[HDDS-7701] Data loss by deleting containers having non-deleted keys

• Type: Bug
• Status: Resolved
• Priority: Blocker
• Resolution: Fixed
• Affects Version/s: 1.3.0
• Fix Version/s: 1.4.0
• Component/s: DN, OM, SCM

Environment

Version: Our forked version based ozone-1.3.0 branch. commit hash is https://github.com/apache/ozone/commit/9c61a8aa497ab96c014ad3bb7b1ee4f731ebfaf8

FSO-enabled bucket

• Target Version/s: 1.4.0
• Assignee: Sumit Agrawal
• Reporter: Kohei Sugihara
• Created: 23/Dec/22 03:14
• Updated: 29/Aug/23 00:13
• Resolved: 28/Aug/23 15:26

Description

A key is unavailable after time spent even if we don't delete the keys. We dug SCM logs and then found that SCM removed a container that consists the key. I think this is not intentional behavior.

I will attach our logs about ReplicationManager for two cases. In each case, the key is 6GB and uploaded via multi-part upload using aws-cli. We experienced the first data loss in Case #1 and overwrote the key to fix it. However, the key was made unavailable again a month later in Case #2.

Case #1

Key was reported a NO_REPLCIA_FOUND error. Container #41524 was deleted. We had not use ContainerBalancer at that time.

