d001820219
FastRestore:getVersionBatchState can be called before VB is init
2020-02-28 14:47:11 -08:00
7c003bfa70
FastRestore:Disable isSchedulable for test
2020-02-28 14:27:45 -08:00
62b9043ff6
FastRestore:DB can be destroyed before master unlock it in simulation
...
Because retore roles run as workload in simulation,
they do not know when DB is destroyed by the backup and restore test workload.
So if DB is destroyed earlier than restore master unlocks DB, which is rare,
restore master should abort the unlocking DB step.
2020-02-28 14:25:58 -08:00
6054c05963
Merge branch 'release-6.2'
...
# Conflicts:
# CMakeLists.txt
# documentation/sphinx/source/release-notes.rst
# fdbserver/fdbserver.actor.cpp
# versions.target
2020-02-28 12:11:05 -08:00
1be1145bc6
FastRestore:Fix corrupted trace event in FastRestoreVersionBatchProgressState
2020-02-28 00:23:58 -08:00
89b121ae25
FastRestore:Fix duplicate type in traceRoleVersionBatchProgress
2020-02-28 00:08:01 -08:00
eaf340652f
FastRestore:ensure setVersionBatchState after version batch is initialized
2020-02-27 23:50:09 -08:00
22b34bc609
FastRestore:getVersionBatchState can be called before version batch is initialized
2020-02-27 23:45:48 -08:00
6018b64d73
FastRestore:Fix undefined ref to vtable error
2020-02-27 21:18:10 -08:00
f4cd0ef74f
FastRestore:Apply clang-format
2020-02-27 20:59:56 -08:00
a6e66da29f
FastRestore:fix compilation error for version batch state
2020-02-27 20:59:34 -08:00
fe8b8bbbff
FastRestore:Change vb state to class from enum
2020-02-27 20:15:25 -08:00
d77177367c
FastRestore:Track each ongoing version batch progress state for applier and loader roles
2020-02-27 19:47:22 -08:00
b8a9d49b2e
FastRestore:Ensure minimum delay between heart beats
2020-02-27 18:46:31 -08:00
2657d41bb2
FastRestore:Add debug msg when memory is over threshold
2020-02-27 18:32:11 -08:00
13e72f7b3b
Merge pull request #2605 from dongxinEric/fix/1977/report-inability-to-flush-trace-log
...
Report inability to flush trace logs.
2020-02-27 12:36:55 -08:00
c3299b8ebe
if tls cannot be initialized, throw an error from createDatabase
2020-02-26 18:53:06 -08:00
bf5a95e6df
Merge commit 'dc39bdfbbf94a7f470386f439df08c044d08d90c' into feature-tls-environment-vars
...
# Conflicts:
# flow/Net2.actor.cpp
2020-02-26 18:02:56 -08:00
97d7eb49b5
FastRestore:Master:Report unavailable role periodically
...
Ping all restore roles and report unavailable ones.
2020-02-26 16:14:55 -08:00
d1598e7c99
set_verify_peers throws an error instead of returning a value
2020-02-26 16:06:16 -08:00
2586bade68
re-added support for configuration TLS options with environment variables
2020-02-26 15:33:48 -08:00
fbb6e8f39d
FastRestore:Create low memory situation in simulation on purpose
2020-02-26 14:54:38 -08:00
06495b90ae
FastRestore:Loader:Use isSchedulable to guard OOM
...
And trigger delayed actors that are blocked on memory to recheck memory.
2020-02-26 14:35:05 -08:00
a354f6ffa2
FastRestore:Applier:Use isSchedulable to guard OOM
2020-02-26 14:12:56 -08:00
ca726fc68e
FastRestore:Introduce OOM protection
...
An actor is schedulable to run if the current worker has enough resourc, i.e.,
the worker's memory usage is below the threshold;
Exception: If the actor is working on the current version batch, we have to schedule
the actor to run to avoid dead-lock.
Future: When we release the actors that are blocked by memory usage, we should release them
in increasing order of their version batch.
2020-02-26 14:09:18 -08:00
0f5c999d4b
Better containment of boost errors related to TLS.
