[
https://issues.apache.org/jira/browse/CASSANDRA-15970?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
David Capwell updated CASSANDRA-15970:
--------------------------------------
Reviewers: Brandon Williams (was: Brandon Williams, David Capwell)
> 3.x fails to start if commit log has range tombstones from a column which is
> also deleted
> -----------------------------------------------------------------------------------------
>
> Key: CASSANDRA-15970
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15970
> Project: Cassandra
> Issue Type: Bug
> Components: Legacy/Local Write-Read Paths, Local/Commit Log
> Reporter: David Capwell
> Assignee: David Capwell
> Priority: Normal
> Fix For: 3.0.x, 3.11.x
>
>
> Cassandra crashes with the following exception
> {code}
> ERROR [node1_isolatedExecutor:1] node1 2020-07-21 18:59:39,048
> JVMStabilityInspector.java:102 - Exiting due to error while processing commit
> log during initialization.
> org.apache.cassandra.db.commitlog.CommitLogReplayer$CommitLogReplayException:
> Unexpected error deserializing mutation; saved to
> /var/folders/cm/08cddl2s25j7fq3jdb76gh4r0000gn/T/mutation6239873170066752296dat.
> This may be caused by replaying a mutation against a table with the same
> name but incompatible schema.
> at
> org.apache.cassandra.db.commitlog.CommitLogReplayer.handleReplayError(CommitLogReplayer.java:731)
> [dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.commitlog.CommitLogReplayer.replayMutation(CommitLogReplayer.java:656)
> [dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.commitlog.CommitLogReplayer.replaySyncSection(CommitLogReplayer.java:609)
> [dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.commitlog.CommitLogReplayer.recover(CommitLogReplayer.java:493)
> [dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.commitlog.CommitLogReplayer.recover(CommitLogReplayer.java:189)
> [dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.commitlog.CommitLog.recover(CommitLog.java:170)
> [dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.commitlog.CommitLog.recover(CommitLog.java:151)
> [dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.distributed.impl.Instance.lambda$startup$7(Instance.java:535)
> [dtest-3.0.21.jar:na]
> at
> java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> ~[na:1.8.0_242]
> at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> ~[na:1.8.0_242]
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> ~[na:1.8.0_242]
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> ~[na:1.8.0_242]
> at
> org.apache.cassandra.concurrent.NamedThreadFactory.lambda$threadLocalDeallocator$0(NamedThreadFactory.java:83)
> ~[dtest-3.0.21.jar:na]
> at java.lang.Thread.run(Thread.java:748) ~[na:1.8.0_242]
> Caused by: java.lang.NullPointerException: null
> at
> org.apache.cassandra.db.ClusteringComparator.validate(ClusteringComparator.java:206)
> ~[dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.partitions.PartitionUpdate.validate(PartitionUpdate.java:494)
> ~[dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.commitlog.CommitLogReplayer.replayMutation(CommitLogReplayer.java:629)
> [dtest-3.0.21.jar:na]
> ... 12 common frames omitted
> {code}
> If you drain in 2.2 before upgrade, you get the following
> {code}
> ERROR [SharedPool-Worker-1] node1 2020-07-21 22:17:25,661
> AbstractLocalAwareExecutorService.java:169 - Uncaught exception on thread
> Thread[SharedPool-Worker-1,5,node1]
> java.lang.RuntimeException: java.lang.NullPointerException
> at
> org.apache.cassandra.service.StorageProxy$DroppableRunnable.run(StorageProxy.java:2537)
> ~[dtest-3.0.21.jar:na]
> at
> java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> ~[na:1.8.0_242]
> at
> org.apache.cassandra.concurrent.AbstractLocalAwareExecutorService$FutureTask.run(AbstractLocalAwareExecutorService.java:165)
> ~[dtest-3.0.21.jar:na]
> at org.apache.cassandra.concurrent.SEPWorker.run(SEPWorker.java:109)
> [dtest-3.0.21.jar:na]
> at java.lang.Thread.run(Thread.java:748) [na:1.8.0_242]
> Caused by: java.lang.NullPointerException: null
> at
> org.apache.cassandra.db.ClusteringComparator.compare(ClusteringComparator.java:131)
> ~[dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.UnfilteredDeserializer$OldFormatDeserializer.compareNextTo(UnfilteredDeserializer.java:391)
> ~[dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.columniterator.SSTableIterator$ForwardReader.handlePreSliceData(SSTableIterator.java:105)
> ~[dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.columniterator.SSTableIterator$ForwardReader.hasNextInternal(SSTableIterator.java:164)
> ~[dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.columniterator.AbstractSSTableIterator$Reader.hasNext(AbstractSSTableIterator.java:336)
> ~[dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.filter.ClusteringIndexNamesFilter$1.hasNext(ClusteringIndexNamesFilter.java:157)
> ~[dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.rows.UnfilteredRowIterator.isEmpty(UnfilteredRowIterator.java:70)
> ~[dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.SinglePartitionReadCommand.queryMemtableAndSSTablesInTimestampOrder(SinglePartitionReadCommand.java:952)
> ~[dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.SinglePartitionReadCommand.queryMemtableAndDiskInternal(SinglePartitionReadCommand.java:679)
> ~[dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.SinglePartitionReadCommand.queryMemtableAndDisk(SinglePartitionReadCommand.java:656)
> ~[dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.SinglePartitionReadCommand.queryStorage(SinglePartitionReadCommand.java:491)
> ~[dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.db.ReadCommand.executeLocally(ReadCommand.java:418)
> ~[dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.service.StorageProxy$LocalReadRunnable.runMayThrow(StorageProxy.java:1793)
> ~[dtest-3.0.21.jar:na]
> at
> org.apache.cassandra.service.StorageProxy$DroppableRunnable.run(StorageProxy.java:2533)
> ~[dtest-3.0.21.jar:na]
> ... 4 common frames omitted
> {code}
> The issue found was as follows
> * The schema had a collection type in 2.1
> * a collection range tombstone happened in 2.1
> * the row only had the RT, no other cells
> * the collection was dropped in 2.1
> * 3.0 detected the collection was deleted and ignored the cell
> * 3.0 produced an empty row with a null clustering key (since we skipped the
> RT)
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