S3G

2022-11-16 18:29:40,727 [qtp1869039062-812583] WARN org.eclipse.jetty.server.HttpChannel: /path/to/file.zip
javax.servlet.ServletException: javax.servlet.ServletException: java.lang.IllegalArgumentException: NO_REPLICA_FOUND
        at org.eclipse.jetty.server.handler.HandlerCollection.handle(HandlerCollection.java:162)
        at org.eclipse.jetty.server.handler.HandlerWrapper.handle(HandlerWrapper.java:127)
        at org.eclipse.jetty.server.Server.handle(Server.java:516)
        at org.eclipse.jetty.server.HttpChannel.lambda$handle$1(HttpChannel.java:388)
        at org.eclipse.jetty.server.HttpChannel.dispatch(HttpChannel.java:633)
        at org.eclipse.jetty.server.HttpChannel.handle(HttpChannel.java:380)
        at org.eclipse.jetty.server.HttpConnection.onFillable(HttpConnection.java:277)
        at org.eclipse.jetty.io.AbstractConnection$ReadCallback.succeeded(AbstractConnection.java:311)
        at org.eclipse.jetty.io.FillInterest.fillable(FillInterest.java:105)
        at org.eclipse.jetty.io.ssl.SslConnection$DecryptedEndPoint.onFillable(SslConnection.java:555)
        at org.eclipse.jetty.io.ssl.SslConnection.onFillable(SslConnection.java:410)
        at org.eclipse.jetty.io.ssl.SslConnection$2.succeeded(SslConnection.java:164)
        at org.eclipse.jetty.io.FillInterest.fillable(FillInterest.java:105)
        at org.eclipse.jetty.io.ChannelEndPoint$1.run(ChannelEndPoint.java:104)
        at org.eclipse.jetty.util.thread.strategy.EatWhatYouKill.runTask(EatWhatYouKill.java:338)
        at org.eclipse.jetty.util.thread.strategy.EatWhatYouKill.doProduce(EatWhatYouKill.java:315)
        at org.eclipse.jetty.util.thread.strategy.EatWhatYouKill.tryProduce(EatWhatYouKill.java:173)
        at org.eclipse.jetty.util.thread.strategy.EatWhatYouKill.run(EatWhatYouKill.java:131)
        at org.eclipse.jetty.util.thread.ReservedThreadExecutor$ReservedThread.run(ReservedThreadExecutor.java:386)
        at org.eclipse.jetty.util.thread.QueuedThreadPool.runJob(QueuedThreadPool.java:883)
        at org.eclipse.jetty.util.thread.QueuedThreadPool$Runner.run(QueuedThreadPool.java:1034)
        at java.base/java.lang.Thread.run(Thread.java:829)
Caused by: javax.servlet.ServletException: java.lang.IllegalArgumentException: NO_REPLICA_FOUND
        at org.glassfish.jersey.servlet.WebComponent.serviceImpl(WebComponent.java:410)
        at org.glassfish.jersey.servlet.WebComponent.service(WebComponent.java:346)
        at org.glassfish.jersey.servlet.ServletContainer.service(ServletContainer.java:366)
        at org.glassfish.jersey.servlet.ServletContainer.service(ServletContainer.java:319)
        at org.glassfish.jersey.servlet.ServletContainer.service(ServletContainer.java:205)
        at org.eclipse.jetty.servlet.ServletHolder$NotAsync.service(ServletHolder.java:1459)
        at org.eclipse.jetty.servlet.ServletHolder.handle(ServletHolder.java:799)
        at org.eclipse.jetty.servlet.ServletHandler$ChainEnd.doFilter(ServletHandler.java:1626)
        at org.apache.hadoop.ozone.s3.RootPageDisplayFilter.doFilter(RootPageDisplayFilter.java:53)
        at org.eclipse.jetty.servlet.FilterHolder.doFilter(FilterHolder.java:193)
        at org.eclipse.jetty.servlet.ServletHandler$Chain.doFilter(ServletHandler.java:1601)
        at org.apache.hadoop.ozone.s3.EmptyContentTypeFilter.doFilter(EmptyContentTypeFilter.java:76)
        at org.eclipse.jetty.servlet.FilterHolder.doFilter(FilterHolder.java:201)
        at org.eclipse.jetty.servlet.ServletHandler$Chain.doFilter(ServletHandler.java:1601)
        at org.apache.hadoop.hdds.server.http.HttpServer2$QuotingInputFilter.doFilter(HttpServer2.java:1681)
        at org.eclipse.jetty.servlet.FilterHolder.doFilter(FilterHolder.java:193)
        at org.eclipse.jetty.servlet.ServletHandler$Chain.doFilter(ServletHandler.java:1601)
        at org.apache.hadoop.hdds.server.http.NoCacheFilter.doFilter(NoCacheFilter.java:48)
        at org.eclipse.jetty.servlet.FilterHolder.doFilter(FilterHolder.java:193)
        at org.eclipse.jetty.servlet.ServletHandler$Chain.doFilter(ServletHandler.java:1601)
        at org.eclipse.jetty.servlet.ServletHandler.doHandle(ServletHandler.java:548)
        at org.eclipse.jetty.server.handler.ScopedHandler.handle(ScopedHandler.java:143)
        at org.eclipse.jetty.security.SecurityHandler.handle(SecurityHandler.java:602)
        at org.eclipse.jetty.server.handler.HandlerWrapper.handle(HandlerWrapper.java:127)
        at org.eclipse.jetty.server.handler.ScopedHandler.nextHandle(ScopedHandler.java:235)
        at org.eclipse.jetty.server.session.SessionHandler.doHandle(SessionHandler.java:1624)
