[
https://issues.apache.org/jira/browse/IGNITE-7389?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16330302#comment-16330302
]
Anton Vinogradov edited comment on IGNITE-7389 at 1/18/18 9:42 AM:
-------------------------------------------------------------------
[~ilantukh]
I see no reason to check cache on each addData.
Let's check it once on DataStreamer creation.
was (Author: avinogradov):
[~ilantukh]
I see no reason to check cache each addData.
Let's check it once on DataStreamer creation.
> DataStreamer hangs if exception was thrown during addData which isn't
> IgniteException
> -------------------------------------------------------------------------------------
>
> Key: IGNITE-7389
> URL: https://issues.apache.org/jira/browse/IGNITE-7389
> Project: Ignite
> Issue Type: Bug
> Reporter: Eduard Shangareev
> Assignee: Ilya Lantukh
> Priority: Major
>
> I have written test which starts cache on one node and right after that
> starts dataStreamer on another node. Which hangs on close method because
> {{resFut}} will never be done.
> {code}
> java.lang.IllegalStateException: Getting affinity for topology version
> earlier than affinity is calculated [locNode=TcpDiscoveryNode
> [id=ad14d7f6-5895-4038-ba5e-cc487ab00000, addrs=[127.0.0.1],
> sockAddrs=[/127.0.0.1:47501], discPort=47501, order=2, intOrder=2,
> lastExchangeTime=1515672065430, loc=true, ver=2.4.0#19700101-sha1:00000000,
> isClient=false], grp=PART-G2, topVer=AffinityTopologyVersion [topVer=4,
> minorTopVer=3], head=AffinityTopologyVersion [topVer=4, minorTopVer=4],
> history=[AffinityTopologyVersion [topVer=4, minorTopVer=4]]]
> at
> org.apache.ignite.internal.processors.affinity.GridAffinityAssignmentCache.cachedAffinity(GridAffinityAssignmentCache.java:603)
> at
> org.apache.ignite.internal.processors.cache.GridCacheAffinityManager.assignment(GridCacheAffinityManager.java:243)
> at
> org.apache.ignite.internal.processors.affinity.GridAffinityProcessor.affinityCache(GridAffinityProcessor.java:375)
> at
> org.apache.ignite.internal.processors.affinity.GridAffinityProcessor.partition0(GridAffinityProcessor.java:187)
> at
> org.apache.ignite.internal.processors.cacheobject.IgniteCacheObjectProcessorImpl.partition(IgniteCacheObjectProcessorImpl.java:267)
> at
> org.apache.ignite.internal.processors.cacheobject.IgniteCacheObjectProcessorImpl.toCacheKeyObject0(IgniteCacheObjectProcessorImpl.java:135)
> at
> org.apache.ignite.internal.processors.cache.binary.CacheObjectBinaryProcessorImpl.toCacheKeyObject(CacheObjectBinaryProcessorImpl.java:805)
> at
> org.apache.ignite.internal.processors.datastreamer.DataStreamerImpl.addData(DataStreamerImpl.java:581)
> at
> org.apache.ignite.internal.processors.datastreamer.DataStreamerImpl.addData(DataStreamerImpl.java:555)
> at
> org.gridgain.grid.internal.processors.cache.database.AbstractSnapshotTest$2.run(AbstractSnapshotTest.java:404)
> at
> org.apache.ignite.testframework.GridTestUtils$6.run(GridTestUtils.java:933)
> at
> org.apache.ignite.testframework.GridTestUtils$9.call(GridTestUtils.java:1278)
> at
> org.apache.ignite.testframework.GridTestThread.run(GridTestThread.java:86)
> {code}
> The best solution is waiting for topology on which cache should be started.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
