[ https://issues.apache.org/jira/browse/GEODE-223?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14703910#comment-14703910 ]
ASF subversion and git services commented on GEODE-223: ------------------------------------------------------- Commit 83029e0e6c9e367ab9297f38c0d54a4f8d8ebbe6 in incubator-geode's branch refs/heads/develop from Vito Gavrilov [ https://git-wip-us.apache.org/repos/asf?p=incubator-geode.git;h=83029e0 ] GEODE-223: Handle region create/destroy remote event in Redis adpater Ignore events where region creation initiated remotely attempts to create a local region reference when the region has already been destroyed. Also, the destruction of a region may be caught the query engine, so I have accounted for that by handling com.gemstone.gemfire.cache.query.RegionNotFoundException. Finally, a the Jedis client timeout has been increased for RedisDistDunitTest to account for concurrent region creation/destruction and an expected exception has been added to not fail over the log scanning. Sometimes when a region is destroyed the PooledMessage Processor will log a regiondestroyed exception, which is ok, but makes the test fail. closes #16 > RedisDistDUnitTest.testConcCreateDestroy NullPointerException > ------------------------------------------------------------- > > Key: GEODE-223 > URL: https://issues.apache.org/jira/browse/GEODE-223 > Project: Geode > Issue Type: Bug > Components: extensions > Reporter: Darrel Schneider > Assignee: Vitaliy Gavrilov > > com.gemstone.gemfire.redis.RedisDistDUnitTest.testConcCreateDestroy failed > because of this suspect string: > Found suspect string in log4j at line 1132 > [error 2015/08/11 11:39:31.020 PDT <P2P message reader for > cc3-rh6(17559)<v0>:13796 shared ordered uid=236 port=51035> tid=0x1506] > Exception occurred in CacheListener > java.lang.NullPointerException > at > java.util.concurrent.ConcurrentHashMap.put(ConcurrentHashMap.java:1124) > at > com.gemstone.gemfire.internal.redis.RegionProvider.createRemoteRegionLocally(RegionProvider.java:222) > at > com.gemstone.gemfire.redis.GemFireRedisServer.afterKeyCreate(GemFireRedisServer.java:532) > at > com.gemstone.gemfire.redis.GemFireRedisServer.access$400(GemFireRedisServer.java:128) > at > com.gemstone.gemfire.redis.GemFireRedisServer$MetaCacheListener.afterCreate(GemFireRedisServer.java:559) > at > com.gemstone.gemfire.internal.cache.EnumListenerEvent$AFTER_CREATE.dispatchEvent(EnumListenerEvent.java:97) > at > com.gemstone.gemfire.internal.cache.LocalRegion.dispatchEvent(LocalRegion.java:9251) > at > com.gemstone.gemfire.internal.cache.LocalRegion.dispatchListenerEvent(LocalRegion.java:7700) > at > com.gemstone.gemfire.internal.cache.LocalRegion.invokePutCallbacks(LocalRegion.java:6466) > at > com.gemstone.gemfire.internal.cache.EntryEventImpl.invokeCallbacks(EntryEventImpl.java:2623) > at > com.gemstone.gemfire.internal.cache.AbstractRegionEntry.dispatchListenerEvents(AbstractRegionEntry.java:157) > at > com.gemstone.gemfire.internal.cache.LocalRegion.basicPutPart2(LocalRegion.java:6316) > at > com.gemstone.gemfire.internal.cache.AbstractRegionMap.basicPut(AbstractRegionMap.java:3214) > at > com.gemstone.gemfire.internal.cache.LocalRegion.virtualPut(LocalRegion.java:6125) > at > com.gemstone.gemfire.internal.cache.DistributedRegion.virtualPut(DistributedRegion.java:394) > at > com.gemstone.gemfire.internal.cache.LocalRegionDataView.putEntry(LocalRegionDataView.java:118) > at > com.gemstone.gemfire.internal.cache.LocalRegion.basicUpdate(LocalRegion.java:6100) > at > com.gemstone.gemfire.internal.cache.AbstractUpdateOperation.doPutOrCreate(AbstractUpdateOperation.java:142) > at > com.gemstone.gemfire.internal.cache.AbstractUpdateOperation$AbstractUpdateMessage.basicOperateOnRegion(AbstractUpdateOperation.java:280) > I think the bug is in this code from RegionProvider: > {code} > r = cache.getRegion(key.toString()); > if (type == RedisDataType.REDIS_LIST) > doInitializeList(key, r); > else if (type == RedisDataType.REDIS_SORTEDSET) > doInitializeSortedSet(key, r); > this.regions.put(key, r); > {code} > The NPE happens on the last line (the put call) because r is null. > I think this code just needs to handle the race in which cache.getRegion > returns null. -- This message was sent by Atlassian JIRA (v6.3.4#6332)