        at org.eclipse.jetty.server.handler.ScopedHandler.nextHandle(ScopedHandler.java:233)
        at org.eclipse.jetty.server.handler.ContextHandler.doHandle(ContextHandler.java:1434)
        at org.eclipse.jetty.server.handler.ScopedHandler.nextScope(ScopedHandler.java:188)
        at org.eclipse.jetty.servlet.ServletHandler.doScope(ServletHandler.java:501)
        at org.eclipse.jetty.server.session.SessionHandler.doScope(SessionHandler.java:1594)
        at org.eclipse.jetty.server.handler.ScopedHandler.nextScope(ScopedHandler.java:186)
        at org.eclipse.jetty.server.handler.ContextHandler.doScope(ContextHandler.java:1349)
        at org.eclipse.jetty.server.handler.ScopedHandler.handle(ScopedHandler.java:141)
        at org.eclipse.jetty.server.handler.HandlerCollection.handle(HandlerCollection.java:146)
        ... 21 more
Caused by: java.lang.IllegalArgumentException: NO_REPLICA_FOUND
        at com.google.common.base.Preconditions.checkArgument(Preconditions.java:145)
        at org.apache.hadoop.hdds.scm.XceiverClientManager.acquireClient(XceiverClientManager.java:164)
        at org.apache.hadoop.hdds.scm.XceiverClientManager.acquireClientForReadData(XceiverClientManager.java:157)
        at org.apache.hadoop.hdds.scm.storage.BlockInputStream.acquireClient(BlockInputStream.java:256)
        at org.apache.hadoop.hdds.scm.storage.BlockInputStream.getChunkInfos(BlockInputStream.java:223)
        at org.apache.hadoop.hdds.scm.storage.BlockInputStream.initialize(BlockInputStream.java:146)
        at org.apache.hadoop.hdds.scm.storage.BlockInputStream.readWithStrategy(BlockInputStream.java:283)
        at org.apache.hadoop.hdds.scm.storage.ExtendedInputStream.read(ExtendedInputStream.java:54)
        at org.apache.hadoop.hdds.scm.storage.ByteArrayReader.readFromBlock(ByteArrayReader.java:57)
        at org.apache.hadoop.ozone.client.io.KeyInputStream.readWithStrategy(KeyInputStream.java:273)
        at org.apache.hadoop.ozone.client.io.KeyInputStream.read(KeyInputStream.java:240)
        at org.apache.hadoop.ozone.client.io.OzoneInputStream.read(OzoneInputStream.java:56)
        at org.apache.commons.io.IOUtils.copyLarge(IOUtils.java:1384)
        at org.apache.hadoop.ozone.s3.endpoint.ObjectEndpoint.lambda$get$1(ObjectEndpoint.java:344)
        at org.glassfish.jersey.message.internal.StreamingOutputProvider.writeTo(StreamingOutputProvider.java:55)
        at org.glassfish.jersey.message.internal.StreamingOutputProvider.writeTo(StreamingOutputProvider.java:37)
        at org.glassfish.jersey.message.internal.WriterInterceptorExecutor$TerminalWriterInterceptor.invokeWriteTo(WriterInterceptorExecutor.java:242)
        at org.glassfish.jersey.message.internal.WriterInterceptorExecutor$TerminalWriterInterceptor.aroundWriteTo(WriterInterceptorExecutor.java:227)
        at org.glassfish.jersey.message.internal.WriterInterceptorExecutor.proceed(WriterInterceptorExecutor.java:139)
        at org.glassfish.jersey.server.internal.JsonWithPaddingInterceptor.aroundWriteTo(JsonWithPaddingInterceptor.java:85)
        at org.glassfish.jersey.message.internal.WriterInterceptorExecutor.proceed(WriterInterceptorExecutor.java:139)
        at org.glassfish.jersey.server.internal.MappableExceptionWrapperInterceptor.aroundWriteTo(MappableExceptionWrapperInterceptor.java:61)
        at org.glassfish.jersey.message.internal.WriterInterceptorExecutor.proceed(WriterInterceptorExecutor.java:139)
        at org.glassfish.jersey.message.internal.MessageBodyFactory.writeTo(MessageBodyFactory.java:1116)
        at org.glassfish.jersey.server.ServerRuntime$Responder.writeResponse(ServerRuntime.java:635)
        at org.glassfish.jersey.server.ServerRuntime$Responder.processResponse(ServerRuntime.java:373)
        at org.glassfish.jersey.server.ServerRuntime$Responder.process(ServerRuntime.java:363)
        at org.glassfish.jersey.server.ServerRuntime$1.run(ServerRuntime.java:258)
        at org.glassfish.jersey.internal.Errors$1.call(Errors.java:248)
        at org.glassfish.jersey.internal.Errors$1.call(Errors.java:244)
        at org.glassfish.jersey.internal.Errors.process(Errors.java:292)
        at org.glassfish.jersey.internal.Errors.process(Errors.java:274)
        at org.glassfish.jersey.internal.Errors.process(Errors.java:244)
        at org.glassfish.jersey.process.internal.RequestScope.runInScope(RequestScope.java:265)
        at org.glassfish.jersey.server.ServerRuntime.process(ServerRuntime.java:234)
        at org.glassfish.jersey.server.ApplicationHandler.handle(ApplicationHandler.java:680)
        at org.glassfish.jersey.servlet.WebComponent.serviceImpl(WebComponent.java:394)
        ... 55 more

