1
0
mirror of https://github.com/apple/foundationdb.git synced 2025-06-02 03:12:12 +08:00

17355 Commits

Author SHA1 Message Date
Yao Xiao
2a1a5af939 Update release-notes-630.rst
Resolve comments
2021-10-21 12:40:10 -07:00
Yao Xiao
1a24ad33dd Update documentation/sphinx/source/release-notes/release-notes-630.rst
Co-authored-by: A.J. Beamon <aj.beamon@snowflake.com>
2021-10-21 12:40:10 -07:00
Yao Xiao
ac23751608 Update release-notes-630.rst 2021-10-21 12:40:10 -07:00
Renxuan Wang
f39e34cabe Use --enable-prof switch when building jemalloc。
I got errors like `<jemalloc>: Invalid conf pair: prof:true` when trying to use jemalloc. Referring to https://stackoverflow.com/questions/27422508/heap-dump-fails-with-jemalloc-mcllctl, seems that we are missing out this flag. https://github.com/jeffgriffith/native-jvm-leaks#building.
2021-10-21 10:08:55 -07:00
Trevor Clinkenbeard
c69364d5aa
Verify that cluster is fully recovered in quietDatabase check ()
* Verify that cluster is fully recovered in quietDatabase check

* Add trace event to waitForQuietDatabase
2021-10-21 09:01:52 -07:00
Zhe Wu
0cf829ef91 Reduce restore error message 2021-10-20 14:02:48 -07:00
Kao Makino
4f0732b510
Merge pull request from sfc-gh-qhoang/qhoang-support-multithreaded-client-mako
Support multithreaded client mako
2021-10-20 09:34:49 -07:00
Markus Pilman
9730f670e1
Merge pull request from sfc-gh-ajbeamon/improved-logging-failed-network-thread
Add some additional logging if the network thread finishes, fails with an error, gets stopped, or is blocked
2021-10-19 13:10:16 -06:00
QA Hoang
087533a6fc Merge branch 'master' into qhoang-support-multithreaded-client-mako 2021-10-19 12:02:49 -07:00
A.J. Beamon
6cf731d5aa Add comment to run loop blocked message about timestamp order 2021-10-19 10:05:29 -07:00
A.J. Beamon
abab45760d Add some additional logging if the network thread finishes, fails with an error, gets stopped, or is blocked. 2021-10-19 10:05:29 -07:00
QA Hoang
8d7d8a40ae init client_threads_per_version to 0 2021-10-19 09:48:26 -07:00
Aaron Molitor
e9091f91c9 make build-ycsb-docker.sh executalbe 2021-10-18 14:55:39 -07:00
Aaron Molitor
7129c2b528 update packaging to align across branches 2021-10-18 14:31:46 -07:00
Trevor Clinkenbeard
504d0b71b2
Fix invalid memory access when dataDistribution actor is cancelled ()
* Fix valgrind error when dataDistribution actor is cancelled

