ad1dd4fd9b
Mark the storage servers that are continually lagging as unhealthy and so this will give the Data Distributor the chance to move data out of this server.
2020-03-31 18:25:39 -07:00
247ab84323
Merge branch 'master' of https://github.com/apple/foundationdb into add-data-distribution-metrics
2020-03-23 17:01:17 -07:00
e08f0201f1
merge release 6.2 into master
2020-03-17 12:51:47 -07:00
7adc916e18
Merge pull request #2806 from ajbeamon/improve-team-request-performance
...
Improve performance of get team requests.
2020-03-16 11:56:45 -07:00
700b13e5f8
Remember the best team from team requests, which will likely be the best again and can save us some computation.
2020-03-13 15:21:33 -07:00
12f2b32770
added additional logging in data distribution
2020-03-13 15:19:33 -07:00
9e99a00c8f
fix: do not use priority 0 left when calculating priorities for empty teams
2020-03-13 13:56:46 -07:00
555db50cd1
Avoid calling into SABTF so frequently. Use a cheaper call that only checks that shards exist.
2020-03-12 11:22:03 -07:00
303df197cf
Merge branch 'release-6.2'
...
# Conflicts:
# CMakeLists.txt
# bindings/c/test/mako/mako.c
# documentation/sphinx/source/release-notes.rst
# fdbbackup/backup.actor.cpp
# fdbclient/NativeAPI.actor.cpp
# fdbclient/NativeAPI.actor.h
# fdbserver/DataDistributionQueue.actor.cpp
# fdbserver/Knobs.cpp
# fdbserver/Knobs.h
# fdbserver/LogRouter.actor.cpp
# fdbserver/SkipList.cpp
# fdbserver/fdbserver.actor.cpp
# flow/CMakeLists.txt
# flow/Knobs.cpp
# flow/Knobs.h
# flow/flow.vcxproj
# flow/flow.vcxproj.filters
# versions.target
2020-03-06 18:22:46 -08:00
1128666840
added additional logging on the log router
2020-03-05 18:17:06 -08:00
e219c1671f
Merge branch 'release-6.2' into feature-dd-region-queue
...
# Conflicts:
# fdbserver/Knobs.h
2020-03-04 16:25:38 -08:00
125bd13198
fix: in multi-region configurations, the data distribution queue could start too much work, expecting that the remote region would contribute to the read workload
2020-03-04 14:17:17 -08:00
6296465e07
Make the DD priority associated with populating a remote region lower than machine failures
2020-03-04 14:07:32 -08:00
924d335aa7
Merge branch 'release-6.2'
...
# Conflicts:
# documentation/sphinx/source/release-notes.rst
# flow/Knobs.cpp
# flow/Knobs.h
2020-02-25 18:25:19 -08:00
c05c95cbe8
forgot to rename the knob
2020-02-25 15:47:39 -08:00
96258b9809
Merge branch 'release-6.2'
...
# Conflicts:
# documentation/sphinx/source/release-notes.rst
# fdbcli/fdbcli.actor.cpp
# fdbclient/ManagementAPI.actor.cpp
# fdbrpc/FlowTransport.actor.cpp
# fdbserver/ClusterController.actor.cpp
# fdbserver/DataDistribution.actor.cpp
# fdbserver/DataDistribution.actor.h
# fdbserver/DataDistributionQueue.actor.cpp
# fdbserver/KeyValueStoreMemory.actor.cpp
# fdbserver/MasterProxyServer.actor.cpp
# fdbserver/QuietDatabase.actor.cpp
# fdbserver/SkipList.cpp
# fdbserver/StorageMetrics.actor.h
# fdbserver/TLogServer.actor.cpp
# fdbserver/fdbserver.actor.cpp
# fdbserver/storageserver.actor.cpp
# fdbserver/workloads/KVStoreTest.actor.cpp
# flow/CMakeLists.txt
# flow/Knobs.cpp
# flow/Knobs.h
# flow/genericactors.actor.cpp
# flow/serialize.h
2020-02-21 19:09:16 -08:00
aa4d1357b3
handle the case that there is only one healthy team
2020-02-21 15:41:01 -08:00
457dbc5215
Update fdbserver/DataDistribution.actor.cpp
...
Co-Authored-By: A.J. Beamon <ajbeamon@users.noreply.github.com>
2020-02-21 15:39:17 -08:00
6a634652c4
Update fdbserver/DataDistribution.actor.cpp
...
