Commit Graph

2268 Commits

Author SHA1 Message Date
Evan Tschannen 48fbc407fd fix: we cannot kill all of the remote tlogs, because we still need their data to copy to the next generation in the same data center 2018-06-08 15:28:44 -07:00
Balachandar Namasivayam 32285ee958 Don't crash if fitness file is corrupted in real production use case. 2018-06-08 14:03:36 -07:00
A.J. Beamon 99c9958db7 Some more trace event normalization 2018-06-08 13:57:00 -07:00
A.J. Beamon 0ca51989bb Merge branch 'master' into trace-log-refactor
# Conflicts:
#	fdbserver/QuietDatabase.actor.cpp
#	fdbserver/Status.actor.cpp
#	flow/Trace.cpp
2018-06-08 13:24:30 -07:00
Evan Tschannen 50779a1860
Merge pull request #448 from bnamasivayam/fix-trprofile-test-bug
Having fixed limits for getRange results in continuously getting tran…
2018-06-08 12:52:50 -07:00
Balachandar Namasivayam 34995d4d64 Address review comments. 2018-06-08 11:51:51 -07:00
Evan Tschannen 943dfa278b
Merge pull request #454 from ajbeamon/normalize-trace-events
Attempt to normalize trace events
2018-06-08 11:28:44 -07:00
Balachandar Namasivayam 20febf5ef9 Address review comments. 2018-06-08 11:24:51 -07:00
A.J. Beamon c12b235080 Fix case in a few commented out trace events 2018-06-08 11:20:06 -07:00
A.J. Beamon e5488419cc Attempt to normalize trace events:
* Detail names now all start with an uppercase character and contain no underscores. Ideally these should be head-first camel case, though that was harder to check.
* Type names have the same rules, except they allow one underscore (to support a usage pattern Context_Type). The first character after the underscore is also uppercase.
* Use seconds instead of milliseconds in details.

Added a check when events are logged in simulation that logs a message to stderr if the first two rules above aren't followed.

