You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "Enis Soztutar (JIRA)" <ji...@apache.org> on 2014/06/28 00:35:28 UTC
[jira] [Commented] (HBASE-11430) lastFlushSeqId has been updated
wrongly during region open
[ https://issues.apache.org/jira/browse/HBASE-11430?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14046485#comment-14046485 ]
Enis Soztutar commented on HBASE-11430:
---------------------------------------
Some more context, we saw this in a big cluster in masters log:
{code}
2014-06-27 12:02:58,641 DEBUG [RpcServer.handler=58,port=16020] master.ServerManager: RegionServer hor10n08.gq1.ygridcore.net,16020,1403856435763 indicates a last flushed sequence id (numberOfStores=1, numberOfStorefiles=0, storefileUncompressedSizeMB=0, storefileSizeMB=0, memstoreSizeMB=0, storefileIndexSizeMB=0, readRequestsCount=0, writeRequestsCount=0, rootIndexSizeKB=0, totalStaticIndexSizeKB=0, totalStaticBloomSizeKB=0, totalCompactingKVs=0, currentCompactedKVs=0, compactionProgressPct=NaN, completeSequenceId=2) that is less than the previous last flushed sequence id (40000002) for region IntegrationTestTimeBoundedRequestsWithRegionReplicas,11111100,1403857312477_0002.20ce78e6a77cb1ddd5b6e360784971f4. Ignoring.
2014-06-27 12:02:58,644 DEBUG [RpcServer.handler=52,port=16020] master.ServerManager: RegionServer hor4n37.gq1.ygridcore.net,16020,1403856407099 indicates a last flushed sequence id (numberOfStores=1, numberOfStorefiles=0, storefileUncompressedSizeMB=0, storefileSizeMB=0, memstoreSizeMB=0, storefileIndexSizeMB=0, readRequestsCount=0, writeRequestsCount=0, rootIndexSizeKB=0, totalStaticIndexSizeKB=0, totalStaticBloomSizeKB=0, totalCompactingKVs=0, currentCompactedKVs=0, compactionProgressPct=NaN, completeSequenceId=2) that is less than the previous last flushed sequence id (40000002) for region IntegrationTestTimeBoundedRequestsWithRegionReplicas,049f49f0,1403857312477_0001.b21a19448b7c2f62a3ded3cfc2ddddb1. Ignoring.
2014-06-27 12:02:58,644 DEBUG [RpcServer.handler=52,port=16020] master.ServerManager: RegionServer hor4n37.gq1.ygridcore.net,16020,1403856407099 indicates a last flushed sequence id (numberOfStores=1, numberOfStorefiles=0, storefileUncompressedSizeMB=0, storefileSizeMB=0, memstoreSizeMB=0, storefileIndexSizeMB=0, readRequestsCount=0, writeRequestsCount=0, rootIndexSizeKB=0, totalStaticIndexSizeKB=0, totalStaticBloomSizeKB=0, totalCompactingKVs=0, currentCompactedKVs=0, compactionProgressPct=NaN, completeSequenceId=2) that is less than the previous last flushed sequence id (40000002) for region IntegrationTestTimeBoundedRequestsWithRegionReplicas,aeeeee40,1403857312477_0001.cb4295b2b6abe2db1e04f1a249cc5667. Ignoring.
{code}
+1 for patch.
> lastFlushSeqId has been updated wrongly during region open
> ----------------------------------------------------------
>
> Key: HBASE-11430
> URL: https://issues.apache.org/jira/browse/HBASE-11430
> Project: HBase
> Issue Type: Bug
> Components: MTTR
> Affects Versions: 0.99.0, 0.98.3
> Reporter: Jeffrey Zhong
> Assignee: Jeffrey Zhong
> Attachments: hbase-11430.patch
>
>
> In HRegion this.lastFlushSeqId is updated when opening a region which is wrong and will cause issue in distributedLogReplay recovery mode(not in log splitting mode) because its value is bumped up for log replay mode.
> This may cause data loss issue when we recover a region in distributedLogReplay mode while no flush for the region because no edits to be replayed. Later the region move to another region server and then future recovery will have issue because the large flush sequence Id value.
--
This message was sent by Atlassian JIRA
(v6.2#6252)