Hello,

Are there any comments on the issue?  Is it a bug? Or may be the
expected behaviour and known incompatibility?

Thanks,

- -

  Sergey


18.11.2021 17:37, Sergey Korotkov пишет:
> Hello,
>
> We have a trouble migrating the persistent data from the Ignite 2.8.1 to
> 2.11.0. The native persistence is used and one the caches has an
> CacheInterceptor configured.
>
> If we run Ignite 2.11.0 against the persistent data created in 2.8.1
> nodes do not start (only the 1st node starts but all others fail) with
> the exception like:
>
> INFO: Received activate cluster request with BaselineTopology[id=0]
> initiator node ID: 43b7cd18-4366-44ef-8bfd-df4778bb4b23
> INFO: Started state transition: activate cluster
> INFO: Started exchange init [topVer=AffinityTopologyVersion [topVer=2,
> minorTopVer=1], crd=true, evt=DISCOVERY_CUSTOM_EVT,
> evtNode=43b7cd18-4366-44ef-8bfd-df4778bb4b23,
> customEvt=ChangeGlobalStateMessage
> [id=d6445623d71-548457da-85c2-4abd-b6ac-0a8c39dec582,
> reqId=02ac8e24-c461-4361-805d-75364a997a7b,
> initiatingNodeId=43b7cd18-4366-44ef-8bfd-df4778bb4b23, state=ACTIVE,
> baselineTopology=BaselineTopology [id=0, branchingHash=-2080344499,
> branchingType='Cluster activation', baselineNodes=[node-1, node-2]],
> forceChangeBaselineTopology=false, timestamp=1637228041860,
> forceDeactivation=true], allowMerge=false, exchangeFreeSwitch=false]
> SEVERE: Got exception while starting (will rollback startup routine).
> class org.apache.ignite.IgniteCheckedException: Cache Interceptor
> mismatch [cacheName=test-cache, localInterceptor=CacheInterceptor,
> remoteInterceptor=null, rmtNodeId=92330e36-3a07-4a6e-b851-9e7fb6422e17]
> Fix cache configuration or set system property -DIGNI
> TE_SKIP_CONFIGURATION_CONSISTENCY_CHECK=true.
>         at
> org.apache.ignite.internal.processors.cache.GridCacheUtils.throwIgniteCheckedException(GridCacheUtils.java:1633)
>         at
> org.apache.ignite.internal.processors.cache.GridCacheUtils.checkAttributeMismatch(GridCacheUtils.java:1002)
>         at
> org.apache.ignite.internal.processors.cache.ClusterCachesInfo.checkCache(ClusterCachesInfo.java:398)
>         at
> org.apache.ignite.internal.processors.cache.ClusterCachesInfo.onKernalStart(ClusterCachesInfo.java:341)
>         at
> org.apache.ignite.internal.processors.cache.GridCacheProcessor.onKernalStart(GridCacheProcessor.java:687)
>         at
> org.apache.ignite.internal.IgniteKernal.start(IgniteKernal.java:1414)
>         at
> org.apache.ignite.internal.IgnitionEx$IgniteNamedInstance.start0(IgnitionEx.java:2141)
>         at
> org.apache.ignite.internal.IgnitionEx$IgniteNamedInstance.start(IgnitionEx.java:1787)
>         at
> org.apache.ignite.internal.IgnitionEx.start0(IgnitionEx.java:1172)
>         at org.apache.ignite.internal.IgnitionEx.start(IgnitionEx.java:668)
>         at org.apache.ignite.internal.IgnitionEx.start(IgnitionEx.java:590)
>         at org.apache.ignite.Ignition.start(Ignition.java:328)
>         at ServerNode.startNode(ServerNode.java:57)
>         at ServerNode.main(ServerNode.java:20)
>         at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
>         at
> sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
>         at
> sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
>         at java.lang.reflect.Method.invoke(Method.java:498)
>         at org.codehaus.mojo.exec.ExecJavaMojo$1.run(ExecJavaMojo.java:293)
>         at java.lang.Thread.run(Thread.java:748)
>
>
> We have prepared the small test to reproduce the problem.  Test starts 2
> ignite nodes with native persistence enabled and creates cache with
> interceptor.  To see the problem test should be run twice:
>
> 1. Create the persistent cache using ignite 2.8.1:
>
>    mvn clean package exec:java -Dignite.version=2.8.1
>
> 2. Try to start ignite 2.11.0 nodes agains created cache data. One of
> the nodes (node-2) will fail to start:
>
>    mvn clean package exec:java -Dignite.version=2.11.0
>
>
> ***
>
> Looks like the issue was introduced in 2.9.1.  Migration to 2.9.0 works
> fine but all 2.9.1, 2.10.0, 2.11.0 fail.
>
> ****
>
> The  error message suggests to use the
> -DIGNITE_SKIP_CONFIGURATION_CONSISTENCY_CHECK=true.  It does  help.  But
> can it be safely used in production environment?  What other important
> checks would be skipped?
>
> Thanks,
>

Reply via email to