevError to SevWarnAlways. 2018-08-14 09:00:43 -07:00
AlvinMooreSr de396d15f2
Merge pull request #706 from AlvinMooreSr/release-6.0.6
Release 6.0.6 prerequisite changes
2018-08-14 08:08:10 -04:00
Alvin Moore 4519aa8397 Changed Windows GUID
Updated documentation version for release
2018-08-14 05:03:38 -07:00
AlvinMooreSr 1431d1bb66
Merge pull request #705 from etschannen/release-6.0
Added a yield to data distribution and fixed a problem with leader election
2018-08-14 07:56:20 -04:00
Evan Tschannen 883050d12f moved the creation of the yieldPromiseStream to properly yield moves from initialDataDistribution 2018-08-13 22:29:55 -07:00
Evan Tschannen d98f843472 updated release notes for 6.0.6 2018-08-13 21:34:32 -07:00
Evan Tschannen f52d841e8a we need to send notifications when the leader fitness becomes worse so that we repopulate availableCandidates to compare with the new lower fitness 2018-08-13 20:56:02 -07:00
Evan Tschannen 2341e5d8ad fix: we must yield when updating shardsAffectedByTeamFailure with the initial shards. A test with 1 million shards caused a 22 second slow task 2018-08-13 19:46:47 -07:00
Evan Tschannen 8fc8aa0493 fix: we must notify every time nextNominee is not present to continue to repopulate availableCandidates 2018-08-13 17:59:47 -07:00
Alec Grieser fad07307a1
Merge pull request #704 from ajbeamon/directory-test-fixes
Prepopulated directories should have their partition state set.
2018-08-13 13:47:54 -07:00
A.J. Beamon 6af8c0d907 Prepopulated directories should have their partition state set. Filter commit_unknow_result errors if encountered in the directory test. 2018-08-13 13:35:48 -07:00
Evan Tschannen fc4f74e0cc
Merge pull request #703 from etschannen/release-6.0
Merge 5.2 into 6.0
2018-08-13 10:13:51 -07:00
Evan Tschannen aaa90de7d9 merge 5.2 into 6.0 2018-08-13 10:13:03 -07:00
A.J. Beamon b8486d4a2a
Merge pull request #700 from etschannen/release-5.2
Fixed a leader election bug
2018-08-13 09:01:02 -07:00
A.J. Beamon d9bb59c780
Merge pull request #701 from etschannen/release-6.0
Make sure we still accept some connections even if we are CPU bound by high priority work
2018-08-13 08:43:53 -07:00
Evan Tschannen 3186fac397 Make sure we still accept some connections even if we are CPU bound by high priority work 2018-08-10 17:47:21 -07:00
Evan Tschannen 4f9dd10644 fix: as long as some leader was sending heartbeats we would keep the currentNominee as leader, even if that currentNominee was not the one sending the heartbeats 2018-08-10 17:11:24 -07:00
A.J. Beamon faea370555
Merge pull request #697 from ajbeamon/release-6.0
Post-release updates for 6.0.5 release
2018-08-10 14:30:03 -07:00
A.J. Beamon 61480bdf4e Post-release updates for 6.0.5 release 2018-08-10 14:28:56 -07:00
A.J. Beamon 55bfd80e20
Merge pull request #696 from ajbeamon/release-6.0
Prepare for Release 6.0
2018-08-10 12:13:05 -07:00
A.J. Beamon 6bc8dd3a2b Prepare for 6.0.5 release 2018-08-10 12:12:17 -07:00
A.J. Beamon dda71f4597
Merge pull request #694 from etschannen/release-6.0
Updated release notes
2018-08-10 12:02:24 -07:00
A.J. Beamon 1379f37f48
Merge pull request #695 from ajbeamon/release-6.0
Merge Release 5.2 into Release 6.0
2018-08-10 11:54:14 -07:00