[jira] [Updated] (IGNITE-21656) Cache dump fails on a cache with a node filter.

["Vladimir Steshin (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/IGNITE-21656?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Vladimir Steshin updated IGNITE-21656:
--------------------------------------
    Fix Version/s: 2.17

> Cache dump fails on a cache with a node filter.
> -----------------------------------------------
>
>                 Key: IGNITE-21656
>                 URL: https://issues.apache.org/jira/browse/IGNITE-21656
>             Project: Ignite
>          Issue Type: Bug
>            Reporter: Vladimir Steshin
>            Assignee: Vladimir Steshin
>            Priority: Major
>              Labels: ise
>             Fix For: 2.17
>
>          Time Spent: 2h 20m
>  Remaining Estimate: 0h
>
> Ignite may fail to create dump of cache with a node filter.
> 1) If there are no partitions on some node, we assign the empty set as the snapshot task result:
> {code:java}
> if (parts.isEmpty() && !withMetaStorage)
>      task0 = new GridFinishedFuture<>(Collections.emptySet());
> else {
>      ...
> }
> {code}
> Cache dump doesn't contain the metastorage. Further, we may cast the result to
> {code:java}
> SnapshotFutureTaskResult res = (SnapshotFutureTaskResult)task0.result();
> {code}
> but it is a
> {code:java}
> Collections#EMPTY_SET
> {code}
> Stacktrace:
> {code:java}
> [ERROR][snapshot-runner-#413%dump.IgniteCacheDumpSelfTest0%][IgniteTestResources] Critical system error detected. Will be handled accordingly to configured handler [hnd=NoOpFailureHandler [super=AbstractFailureHandler [ignoredFailureTypes=UnmodifiableSet [SYSTEM_WORKER_BLOCKED, SYSTEM_CRITICAL_OPERATION_TIMEOUT]]], failureCtx=FailureContext [type=CRITICAL_ERROR, err=java.lang.ClassCastException: class java.util.Collections$EmptySet cannot be cast to class o.a.i.i.processors.cache.persistence.snapshot.SnapshotFutureTaskResult (java.util.Collections$EmptySet is in module java.base of loader 'bootstrap'; o.a.i.i.processors.cache.persistence.snapshot.SnapshotFutureTaskResult is in unnamed module of loader 'app')]]
> java.lang.ClassCastException: class java.util.Collections$EmptySet cannot be cast to class org.apache.ignite.internal.processors.cache.persistence.snapshot.SnapshotFutureTaskResult (java.util.Collections$EmptySet is in module java.base of loader 'bootstrap'; org.apache.ignite.internal.processors.cache.persistence.snapshot.SnapshotFutureTaskResult is in unnamed module of loader 'app')
> at org.apache.ignite.internal.processors.cache.persistence.snapshot.IgniteSnapshotManager.lambda$initLocalFullSnapshot$923db49$1(IgniteSnapshotManager.java:1221) ~[classes/:?]
> at org.apache.ignite.internal.util.future.GridFinishedFuture.lambda$chain$d8c8d23e$1(GridFinishedFuture.java:180) ~[classes/:?]
> at org.apache.ignite.internal.util.future.GridFinishedFuture.lambda$chain$0(GridFinishedFuture.java:163) ~[classes/:?]
> {code}
> 2) Even we fix the cast, another failure may arise. If cache with node filter was created by a client or not by the excluded node (or may be not from the coordinator), the dump content may be checked before anything is written.