Co-Authored-By: A.J. Beamon <ajbeamon@users.noreply.github.com>
2020-02-21 15:39:06 -08:00
08914a2acd
Once available space ratio falls below 0.3 avoid moving data to teams with less free space than the median team
2020-02-21 15:14:32 -08:00
819c55556c
More aggressively attempt to find teams that do not have low disk space
2020-02-20 16:47:50 -08:00
e1fb568fd1
Merge branch 'release-6.2' into dd-use-available-space
...
# Conflicts:
# fdbserver/DataDistribution.actor.cpp
# fdbserver/DataDistribution.actor.h
# fdbserver/DataDistributionQueue.actor.cpp
2020-02-20 16:12:42 -08:00
e4b483796d
Combine some logic that was doing similar computations for free space ratio.
2020-02-20 14:52:08 -08:00
4c9c736253
Data distribution uses available space instead of free space when evaluating whether processes are low on space and penalizing them.
2020-02-20 11:21:03 -08:00
3a1ba5a077
Rename variable for clarity
2020-02-20 10:59:52 -08:00
c164acb88d
Add new criteria to DD's GetTeamRequest that allow you to require shards be present on the team and that the team have a minimum free ratio. This avoids scenarios where the team chosen when processing the request is later rejected by the requestor, causing rebalancing movements to get stuck.
2020-02-20 09:32:00 -08:00
5a9d420cb7
Merge remote-tracking branch 'upstream/release-6.2' into release-merges/20200210
2020-02-10 10:02:05 -08:00
b8a252da40
Clarify the names of a couple trace fields
2020-02-10 08:15:00 -08:00
c9363e7e28
Merge branch 'master' of https://github.com/apple/foundationdb into add-data-distribution-metrics
2020-01-22 21:02:21 -08:00
3f9d9d8b84
Merge branch 'release-6.2'
...
# Conflicts:
# CMakeLists.txt
# cmake/FlowCommands.cmake
# documentation/sphinx/source/release-notes.rst
# fdbclient/StorageServerInterface.h
# fdbserver/DataDistributionTracker.actor.cpp
# fdbserver/MasterProxyServer.actor.cpp
# fdbserver/fdbserver.actor.cpp
# flow/Knobs.h
# flow/Platform.cpp
# versions.target
2020-01-16 18:37:47 -08:00
1d6ac716a1
Merge remote-tracking branch 'origin' into add-data-distribution-metrics
2020-01-15 13:20:04 -08:00
9b80498180
Added a trace event to warn if a shard is merged before enough time has elapses from becoming low bandwidth
2020-01-10 14:58:38 -08:00
c2608f0af9
fix: completeSources could be larger than the teamSize, so we need to check all completeSources
...
we do not need to track bestSize, since all teams in the list will be the same size
2020-01-10 14:46:40 -08:00
ab7071932f
Data distribution no longer attempts to pick teams which share members of the source unless the team matches exactly
2020-01-09 16:59:37 -08:00
83ad9caf54
implemented a load balancing algorithm which evens out the number of requests processes by each proxy
2020-01-08 01:59:01 -08:00
59738e8ef1
fixed compiler error
2019-11-22 16:19:34 -08:00
3c769fcf60
Merge branch 'release-6.2'
...
# Conflicts:
# CMakeLists.txt
# documentation/sphinx/source/release-notes.rst
# fdbserver/ClusterController.actor.cpp
# fdbserver/MasterProxyServer.actor.cpp
# versions.target
2019-11-22 15:39:19 -08:00
3a3ab5664b
fix: team trackers for bad teams that contain a removed servers must be cancelled or the cluster will falsely report those teams as failed
2019-11-22 10:20:13 -08:00
d4de608bb6
Fix OPEN_FOR_IDE build
2019-10-25 10:42:22 -07:00
f8e44d2f71
fix: If a storage server was offline, it would not be checked for being in an undesired dc
2019-10-23 23:04:39 -07:00
d2b6626d5c
Merge branch 'master' of https://github.com/apple/foundationdb into mark-ss-failed
2019-10-21 13:47:06 -07:00
688940b685
merge 6.2 into master
2019-10-21 11:43:46 -07:00
b1fd6b4443
addressed review comments
2019-10-18 09:43:25 -07:00
896701006f
addressed code review changes
2019-10-16 11:30:20 -07:00
86bcb84b45
Raised the data distribution priority of splitting shards above restoring fault tolerance to avoid hot write shards
2019-10-11 17:50:43 -07:00
34baa37e60
Merge branch 'master' of https://github.com/apple/foundationdb into mark-ss-failed
2019-10-10 10:14:58 -07:00
1bd6151f54
Update fdbserver/DataDistribution.actor.cpp
...