ReplicationManager @ SCM

2022-11-15 16:30:16,680 [EventQueue-CloseContainerForCloseContainerEventHandler] INFO org.apache.hadoop.hdds.scm.container.CloseContainerEventHandler: Close container Event triggered for container : #41524
2022-11-15 16:30:37,073 [EventQueue-CloseContainerForCloseContainerEventHandler] INFO org.apache.hadoop.hdds.scm.container.CloseContainerEventHandler: Close container Event triggered for container : #41524
2022-11-15 16:30:37,279 [EventQueue-CloseContainerForCloseContainerEventHandler] INFO org.apache.hadoop.hdds.scm.container.CloseContainerEventHandler: Close container Event triggered for container : #41524
2022-11-15 16:30:41,806 [EventQueue-CloseContainerForCloseContainerEventHandler] INFO org.apache.hadoop.hdds.scm.container.CloseContainerEventHandler: Close container Event triggered for container : #41524
2022-11-15 16:30:41,815 [FixedThreadPoolWithAffinityExecutor-8-0] INFO org.apache.hadoop.hdds.scm.container.IncrementalContainerReportHandler: Moving container #41524 to CLOSED state, datanode DN35-UUID{<snip>} reported CLOSED replica.
2022-11-16 09:18:55,963 [ReplicationMonitor] INFO org.apache.hadoop.hdds.scm.container.replication.LegacyReplicationManager: Sending delete container command for container #41524 to datanode DN21-UUID{<snip>}
2022-11-16 09:18:55,963 [ReplicationMonitor] INFO org.apache.hadoop.hdds.scm.container.replication.LegacyReplicationManager: Sending delete container command for container #41524 to datanode DN35-UUID{<snip>}
2022-11-16 09:18:55,963 [ReplicationMonitor] INFO org.apache.hadoop.hdds.scm.container.replication.LegacyReplicationManager: Sending delete container command for container #41524 to datanode DN18-UUID{<snip>}

sh key info & admin container info

SCM knows containers that construct the key, but actually, ReplicationManager removed one of the containers.