* Trace Sev30 when dataDistribution actor is cancelled outside of simulation

* Rethrow actor_cancelled error in dataDistribution catch block
2021-10-18 14:21:29 -07:00
Aaron Molitor
82bdd03268 add ycsb docker build script 2021-10-18 13:57:39 -07:00
Aaron Molitor
73525085cf cleanup docker build scripts 2021-10-18 13:57:39 -07:00
sfc-gh-tclinkenbeard
421dee532c Add const qualifiers in KeyValueStoreRocksDB.actor.cpp 2021-10-18 13:40:47 -07:00
sfc-gh-tclinkenbeard
9e06b6e6e3 Make IClosable interface const-correct 2021-10-18 13:40:47 -07:00
Daniel Smith
faf16fb29e
Merge pull request from Daniel-B-Smith/ikvs-read-type
Add an enum to IKeyValueStore to indicate the source/priority of the read
2021-10-18 13:21:20 -04:00
Daniel Smith
9713a14ef1 Reverse order of read type and debug ID args 2021-10-18 12:23:09 -04:00
A.J. Beamon
507a09893c
Add ClientCount to ClusterControllerMetrics () 2021-10-17 20:47:11 -07:00
Daniel Smith
df53cc9580 Add an enum to IKeyValueStore to indicate the source/priority of the read 2021-10-15 14:35:59 -04:00
Aaron Molitor
0e4a6cc921 add ycsb Dockerfile 2021-10-15 00:27:20 -07:00
He Liu
5688672b50
Merge pull request from liquid-helium/cc-interface-changes
Added new endpoints in Cluster Controller interface.
2021-10-14 17:03:36 -07:00
QA Hoang
5af2915cea added more input validation logic 2021-10-12 16:38:42 -07:00
He Liu
ddd08a99d7 fmt 2021-10-12 16:27:51 -07:00
He Liu
422d8a041d Added comments. 2021-10-12 15:43:18 -07:00
He Liu
8174c57714
Merge pull request from liquid-helium/add-logs
Added logs for worker_removed() errors in SS.
2021-10-12 15:16:11 -07:00
Yao Xiao
e543570343
Add locality cache size to TransactionMetrics. ()
Co-authored-by: Yao Xiao <yaoxiao@Yaos-MacBook-Pro.local>
2021-10-12 13:10:15 -07:00
He Liu
9f974ef21f Added logs for worker_removed() errors in SS. 2021-10-12 10:12:20 -07:00
Josh Slocum
0dafb95bbf Fixing tss private mutations ranges 2021-10-11 18:14:29 -07:00
Yao Xiao
c8e6819a10
Add FastAlloc memory utilization trace. ()
Co-authored-by: Yao Xiao <yaoxiao@Yaos-MacBook-Pro.local>
2021-10-11 15:06:43 -07:00
Markus Pilman
5066d67dec
Merge pull request from sfc-gh-ljoswiak/fixes/tracing-various
Add knob to control distributed trace recording percentage
2021-10-11 14:03:53 -06:00
Markus Pilman
cd6049f178
Merge pull request from sfc-gh-ljoswiak/fixes/sampling-crashes
Fix some common sampling crashes at high sampling frequencies
2021-10-11 14:02:11 -06:00
Steve Atherton
2ebaddcc1e
Bug fix: CommitQuorum::addRequestActor() accesses self after destruction due to ignoring actor_cancelled error. () 2021-10-11 12:17:09 -07:00
QA Hoang
b4f56c7f76 added args validation 2021-10-11 11:11:48 -07:00
QA Hoang
85dbbd325d added debug msg 2021-10-11 11:06:11 -07:00
He Liu
2e03dd02cd Added new endpoints in Cluster Controller interface.
moveShard is for manual data move;
repairSystemData is for repairing system metadata when some or all of
storage servers hosting system metadata are lost.

The complete implementation can be fund in .
2021-10-11 10:41:46 -07:00
Steve Atherton
6abeb89f95
Merge pull request from sfc-gh-satherton/dd-crash-fix
Bug fix: rare crash in DataDistribution
2021-10-11 10:30:36 -07:00
Steve Atherton
14efc4c251 Bug fix which only affects simulation: dataCopy was created after a potential wait so the source memory may have already been released. 2021-10-11 08:23:15 -06:00
Steve Atherton
f339b603a5 Bug fix: printSnapshotTeamsInfo() could crash when looking up status for a storage server that was very recently added because its entry in server_status was not yet created.
Bug fix:  printSnapshotTeamsInfo()'s local server_status map would not see status updates for server UIDs that already existed in the map.
2021-10-10 01:48:31 -07:00
Steve Atherton
0cce774325
Merge pull request from sfc-gh-satherton/kvs-write-version
Refactored how Redwood handles commit version
2021-10-09 20:26:52 -07:00
Steve Atherton
13e6ac7c53
Merge pull request from sfc-gh-satherton/queue-read-during-shutdown-fix
Fix rare Redwood crash after shutdown
2021-10-09 17:43:01 -07:00
Markus Pilman
4fbc193e19 fix priority 2021-10-09 16:40:24 -06:00
Markus Pilman
9077e15b11 Address review comments 2021-10-09 16:40:24 -06:00
Markus Pilman
4e723ba3d9 Apply suggestions from code review
Co-authored-by: A.J. Beamon <aj.beamon@snowflake.com>
2021-10-09 16:40:24 -06:00
Markus Pilman
ffdba4a133 assert we never get a stale read version 2021-10-09 16:40:24 -06:00
Markus Pilman
424b35de63 verify FLAG_USE_PROVISIONAL_PROXIES on the server 2021-10-09 16:40:24 -06:00
Steve Atherton
abc14c4921 Redwood crash after shutdown which occurs if a queue page disk read was in progress during pager shutdown but the callback was fired after shutdown completed. 2021-10-09 02:04:23 -07:00