This probably doesn't address every instance of the above problems, but all of the events I was able to hit in simulation pass the check.
2018-06-08 11:11:08 -07:00
Alec Grieser 42e6f2c216
Merge pull request #444 from ajbeamon/release-5.2
Update version-specific upgrade notes
2018-06-08 11:03:36 -07:00
A.J. Beamon 4e921417df Make wording consistent 2018-06-08 10:49:43 -07:00
A.J. Beamon c005560804 Merge branch 'release-5.2' of github.com:apple/foundationdb into release-5.2 2018-06-08 10:46:22 -07:00
A.J. Beamon f1d389448c
Merge pull request #453 from apple/release-5.2
Merge release-5.2 into master
2018-06-08 10:41:44 -07:00
A.J. Beamon 6461478695
Merge pull request #452 from apple/release-5.1
Merge release-5.1 into release-5.2
2018-06-08 10:41:13 -07:00
Evan Tschannen d7d38c3544
Merge pull request #430 from ajbeamon/rename-logGroup-attribute
Rename trace file logGroup attribute to LogGroup
2018-06-08 10:30:45 -07:00
Evan Tschannen 953c27e570
Merge pull request #431 from ajbeamon/tlog-rename-variables
Rename several variables in TLogServer.actor.cpp to follow our normal camel case conventions.
2018-06-08 10:30:22 -07:00
Evan Tschannen 12c45ccf79
Merge pull request #451 from ajbeamon/release-5.1
Fix case of newSeverity detail in StderrSeverity trace event
2018-06-08 10:28:30 -07:00
A.J. Beamon c9543791fd Fix case of newSeverity detail in StderrSeverity trace event 2018-06-08 10:24:12 -07:00
Jay Kominek fb33412b3a use inspect.getfullargspec when available
getargspec was deprecated in python3, this should use
getfullargspec when available, and degrade gracefully
otherwise.
2018-06-08 01:07:18 -06:00
Evan Tschannen 7d392689fe fix: only update metrics for healthy destinations, because unhealthy destinations are already in the source 2018-06-07 18:12:04 -07:00
Evan Tschannen e4d5817679 fix: we must server getTeam requests before readyToStart is set because we cannot complete relocateShard requests without getTeam responses from both team collections 2018-06-07 16:14:40 -07:00
Balachandar Namasivayam 514b0e3c20 Having fixed limits for getRange results in continuously getting transaction_too_old error in some scenarios.
Cutting the limits by half in such cases allows to test to progress.
2018-06-07 15:27:05 -07:00
Evan Tschannen 9f0c16f062 do not build teams which contain failed servers 2018-06-07 14:05:53 -07:00
Balachandar Namasivayam 11b79c6c94 Save fitness info of a process to become a cluster controller. This info is currently lost after a reboot. Save this info and reload it to avoid unnecessary re-recruitments. 2018-06-07 13:07:19 -07:00
Evan Tschannen b423d73b42 fix: do not finish a shard relocation until all of the storage servers have made the current recovery version durable. This is to prevent dropping a needed storage server as a source for a shard after dropping a remote configuration 2018-06-07 12:29:25 -07:00
A.J. Beamon f463245ca0 Update version-specific upgrade notes 2018-06-06 15:42:27 -07:00
A.J. Beamon f9cec3c6bb
Merge pull request #443 from ajbeamon/master
Merge release-5.2
2018-06-06 15:28:33 -07:00
A.J. Beamon 3fde6cbaa7 Change MSI GUID in 6.0 2018-06-06 15:27:36 -07:00
A.J. Beamon 216404de45 Merge branch 'release-5.2' of github.com:apple/foundationdb
# Conflicts:
#	documentation/sphinx/source/release-notes.rst
#	versions.target
2018-06-06 15:25:37 -07:00
A.J. Beamon b52681c763
Merge pull request #442 from ajbeamon/release-5.2
Update versions.target and MSI package for 5.2.4
2018-06-06 15:21:39 -07:00
A.J. Beamon cf7ab15c3e Update versions.target and MSI package for 5.2.4 2018-06-06 15:20:50 -07:00
A.J. Beamon b8efd4c2b5
Merge pull request #441 from ajbeamon/release-5.2
Updates for release 5.2.3. This excludes required changes to administ…
2018-06-06 14:14:23 -07:00
A.J. Beamon bd90cdbc59 Updates for release 5.2.3. This excludes required changes to administration.rst pending some clarification about how certain upgrades work. 2018-06-06 14:13:20 -07:00
A.J. Beamon 59caa968dd
Merge pull request #440 from etschannen/release-5.2
backup created large transactions when erasing log ranges
2018-06-06 13:45:43 -07:00
Evan Tschannen e82985aea2 fix: continue setting beginVersion so that versions between 5.2.0 and 5.2.2 do not crash when decoding tasks created by 5.2.3 2018-06-06 13:34:22 -07:00
Evan Tschannen 7c289c12d6 updated release notes 2018-06-06 13:11:34 -07:00
Evan Tschannen 4120062bb9 fix: backup initialized its begin version at 1 instead of the read version of the starting transaction
fix: erasing log ranges did not properly divide up work between transactions to prevent making transactions which were too large
2018-06-06 13:05:53 -07:00
Richard Low 7a84375f6d Better 5.2.2 release notes wording for TLS 2018-06-05 14:52:49 -07:00
Evan Tschannen e659dc7c2c
Merge pull request #438 from ajbeamon/release-5.2
Add read-only commit fix and some PR links to release notes.
2018-06-05 13:44:05 -07:00
Evan Tschannen f26a2f771d fix: log router popped one too many versions from messageBlocks 2018-06-05 13:42:48 -07:00
A.J. Beamon eeb92fbfa6 Add read-only commit fix and some PR links to release notes. 2018-06-05 13:41:37 -07:00
Evan Tschannen 6fbb046320
Merge pull request #437 from ajbeamon/fix-commit-readonly-transactions
fix: Read-only transactions that get committed would fail if the...
2018-06-05 13:13:11 -07:00
A.J. Beamon e4e06321c7 fix: Read-only transactions that get committed would fail if the readOnly option is set. They would also be counted in the transactionsCommitStarted metric. 2018-06-05 12:10:28 -07:00
Evan Tschannen be06938d9d fix: dropping the remote replication will cause all remote storage servers to die. Make sure we are not restoring redundancy before doing this to prevent data loss in simulation. 2018-06-04 18:46:09 -07:00
Evan Tschannen 6cf9508aae finished a comment 2018-06-03 19:38:51 -07:00
Evan Tschannen e95f663ebc fix: the log router could pop too much data from the logs in rare situations 2018-06-03 19:34:24 -07:00
Evan Tschannen bf65e745a9 tlogs do not index tags for other localities 2018-06-01 22:51:08 -07:00
Evan Tschannen c519339adb avoid peeking from logs that do not match the tag’s locality 2018-06-01 18:42:48 -07:00
Evan Tschannen ce6a2f0563
Merge pull request #425 from bnamasivayam/leader-election-optimize
Optimize client and server connection times to cluster controller, es…
2018-06-01 18:35:27 -07:00