[
https://issues.apache.org/jira/browse/HBASE-5179?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13188248#comment-13188248
]
chunhui shen commented on HBASE-5179:
-------------------------------------
@Zhihong
For the first question: this.serverManager.isServerOnline() returns true and
catalogTracker.verifyMetaRegionLocation(timeout) returns false, if the meta
server is really dead and we assign meta without splitting log, causing meta
data loss.
Through the following code:
{code}
if (!catalogTracker.verifyRootRegionLocation(timeout)) {
ServerName currentRootServer = this.catalogTracker.getRootLocation();
if (expireIfOnline(currentRootServer)) {
// We are expiring this server. The processing of expiration will assign
// root so don't do it here.
expiredServer = currentRootServer;
} else {
// Root was not on an online server when we failed verification
this.assignmentManager.assignRoot();
}
private boolean expireIfOnline(final ServerName sn) {
if (sn == null) return false;
if (!this.serverManager.isServerOnline(sn)) return false;
LOG.info("Forcing expiration of " + sn);
this.serverManager.expireServer(sn);
return true;
}
{code}
We could find currentRootServer is considered dead, with the same condition,
could we consider currentMetaServer dead?
> Concurrent processing of processFaileOver and ServerShutdownHandler may cause
> region to be assigned before log splitting is completed, causing data loss
> --------------------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-5179
> URL: https://issues.apache.org/jira/browse/HBASE-5179
> Project: HBase
> Issue Type: Bug
> Components: master
> Affects Versions: 0.90.2
> Reporter: chunhui shen
> Assignee: chunhui shen
> Priority: Critical
> Fix For: 0.92.0, 0.94.0, 0.90.6
>
> Attachments: 5179-90.txt, 5179-90v2.patch, 5179-90v3.patch,
> 5179-90v4.patch, 5179-90v5.patch, 5179-90v6.patch, 5179-90v7.patch,
> 5179-90v8.patch, 5179-90v9.patch, 5179-v2.txt, 5179-v3.txt, 5179-v4.txt,
> hbase-5179.patch, hbase-5179v5.patch, hbase-5179v6.patch, hbase-5179v7.patch,
> hbase-5179v8.patch, hbase-5179v9.patch
>
>
> If master's processing its failover and ServerShutdownHandler's processing
> happen concurrently, it may appear following case.
> 1.master completed splitLogAfterStartup()
> 2.RegionserverA restarts, and ServerShutdownHandler is processing.
> 3.master starts to rebuildUserRegions, and RegionserverA is considered as
> dead server.
> 4.master starts to assign regions of RegionserverA because it is a dead
> server by step3.
> However, when doing step4(assigning region), ServerShutdownHandler may be
> doing split log, Therefore, it may cause data loss.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