Co-Authored-By: Jingyu Zhou <jingyuzhou@gmail.com>
2019-10-09 21:17:03 -07:00
26e1d565f6
StorageServerTracker:Fix OOM bug caused by server healthyness toggles infinitely
...
When there is only one healthy team, the bug will set a server's status as unhealthy;
which causes the healthyTeam to 0, triggering StorageServerTracker to loop back;
which resets the server's status to healthy, and thus the healthyTeam to non-zero.
This pattern will cause infinite loop.
Infinite loop will prevent TraceEvent from flushing, which causes
TraceEvent to use most of memory and out-of-memory.
Kudos to JingYu Zhou (jingyu_zhou@apple.com ) who is the main contributor who found the bug!
2019-10-09 17:45:09 -07:00
eb41e32876
add extra dd safety check to deny exclude if only 1 team exists
2019-10-08 16:10:09 -07:00
ff563c1672
fixed trace line description
2019-10-07 12:45:27 -07:00
d146f1d636
Merge branch 'master' of https://github.com/apple/foundationdb into mark-ss-failed
2019-10-07 11:27:15 -07:00
22c3fa867c
Merge pull request #2074 from xumengpanda/mengxu/fix-correctness-bug
...
DD:fix:getTeam may fail to get a team when it can
2019-10-07 09:33:57 -07:00
cc41cd610d
force recheck in tracker if team contains failed server
2019-10-04 14:02:12 -07:00
6dae95a4ca
Update fdbserver/DataDistribution.actor.cpp
...
Co-Authored-By: A.J. Beamon <ajbeamon@users.noreply.github.com>
2019-10-04 10:02:11 -07:00
31332da57d
move option setting inside retry loop
2019-10-03 13:24:28 -07:00
abb46d527b
fixed implementation of watches in TrackExcludedServers
2019-10-03 13:18:05 -07:00
d96a7b2c69
Merge branch 'master' of https://github.com/apple/foundationdb into mark-ss-failed
2019-10-03 09:47:45 -07:00
5286523a43
StorageServerTracker:Distinguish wrongDC from invalidLocality
2019-10-02 14:50:53 -07:00
9e7dfa358c
Replace isCorrectDC with isCorrectLocality
...
This incorporates the change for defending DD from misconfigured
locality entries.
The check for misconfigured locality was in keyValueStoreTypeTracker,
but the storage engine switch PR moves the isCorrectDC checking out of
the tracker and move it into storageServerTracker
2019-10-02 14:05:43 -07:00
d0147e5e5d
Merge branch 'release-6.2' into mengxu/merge-release620-to-master-v3
...
Resolved Conflicts:
documentation/sphinx/source/release-notes.rst
fdbserver/DataDistribution.actor.cpp
versions.target
2019-10-02 13:22:56 -07:00
4b5080fbea
added a few more missing data distribution priorities
2019-09-27 19:39:53 -07:00
324d0bd3b0
Merge branch 'release-6.2' of github.com:apple/foundationdb into feature-cleanup-mutations
2019-09-27 19:15:14 -07:00
3bb62e008c
lowered the priority of some delays in data distribution so that the process will prefer other work
2019-09-27 18:33:13 -07:00
efa2f5df6b
avoid double overwrite on failed servers and reordered removal of keys
2019-09-26 13:52:53 -07:00
450a09e117
Code Review Changes
2019-09-24 15:48:50 -07:00
32ebd08f9f
DD:Trigger storage recruitment when an invalid address locality is corrected
2019-09-24 13:35:38 -07:00
515689d07b
Update fdbserver/DataDistribution.actor.cpp
...
Co-Authored-By: A.J. Beamon <ajbeamon@users.noreply.github.com>
2019-09-18 14:45:18 -07:00
d175b61a62
DD:Trace when invalid locality is corrected
...
Change getWorkers param from txn handler to db
2019-09-18 13:52:54 -07:00
40ad6f0931
created new reply types to work with flatbuffer serialization requirements
2019-09-18 13:40:18 -07:00
471e283128
Merge branch 'master' of https://github.com/apple/foundationdb into mark-ss-failed
2019-09-18 11:49:07 -07:00
93bbc26a35
set:erase:Use return value of erase iterator as next iterator
2019-09-17 15:28:30 -07:00
d2fd1f4931
DD:MisconfiguredLocality:Fix review comments
2019-09-17 13:04:21 -07:00
37d2318eed
DD:Handle worker with incorrect locality
...