% ozone sh key info /path/to/file.zip > key-info.log
% jq '.ozoneKeyLocations[].containerID' < key-info.log | sort | uniq > containers.list
% cat containers.list | xargs echo
41196 41198 41200 41201 41204 41205 41206 41207 41208 41209 41210 41211 41212 41213 41216 41217 41218 41219 41220 41222 41224 41225 41226 41228 41229 41230 41231 41232 41233 41234 41235 41236 41237 41238 41239 41241 41242 41243 41244 41245 41246 41247 41248 41249 41250 41251 41252 41253 41254 41255 41256 41257 41258 41259 41260 41261 41263 41264 41265 41266 41267 41268 41269 41270 41272 41273 41275 41276 41277 41278 41279 41280 41281 41282 41283 41284 41285 41287 41288 41289 41290 41291 41292 41295 41296 41297 41298 41299 41300 41301 41303 41304 41305 41306 41307 41308 41309 41311 41312 41313 41314 41315 41316 41317 41318 41321 41322 41323 41324 41325 41326 41327 41328 41329 41330 41331 41332 41333 41334 41335 41336 41337 41338 41339 41340 41341 41342 41343 41344 41345 41346 41347 41348 41349 41350 41351 41352 41353 41355 41356 41357 41359 41360 41361 41362 41363 41364 41365 41366 41367 41368 41369 41370 41371 41372 41373 41374 41375 41376 41377 41378 41379 41380 41381 41382 41383 41384 41385 41386 41387 41388 41389 41390 41391 41392 41393 41394 41395 41396 41397 41398 41399 41400 41401 41402 41403 41404 41405 41406 41407 41408 41409 41410 41411 41412 41413 41414 41415 41416 41417 41418 41419 41420 41421 41422 41423 41424 41425 41426 41427 41428 41429 41431 41432 41433 41434 41435 41436 41437 41438 41439 41440 41442 41443 41446 41448 41449 41450 41451 41452 41453 41454 41455 41456 41457 41458 41459 41460 41461 41463 41464 41465 41466 41467 41469 41470 41471 41473 41474 41475 41476 41478 41479 41480 41481 41483 41484 41486 41487 41488 41489 41490 41491 41492 41493 41494 41495 41496 41497 41498 41499 41500 41501 41504 41505 41506 41508 41509 41510 41511 41512 41513 41514 41515 41516 41518 41519 41521 41522 41523 41524 41525 41526 41527 41528 41529 41530 41531 41532 41533 41534 41535 41536 41537 41538 41539 41540 41541 41543 41545 41546 41547 41548 41549 41550 41551 41552 41553 41554 41555 41556 41557 41558 41559 41560 41561 41562 41564 41565 41567 41568 41569 41570 41571 41572 41573 41574 41577 41578 41579 41580 41581 41582 41583 41584 41585 41586 41587 41588 41589 41590 41591 41592 41593 41594 41595 41596 41597 41598 41599 41600 41601 41602 41604 41605 41606 41607 41608 41609 41610 41611
% ozone admin container info 41524
Container id: 41524
Pipeline id: 7a478e5f-d948-4520-9147-323299d4d10a
Container State: DELETED
Datanodes: []
Replicas: []

Case #2

After re-uploading the file to the same path and a month later, Container #55905 was deleted. ContainerBalancer is running but I don't make sure that this is relevant. Ozone CLI and S3G logs are the same like the above.

2022-12-21 17:21:45,797 [ContainerBalancerTask] INFO org.apache.hadoop.hdds.scm.container.balancer.ContainerBalancerTask: ContainerBalancer matched source datanode DN19-UUID with target datanode DN14-UUID for container move.
2022-12-21 17:21:45,797 [ContainerBalancerTask] INFO org.apache.hadoop.hdds.scm.container.balancer.ContainerBalancerTask: ContainerBalancer is trying to move container #55905 with size 247113087B from source datanode DN19-UUID to target datanode DN14-UUID
2022-12-21 17:21:45,797 [ContainerBalancerTask] INFO org.apache.hadoop.hdds.scm.container.replication.LegacyReplicationManager: Sending replicate container command for container #55905 to datanode DN14-UUID{<snip>}]
2022-12-21 17:21:45,797 [ContainerBalancerTask] INFO org.apache.hadoop.hdds.scm.container.replication.LegacyReplicationManager: receive a move request about container #55905 , from DN19-UUID to DN14-UUID
2022-12-21 17:53:31,409 [ReplicationMonitor] WARN org.apache.hadoop.hdds.scm.container.balancer.ContainerBalancerTask: Container move for container #55905 from source DN19-UUID to target DN14-UUID failed: REPLICATION_FAIL_TIME_OUT
...
2022-12-21 18:23:34,070 [ReplicationMonitor] INFO org.apache.hadoop.hdds.scm.container.replication.LegacyReplicationManager: Container #55905 is over replicated. Expected replica count is 3, but found 4.
2022-12-21 18:23:34,070 [ReplicationMonitor] INFO org.apache.hadoop.hdds.scm.container.replication.LegacyReplicationManager: Sending delete container command for container #55905 to datanode DN9-UUID{<snip>}
...
2022-12-21 18:38:35,024 [ReplicationMonitor] INFO org.apache.hadoop.hdds.scm.container.replication.LegacyReplicationManager: Sending delete container command for container #55905 to datanode DN8-UUID{<snip>}
2022-12-21 18:38:35,025 [ReplicationMonitor] INFO org.apache.hadoop.hdds.scm.container.replication.LegacyReplicationManager: Sending delete container command for container #55905 to datanode DN14-UUID{<snip>}
2022-12-21 18:38:35,025 [ReplicationMonitor] INFO org.apache.hadoop.hdds.scm.container.replication.LegacyReplicationManager: Sending delete container command for container #55905 to datanode DN19-UUID{<snip>}