2020-02-26 12:26:43 -08:00
087c6fa33d
Merge branch 'master' into feature-redwood
2020-02-26 12:25:04 -08:00
fbf5020af9
FastRestore:Applier:Add fetchKeys counter
2020-02-26 11:37:40 -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
12b5064041
a high free_space_ratio_cutoff is not needed anymore because avoid teams with low disk space is no longer the responsibility of getLoadBytes()
2020-02-25 15:47:10 -08:00
6e7d2ff7dd
prevent the proxy from delaying too long based on an incorrect estimate of the compute time
2020-02-25 15:46:13 -08:00
f20619c9fb
Resolve review comments. Changed how issues got cleared
2020-02-25 15:39:51 -08:00
fce71e4516
Added a TODO for the usage of 'issues' in 'monitorServerDBInfo'
2020-02-25 15:39:38 -08:00
090c89e90a
Addressed review comments. Fix the bug where issues on a worker may be wrongly cleared by subsequent GetDBinfo request.
2020-02-25 15:39:38 -08:00
6325c40336
Apply suggestions from code review
...
Co-Authored-By: A.J. Beamon <ajbeamon@users.noreply.github.com>
2020-02-25 15:39:09 -08:00
288e95c7e1
Reallocate the issues set after each get. Changed an issues name to be accurate
2020-02-25 15:39:09 -08:00
f4f860bfa8
Changed issue reporting to be thread safe. Also changed the liveness ping to be thread safe.
2020-02-25 15:38:14 -08:00
a6580dc15f
Added the ability to ping a trace log writer thread and the monitoring in worker.actor.cpp. The current solution is simple a loose check. We can change this to be accurate check by using 'pthread_kill(writer_thread, 0)'
2020-02-25 15:37:53 -08:00
0b0414fb94
Addressded review comments. Change the issue reporting from 'ITraceLogWriter' to be a more generic way.
2020-02-25 15:37:53 -08:00
034dfe5e42
Now the inability to flush trace logs will be reported to both 'stderr' and also the status json object.
...
- Since the first flush failure, if the accumulated consecutive failure count exceeds the value defined in knobs, it will trigger the current worker process to report this issue via the 'GetServerDBInfo' interface of the cluster controler
- A successful flush will reset the accumulated counter.
Notice that the current solution does not take the time into consideration. The assumption is that flush failures tend to only happen in a clustered manner. The intermittent, but short, periods of flush failures are not considered as a problem since the memory pressure built by them should be negligible.
2020-02-25 15:37:32 -08:00
636c256606
Simplified return value of CommitSubtree to a single ChildLinkSetRef, which doesn't contain a version. Removed unnecessary record copying in several places in CommitSubtree.
2020-02-24 15:18:34 -08:00
e79faae175
Add some more tail latencies to the latency trace events in the ReadWrite workload.
2020-02-24 11:06:13 -08:00
72a549df14
Added basic mutation buffer insert/lookup test.
2020-02-24 00:31:09 -08:00
1e45bb2be3
Further refactoring of MutationBuffer to make upcoming ART implementation easier to plug in. Key and mutation for an iterator position are now accessed via member functions which gives the implementation an opportunity to decode or reconstitute keys on-demand.
2020-02-23 23:44:15 -08:00
0f64505d0b
Merge branch 'release-6.2' of github.com:apple/foundationdb
...
Needed to pull in changes to build docker
2020-02-23 23:27:53 -08:00
eb7016a09a
Some more refactoring of MutationBuffer so it will be easier to have multiple implementations.
2020-02-23 22:48:43 -08:00
710bc3ecb1
Added optimization for single key clears to avoid an additional mutation buffer boundary key insertion. Made mutation buffer members private and created additional member functions to avoid needing access to the private members.
2020-02-23 22:24:16 -08:00
712aa27896
Merge branch 'release-6.2' of github.com:apple/foundationdb into feature-redwood
2020-02-23 00:30:27 -08:00
3618dff07b
Corrected statement in test output to reflect new logic.
2020-02-23 00:21:39 -08:00
078f85fea7
Bug fix. Pager could return an an oldest version that is no longer readable because it is being expired in the current commit cycle.
2020-02-23 00:15:35 -08:00
648040b070
Removed versioned records from the mutation buffer and commit path because support for this was incomplete and has high overhead costs when only exact committed versions are being read. The BTree could still contain versioned records in the future as the read cursors still support this, but some optimizations were added for when all internal records are at version 0 which is the case and has been for quite some time. Mutation buffer now stores keys and values in one arena per buffer.
