4086e3a275
Merge pull request #3999 from dongxinEric/misc/attach-primary-peek-location-to-logRouterMetrics
...
Add a way to decorate the trace event created by `traceCounters` with more details.
2020-11-05 09:54:39 -08:00
5d7ec6555a
Update fdbserver/LogRouter.actor.cpp
2020-11-04 16:34:32 -08:00
93f17e4a9f
Merge pull request #4004 from vishesh/task/issue-3799-conn-reset-log
...
Make ConnectionReset logs query friendly #3977
2020-11-04 16:29:10 -08:00
7b28de8a41
Add IDs to ConnectionReset TraceEvents
2020-11-04 14:06:49 -08:00
44cdc4dfa6
Update fdbserver/LogRouter.actor.cpp
2020-11-04 09:44:28 -08:00
22b16302c3
Make ConnectionReset logs easier to query #3977
...
All TraceLogs that are related to ConnectionReset should be prefixed with
ConnectionReset. This should make it easy to query and aggregate by address and
role.
2020-11-02 15:10:51 -08:00
32c87bbb33
Lightweight, power of two spaced histogram implementation + automatic reporting
2020-11-02 11:13:16 -08:00
46150d22c3
Attach generation(recovery count) to TLog metrics and LogRouter metrics.
2020-11-01 11:24:23 -08:00
d302f60925
Fix build error.
2020-10-30 17:06:22 -07:00
566365accd
Fix typo.
2020-10-30 16:28:05 -07:00
af7e65110f
Allow the caller to decorate role metrics trace event with more details.
2020-10-30 16:20:08 -07:00
f2a6a6101e
Fix build error.
2020-10-30 13:43:39 -07:00
eead86f006
Add primary peek location, aks paring TLog ID to LogRouterMetrics.
2020-10-30 11:42:09 -07:00
cedcfc7d87
Removed unused uid field inside ParallelTCInfo
2020-10-26 16:48:09 -07:00
be7944773f
Fix a typo
2020-10-26 16:44:52 -07:00
9ef29d0cea
Changed getTeamID() to return a string instead of UID as suggested by reviews.
2020-10-26 16:44:52 -07:00
9b5a02b552
Resolve review comments
2020-10-26 16:44:52 -07:00
bec2cfb167
Fix typos.
2020-10-26 16:44:52 -07:00
21ad448ad3
Fix macOS build.
2020-10-26 16:44:52 -07:00
0bc51bb780
Resolve review comments
2020-10-26 16:44:50 -07:00
7ebb2e5c09
Piggy back this PR to polish more TraceEvent by:
...
- Making it clear that it's tracking machine team info or server team info
- Added ID to both machine team and server team for better trackability
- Attach distributor id to some trace events.
2020-10-26 16:44:09 -07:00
c037bfd001
Added detailed logging when there is no servers left in a server team, because that may indicate a data loss incident.
2020-10-26 16:44:07 -07:00
b0d78ecf37
Merge pull request #3912 from dongxinEric/misc/log-detailed-teams-info-in-dd
...
Added a way to print detailed information about team collection for d…
2020-10-24 19:06:10 -07:00
6395b76d8c
Address more review comments
2020-10-23 15:29:08 -07:00
f757cae786
Address review comments
2020-10-23 14:01:53 -07:00
7e13c99d3c
Merge pull request #3938 from sears/rocksdb_fixes
...
Rocksdb fixes
2020-10-23 15:11:33 -04:00
c7345b8ab5
Fix static linking of lz4
...
Conflicts:
fdbserver/CMakeLists.txt
2020-10-21 10:26:55 -07:00
6a6ea56596
Restore line that stores the data lag seconds of a storage server. This value is used to add a data lag message to status.
2020-10-20 10:12:00 -07:00
1b82adde3a
Use lz4
...
Conflicts:
cmake/CompileRocksDB.cmake
fdbserver/CMakeLists.txt
2020-10-19 13:33:43 -07:00
8d0aa02a63
Do not periodically print detailed DD teams info
2020-10-16 16:11:14 -07:00
92e31dd338
Address review comments
2020-10-15 15:25:00 -07:00
1d43729cc9
Added a way to print detailed information about team collection for debugging.
2020-10-15 10:01:56 -07:00
89469921bb
Merge pull request #3891 from etschannen/feature-reset-proxy-connections
...
Reset a proxy's network connection with the master or resolvers if it is too far behind
2020-10-12 11:21:24 -07:00
1378ecba4d
If a proxy is sufficiently far behind, reset network connections to attempt to fix the problem
2020-10-11 23:06:26 -07:00
91a8367acb
Avoid slow task in ~DataDistributionTracker
2020-10-01 11:44:55 -07:00
d750ed303a
fixed whitespace
2020-08-30 22:34:03 -07:00
331a49a62a
do not allow a proxy to reset a connection with the logs immediately upon starting up
2020-08-30 18:50:19 -07:00
ce1139e588
added missing dumpToken trace events
2020-08-27 17:17:27 -07:00
86cba16d7d
Added logging whenever a wait failure client triggers
2020-08-27 17:07:05 -07:00
74edf61745
Merge pull request #3674 from sfc-gh-anoyes/anoyes/read-conflict-dbLocked
...
Add read conflict on \xff/dbLocked
2020-08-26 12:18:42 -07:00
507c67c930
Added additional information to trace events
2020-08-26 11:42:23 -07:00
28cb5f242c
another fix
2020-08-26 11:01:40 -07:00
e81ccd2dc9
another compiler fix
2020-08-26 10:59:06 -07:00
e531046b53
fix compiler errors
2020-08-26 10:56:21 -07:00
fd1a4304fa
fix: made ConnectionResetInfo reference counted
2020-08-26 10:53:17 -07:00
4dc2975f5a
Only switch primary if extraDB != nullptr
2020-08-20 18:06:24 +00:00
30c086ed96
Improve comment
...
Also use push_back. The lifetime of databaseLockedKey and
databaseLockedKeyEnd will outlive trRequest.arena, since they're
literals.
2020-08-20 18:05:03 +00:00
9cf3041393
Add read conflict for metadata txs that aren't lock-aware
2020-08-20 02:52:37 +00:00
fbc97a12b8
Add LockDatabaseFrequently workload
2020-08-19 23:17:38 +00:00
8ede143941
Track tlog push latencies and reset connections if they are above 500ms
2020-08-18 08:43:14 -07:00
Xin Dong