When a worker has incorrect locality, the worker will be excluded from
storage recruitment.
When the worker has its locality corrected by system operators,
the worker will be reincluded for storage recruitment.
2019-09-14 12:12:56 -07:00
c3960aba17
DD:initializeStorage:Exclude worker with invalid locality
2019-09-13 22:05:41 -07:00
75460089e1
DD_VALIDATE_LOCALITY:Add comment for our future selves
...
When we add simulation test that misconfigure a cluster by not setting some
locality entries, we should set DD_VALIDATE_LOCALITY always true.
Otherwise, simulation tests may fail.
2019-09-13 16:26:54 -07:00
78b8e48cef
DD:ValidLocality:Resolve review comment
2019-09-13 15:35:16 -07:00
e1dcdbf3d2
LocalityData:Remove verbose check for valid locality
2019-09-13 15:11:13 -07:00
8970d9858b
DD:isValidLocality:A generic way to check any replicationPolicy
2019-09-13 14:55:51 -07:00
1196841b3d
DD:IsValidLocality:Clang format
2019-09-13 13:56:43 -07:00
e8878b16d4
DD:Valid locality includes an empty but set locality entry
2019-09-13 13:55:46 -07:00
1596e2e4a5
DD:TCMachine:Use processID as machineID if zoneID is unset
2019-09-13 13:43:41 -07:00
3ad7e3adb3
DD:DD_VALIDATE_LOCALITY:Guard the checking of locality validity
2019-09-13 13:19:35 -07:00
90d6a27a0d
DD:IsValidLocality:Consider configured replica policy
2019-09-13 12:04:49 -07:00
52f6297b52
DD:Introduce isValidLocality
...
A server or machine has a valid locality only if it sets correct
locality entries.
Build teams should only use the valid locality servers or machines
2019-09-13 11:30:26 -07:00
b495cc697b
Merge branch 'release-6.2'
...
# Conflicts:
# CMakeLists.txt
# documentation/sphinx/source/release-notes.rst
# versions.target
2019-09-13 09:25:08 -07:00
cc41f3e2fc
fix: an unhealthy server with a low number of teams could cause data distribution to build every possible teams
2019-09-12 14:18:10 -07:00
2081302a89
DD:GetTeam:Add trace event GetTeamReturnEmpty
...
Log BestTeamStuck reason when we have healthy teams
but they do not have healthy free space
2019-09-11 14:48:39 -07:00
3c4dc1003d
DD:clang-format the PR
2019-09-11 11:16:29 -07:00
a83afa5f64
DD:BgDDValleyFiller:Do not move shard to a team with no healthy space
2019-09-11 11:15:00 -07:00
8e47a05326
DD:TeamInfo:Add HasHealthyFreeSpace field
2019-09-09 18:42:18 -07:00
259ca36aa2
DD:Random pick from valid teams
2019-09-09 18:28:39 -07:00
0b785e5c1c
DD:getTeam may fail to get a team when it can
...
Due to randomness, when unhealthy teams are majority while there still
exists healthy teams, getTeam function may be unlucky to find
any feasible (ok) team, which leads to BestTeamStuck situation.
This commit increases the tries from 10 to 20.
A long-term solution may first find all feasible teams and choose a random
one from them. Since This can affect the statistics of which team is picked.
So it is not included in this commit.
Non-functional change: This commit removes unneeded printf introduced by
fast restore PR 1404.
2019-09-07 20:08:58 -07:00
def1294aab
moved wait statement to avoid context switch between check and execution
2019-09-04 12:42:27 -07:00
8f9ba3bc09
StorageEngineSwitch:Remove unused code
2019-09-03 17:18:56 -07:00
dbc993a138
change use of reference to raw ptr to avoid unwanted delay of destructors
2019-09-03 16:27:14 -07:00
bd80a67d46
Merge branch 'master' into mengxu/storage-engine-switch-PR-v2
2019-09-03 14:11:33 -07:00
5d87443323
improved error msgs for snapshot cmd
2019-08-27 16:43:52 -07:00
202900bd79
adjusted priority of relocateShard requests if team contains failed server
2019-08-27 14:39:44 -07:00
93079c6657
always attempt to send RelocateShard requests if team has a failed server
2019-08-27 14:39:44 -07:00
3666c0c776
added more trace lines and added timeout to safety check in test workload
2019-08-27 14:39:44 -07:00
Balachandar Namasivayam