2020-02-22 20:21:06 -08:00
65fbe0d0bc
revert AcceptSocket priority change because of bad performance results
2020-02-21 19:22:14 -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
f1ec780b31
Merge branch 'release-6.2' of github.com:apple/foundationdb into feature-redwood
2020-02-21 17:43:11 -08:00
4c696d5bf2
Merge branch 'release-6.2' into dd-better-rebalance-logging
...
# Conflicts:
# fdbserver/DataDistributionQueue.actor.cpp
2020-02-21 17:41:00 -08:00
dfa5f76c01
Remove unused parameter. Don't put check for g_network presence in ASSERT_WE_THINK.
2020-02-21 16:28:03 -08:00
5a1884d054
Leaf page modifications are now made incrementally until the DeltaTree is out of space or the tree is too imbalanced, at which point it falls back to a linear merge and page(s) rebuild.
2020-02-21 16:26:44 -08:00
8506bce493
FastRestore:Reuse getBatchReplies for sendBatchRequests
...
Remove old sendBatchRequests and getBatchReplies as well.
2020-02-21 16:15:53 -08:00
4dd206b1b8
FastRestore:Use new getBatchReplies that profile request latency
2020-02-21 15:59:57 -08:00
505997ba0a
FastRestore:Switch to new sendBatchRequests that tracks performance and straggler
2020-02-21 15:45:32 -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
05ea79f584
FastRestore:Profile performance for getBatchReplies
...
Generic approach to profile getBatchReplies performance
and detect straggler.
2020-02-21 15:20:22 -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
ab2dd36bdc
FastRestore:Generic way to detect stragger
2020-02-21 14:30:08 -08:00
e422874758
fix: reboot does not work un unreliable processes
2020-02-21 14:29:42 -08:00
2e699fef55
Don't suppress actor cancellation because we've already initialized the trace event by adding details.
2020-02-21 11:28:59 -08:00
6810a03283
Add more logging to valley filler and mountain chopper
2020-02-21 10:55:14 -08:00
9042cab7bc
Changed ordering of link libraries
2020-02-21 08:56:52 -08:00
87751df40a
Fixed problem with linking pthread
2020-02-21 08:45:39 -08:00
a27ea63500
Merge branch 'release-6.2' into feature-boost-ssl
2020-02-20 23:06:22 -08:00
59ff782927
fix: only delete the processId file on binaryReader errors
2020-02-20 23:04:39 -08:00
986e533c8c
Bug fix, DeltaTree was not updating numItems after erase / insert operations which only change the deleted flag of an existing equivalent entry.
2020-02-20 22:49:36 -08:00
6f1d3ccd35
Merge branch 'release-6.2' into feature-boost-ssl
2020-02-20 20:03:40 -08:00
f04e311a1e
Merge commit 'b46d6e25e24993ab5a5f04091fd3235050b7cd09' into feature-boost-ssl
...
# Conflicts:
# fdbserver/SimulatedCluster.actor.cpp
# flow/Net2.actor.cpp
2020-02-20 17:36:38 -08:00
927cff3317
Report errors on TLS misconfigurations ... or at least try to.
2020-02-20 16:57:29 -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
9e84fa965d
Merge pull request #2703 from ajbeamon/fix-stuck-dd-rebalancing
...
Fix issue with rebalancing data movement doing no work
2020-02-20 15:56:04 -08:00
8222089d21
Merge branch 'master' into mengxu/fast-restore-enable-buggify-test-PR
2020-02-20 15:34:20 -08:00
92c38f49bb
Merge pull request #2665 from xumengpanda/mengxu/fast-restore-applier-multi-applying-PR
...
Performant restore [17/xx]: Improve performance of restore appliers
2020-02-20 15:33:21 -08:00
d7c841a28a
Merge pull request #2589 from etschannen/feature-proxy-delay
...
Improve version pipelining on the proxy
2020-02-20 15:23:30 -08:00
e4b483796d
Combine some logic that was doing similar computations for free space ratio.
2020-02-20 14:52:08 -08:00
8b768e66df
Merge pull request #2694 from dongxinEric/feature/2663/specialize-policy-for-zoneid-in-cc
...
Added a specialized algorithm for PolicyOne and PolicyAcross(,'zoneId…
2020-02-20 14:46:23 -08:00
8129f74a10
Merge pull request #2698 from etschannen/feature-recruit-delay
...