Container #55905 is the part of the key.

% cat containers.list | xargs echo
55709 55710 55713 55715 55718 55722 55723 55724 55726 55727 55728 55730 55731 55732 55734 55738 55739 55740 55741 55742 55745 55747 55748 55749 55751 55752 55753 55754 55755 55756 55758 55759 55760 55761 55762 55764 55765 55766 55767 55768 55772 55773 55774 55775 55777 55778 55780 55781 55783 55784 55785 55789 55790 55791 55793 55794 55795 55796 55797 55798 55799 55800 55801 55802 55803 55804 55806 55807 55808 55809 55810 55811 55812 55813 55814 55815 55816 55817 55818 55819 55820 55821 55822 55824 55825 55827 55828 55829 55830 55831 55834 55835 55836 55837 55838 55839 55840 55841 55842 55843 55844 55845 55846 55847 55848 55849 55852 55854 55855 55856 55857 55858 55859 55860 55861 55862 55863 55864 55865 55866 55868 55869 55870 55871 55872 55874 55875 55876 55877 55878 55879 55880 55881 55882 55883 55884 55885 55887 55889 55890 55891 55892 55893 55894 55895 55896 55897 55898 55899 55900 55901 55902 55903 55904 55905 55906 55907 55909 55910 55912 55913 55914 55915 55916 55917 55918 55919 55921 55922 55923 55924 55925 55926 55927 55928 55929 55930 55931 55932 55933 55934 55935 55936 55937 55938 55939 55940 55941 55942 55943 55944 55945 55946 55947 55948 55949 55950 55951 55952 55953 55954 55955 55957 55958 55959 55960 55961 55962 55963 55964 55965 55966 55967 55968 55969 55970 55971 55972 55973 55974 55975 55976 55978 55980 55981 55982 55983 55984 55985 55986 55987 55988 55989 55990 55991 55992 55994 55995 55996 55997 55998 55999 56001 56002 56003 56004 56005 56007 56008 56009 56012 56013 56014 56015 56016 56017 56018 56019 56020 56021 56022 56023 56024 56025 56026 56027 56028 56029 56031 56032 56033 56034 56035 56036 56037 56038 56039 56040 56041 56042 56043 56044 56045 56046 56047 56049 56050 56051 56052 56053 56054 56055 56056 56057 56058 56059 56060 56061 56062 56063 56064 56065 56066 56067 56068 56069 56070 56071 56072 56073 56074 56075 56076 56078 56079 56080 56081 56083 56084 56086 56087 56088 56089 56091 56092 56093 56094 56096 56097 56098 56099 56100 56101 56102 56103 56104 56105 56106 56107 56108 56109 56110 56111 56113 56114 56115 56116 56118 56119 56120 56122 56123 56124 56125 56128 56129 56130 56131 56132 56133 56134 56135 56137

% ozone admin container info 55905

Container id: 55905
Pipeline id: abd81f12-9194-4a1c-9b85-a827b1fd060e
Container State: DELETED
Datanodes: []
Replicas: []

Issue Links

• is related to: HDDS-8129 ContainerStateMachine allows two different tasks with the same container id running in parallel
• relates to: HDDS-8118 Check if container directory is empty before deleting container