> {code:java}
> Caused by: class org.apache.ignite.IgniteException: Ouch! Argument is invalid: binary metadata directory not exists
> at org.apache.ignite.internal.processors.cache.persistence.snapshot.SnapshotPartitionsVerifyHandler.checkDumpFiles(SnapshotPartitionsVerifyHandler.java:382)
> at org.apache.ignite.internal.processors.cache.persistence.snapshot.SnapshotPartitionsVerifyHandler.invoke(SnapshotPartitionsVerifyHandler.java:183)
> at org.apache.ignite.internal.processors.cache.persistence.snapshot.SnapshotPartitionsQuickVerifyHandler.invoke(SnapshotPartitionsQuickVerifyHandler.java:62)
> at org.apache.ignite.internal.processors.cache.persistence.snapshot.SnapshotPartitionsQuickVerifyHandler.invoke(SnapshotPartitionsQuickVerifyHandler.java:37)
> at org.apache.ignite.internal.processors.cache.persistence.snapshot.IgniteSnapshotManager$SnapshotHandlers.invoke(IgniteSnapshotManager.java:3396)
> at org.apache.ignite.internal.processors.cache.persistence.snapshot.IgniteSnapshotManager$SnapshotHandlers.lambda$invokeAll$8d56ceed$1(IgniteSnapshotManager.java:3328)
> at org.apache.ignite.internal.util.IgniteUtils.doInParallel(IgniteUtils.java:11715)
> at org.apache.ignite.internal.util.IgniteUtils.doInParallel(IgniteUtils.java:11617)
> at org.apache.ignite.internal.util.IgniteUtils.doInParallel(IgniteUtils.java:11597)
> at org.apache.ignite.internal.processors.cache.persistence.snapshot.IgniteSnapshotManager$SnapshotHandlers.invokeAll(IgniteSnapshotManager.java:3325)
> at org.apache.ignite.internal.processors.cache.persistence.snapshot.IgniteSnapshotManager.lambda$initLocalFullSnapshot$923db49$1(IgniteSnapshotManager.java:1254)
> at org.apache.ignite.internal.util.future.GridFinishedFuture.lambda$chain$d8c8d23e$1(GridFinishedFuture.java:180)
> at org.apache.ignite.internal.util.future.GridFinishedFuture.lambda$chain$0(GridFinishedFuture.java:163)
> ... 3 more
> Caused by: java.lang.IllegalArgumentException: Ouch! Argument is invalid: binary metadata directory not exists
> at org.apache.ignite.internal.util.GridArgumentCheck.ensure(GridArgumentCheck.java:110)
> at org.apache.ignite.internal.processors.cache.persistence.snapshot.dump.Dump.standaloneKernalContext(Dump.java:169)
> at org.apache.ignite.internal.processors.cache.persistence.snapshot.dump.Dump.(Dump.java:149)
> at org.apache.ignite.internal.processors.cache.persistence.snapshot.SnapshotPartitionsVerifyHandler.checkDumpFiles(SnapshotPartitionsVerifyHandler.java:370)
> {code}
> 3) We might fix this with creating normal dump task instead of
> {code:java}
> task0 = new GridFinishedFuture<>(new SnapshotFutureTaskResult(Collections.emptySet(), null))
> {code}
> It could write the nessesary dump content before checking it. 
> But one more fauilure is met:
> {code:java}
> Caused by: class org.apache.ignite.IgniteException: Wrong number of group directories: 0
> at org.apache.ignite.internal.processors.cache.persistence.snapshot.dump.Dump.dumpGroupDirectory(Dump.java:384)
> at org.apache.ignite.internal.processors.cache.persistence.snapshot.dump.Dump.configs(Dump.java:237)
> at org.apache.ignite.dump.DumpReader.lambda$run$2(DumpReader.java:107)
> at java.base/java.util.stream.ReferencePipeline$7$1.accept(ReferencePipeline.java:271)
> at java.base/java.util.HashMap$EntrySpliterator.tryAdvance(HashMap.java:1785)
> at java.base/java.util.stream.StreamSpliterators$WrappingSpliterator.lambda$initPartialTraversalState$0(StreamSpliterators.java:294)
> at java.base/java.util.stream.StreamSpliterators$AbstractWrappingSpliterator.fillBuffer(StreamSpliterators.java:206)
> at java.base/java.util.stream.StreamSpliterators$AbstractWrappingSpliterator.doAdvance(StreamSpliterators.java:161)
> at java.base/java.util.stream.StreamSpliterators$WrappingSpliterator.tryAdvance(StreamSpliterators.java:300)
> at java.base/java.util.Spliterators$1Adapter.hasNext(Spliterators.java:681)
> at java.base/java.util.Iterator.forEachRemaining(Iterator.java:132)
> at org.apache.ignite.internal.processors.cache.persistence.snapshot.dump.AbstractCacheDumpTest$1.onCacheConfigs(AbstractCacheDumpTest.java:370)
> at org.apache.ignite.dump.DumpReader.run(DumpReader.java:106)
> ... 12 more
> {code}
> There is no cache group directory at all on the filtered node.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)