The CC waits until no new workers register before starting a bad recruitment
2020-02-20 14:42:37 -08:00
7d54acf4ca
removed an unnecessary yield
2020-02-20 14:41:49 -08:00
6bd4703a9f
FastRestore:Resolve review comments
2020-02-20 14:27:34 -08:00
574e88ba8e
updateGoodRemoteRecruitmentTime was unnecessary because the only way findRemoteWorkers would return would be after a new server has joined which already resets goodRemoteRecruitmentTime
2020-02-20 13:46:22 -08:00
5586e6f6d8
Merge pull request #2697 from etschannen/feature-correctness-fixes
...
A variety of correctness fixes
2020-02-20 13:32:18 -08:00
4f1301b2dd
Merge pull request #2583 from etschannen/feature-keep-status-connected
...
Clients should not disconnect from the CC after fetching status
2020-02-20 13:12:30 -08:00
fcbdcda490
Merge pull request #2650 from ajbeamon/fix-reverse-range-read-byte-limit-bug
...
Fix reverse range read performance bug
2020-02-20 12:47:17 -08:00
6d9decdf59
Merge pull request #2672 from ajbeamon/improve-tlog-role-event
...
Improve TLog "Role" event
2020-02-20 12:45:25 -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
08c318d28a
re-added the connect lock in the fdbcli so that the timeout is not spent before a connection has been initiated (because of the handshake lock)
2020-02-20 10:43:34 -08:00
893729e3a8
Bug fix, Redwood was not using direct IO where available.
2020-02-20 10:29:20 -08:00
99095c9224
Again make Clang happy.
2020-02-20 09:50:22 -08:00
298d6cb3d7
Address review comments.
2020-02-20 09:34:01 -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
52ba093f23
BackupAndParallelRestoreTest:Apply Steve change in BackupCorrectness on 604ad062d5
...
The backup container behavior was changed at commit 604ad062d5
2020-02-20 08:07:03 -08:00
0507e9703c
FastRestore:Disable abort fast restore test
2020-02-20 06:54:14 -08:00
761da5a059
code cleanup
2020-02-19 17:59:45 -08:00
898a1ea3ed
FastRestore:Applier:Handle mutations at same version
2020-02-19 17:29:46 -08:00
fbd45963d8
The cluster controller waits until no new workers register for 1.0 before starting a bad recruitment
2020-02-19 16:48:30 -08:00
d5d26f589f
FastRestore:Cosmetic change to improve code readability
2020-02-19 15:43:51 -08:00
7897b1658f
FastRestore:Applier:Rename new apply actor names
2020-02-19 15:29:32 -08:00
e4258d73f5
FastRestore:Applier:Remove applying actors that do not have good perf
2020-02-19 15:27:59 -08:00
cf4efca852
fix: buffered cursor should always make sure all of the sub-cursors are completely exhausted before calculating minVersion. It is not legal to advance a cursor version past an epochEnd (+100 million versions) without also returning the epochEnd mutation, or the storage servers might not be able to rollback far enough because the end of the previous epoch will be made durable
2020-02-19 15:24:32 -08:00
fe75a4cafb
FastRestore:Apply clang-format
2020-02-19 15:22:52 -08:00
03f699f2f9
Merge branch 'master' into mengxu/fast-restore-applier-multi-applying-PR
2020-02-19 15:22:33 -08:00
9b3254d5f4
A corrupted processId file should be deleted in simulation, as that is the manual operation that would fix the problem in the real world
2020-02-19 15:21:42 -08:00
4326984b1d
fix: wait metrics can take a really long time to detect that two shards have been merged into one if both shards are assigned to the same team. Additional information should be added to the request to improve this.
2020-02-19 15:20:38 -08:00
fe78524bbc
Merge pull request #2678 from sears/networktest_perf
...
Add some tuning knobs to networktestclient; also, measure latency directly
2020-02-19 14:38:09 -08:00
551f1ba4d2
FastRestore:Minor revision for code review
2020-02-19 11:52:24 -08:00
956a3efa80
Pull request comments
2020-02-19 10:55:05 -08:00
89fcbb2055
Make clang happy
2020-02-19 09:44:15 -08:00
efc0d7f9d5
Added a specialized algorithm for PolicyOne and PoilcyAcross(,'zoneId',PolicyOne()) to find a set of TLog servers which will be able to fulfill the policy later.
2020-02-19 09:25:57 -08:00
88d36af9c7
Fix --tls_password and add better error logging
...
This refactors all tls settings into a TLSParams object so that we can
set the password before loading any certificates.
It turns out that the FDBLibTLS code did really nice things with error
logging, but I just didn't understand openssl enough before to realize
what pieces I should be copying.
2020-02-19 00:57:05 -08:00
085a6af9ff
FastRestore:Fix compilation errors
2020-02-18 16:44:20 -08:00
94d799552e
FastRestore:Apply clang-format against master
2020-02-18 16:41:59 -08:00
132f5aa9ba
FastRestore:Improve trace name and cosmetic change
2020-02-18 16:41:19 -08:00
31a6ec34b7
Merge branch 'master' into mengxu/fast-restore-agent-PR
2020-02-18 16:17:59 -08:00
a12a161fb3
Merge branch 'master' into mengxu/fast-restore-pipeline-PR
2020-02-18 14:49:52 -08:00
c603b20e7e
FastRestore:Resolve review comments
2020-02-18 14:08:27 -08:00
d5ce9ec924
Merge pull request #2690 from bnamasivayam/master
...
Fix an incorrect if else check.
2020-02-17 18:35:37 -08:00
1be6915a38
Fix an incorrect if else check.
2020-02-17 17:31:41 -08:00
78f20398f1
Merge pull request #2671 from Daniel-B-Smith/moar-branches
...
Remove more branches from KeyInfo comparisons
2020-02-17 17:13:09 -08:00
9144c3e8ca
Merge pull request #2087 from atn34/issue-1226
...
Allow member actors access to private variables
2020-02-17 14:39:31 -08:00
b81e2e0e1d
Merge pull request #2685 from jzhou77/backup-fix
...
Fix key_not_found error due to deleted BackupConfig
2020-02-15 21:45:01 -08:00
9e4668d656
Fix key_not_found error due to deleted BackupConfig
...
Since backup worker doesn't catch this error, a deleted BackupConfig can cause
the backup worker to get key_not_found error. Fix by adding a check if the
container can be found.
2020-02-14 19:37:58 -08:00
94e7f790d8
Merge pull request #2667 from atn34/atn34/remove-flatbuffers-knob
...
Remove USE_OBJECT_SERIALIZER knob
2020-02-14 15:44:38 -08:00
1d9140d874
Removed TLogVersion logging.
...
Added logging of SharedTLog ID for each TLog.
Switched ID logged for TLogRejoining event to the TLog instead of the SharedTLog.
Made some parameters to startRole passed by reference.
2020-02-14 12:33:43 -08:00
a41aa41816
Merge pull request #2670 from Daniel-B-Smith/skip-memcmp
...
Revert to memcmp comparison in SkipList
2020-02-14 09:03:08 -08:00
c34a69df32
FastRestore:Applier:Remove unused func
2020-02-13 23:06:56 -08:00
3e2c19630a
FastRestore:Applier:atomicOp can work on an empty key
2020-02-13 23:05:54 -08:00
b57583a504
FastRestore:Applier:Handle multiple gets in parallel
2020-02-13 23:05:31 -08:00
53f427c319
FastRestore:Applier:fix getAndComputeStagingKeys
2020-02-13 22:11:30 -08:00
c859f859bc
Remove certBytes.
2020-02-13 21:34:23 -08:00
f2d30a9954
comment out certBytes to fix cmake builds
2020-02-13 21:31:36 -08:00
68a6f59830
Set options _within_ the retry loop
2020-02-13 16:15:41 -08:00
0d668ea0c3
FastRestore:Applier:Add more trace for perf tracking
2020-02-13 15:50:10 -08:00
0b27786811
FastRestore:Applier:Minor change for clang-format
2020-02-13 13:17:32 -08:00
7724c644e5
Add some tuning knobs to networktestclient; also, measure latency directly.
2020-02-13 13:11:54 -08:00
b5e60585aa
FastRestore:Applier:Fix precompute mutation result
2020-02-13 12:57:47 -08:00
b1b44d4477
FastRestore:Applier:Handle CompareAndClear atomicOp
2020-02-13 11:11:29 -08:00
58dad5373b
FastRestore:Applier:Handle CompareAndClear atomicOp
2020-02-13 11:06:20 -08:00
d3c01763d9
FastRestore:Applier:Handle version stamped key values
2020-02-13 10:48:36 -08:00
b008df97eb
FastRestore:Applier:Multiple set-clear mutations at same version
2020-02-13 10:13:46 -08:00
Meng Xu