[
https://issues.apache.org/jira/browse/HBASE-12131?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14261991#comment-14261991
]
Esteban Gutierrez commented on HBASE-12131:
-------------------------------------------
I revisited the issue and I think we just should let the loop in
HBaseFsckRepair.closeRegionSilentlyAndWait() not return if we get an IOE and
just let things go until we hit hbase.hbck.close.timeout.
{code}
while (System.currentTimeMillis() < expiration) {
try {
HRegionInfo rsRegion = rs.getRegionInfo(region.getRegionName());
if (rsRegion == null)
return;
} catch (IOException ioe) {
return; <---
}
{code}
Ideally we should have an option to let hbck kill a RS if the RSs is
unresponsive (something like HBASE-6290)
> [hbck] undeployRegions should handle gracefully network partitions and other
> exceptions to avoid the same region deployed multiple times
> ----------------------------------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-12131
> URL: https://issues.apache.org/jira/browse/HBASE-12131
> Project: HBase
> Issue Type: Bug
> Components: hbck
> Affects Versions: 0.94.23
> Reporter: Esteban Gutierrez
> Assignee: Esteban Gutierrez
> Priority: Critical
>
> If we get an IOE (we currently ignore it) while regions are being undeployed
> by hbck we should make sure that we don't re-assign that region in the master
> before we know that RS was marked as dead and optionally let the user to
> confirm that action or we will end in a split brain situation with clients
> talking to different RSs serving the same region.
> The offending part is here in HBaseFsck.undeployRegions():
> {code}
> private void undeployRegions(HbckInfo hi) throws IOException,
> InterruptedException {
> for (OnlineEntry rse : hi.deployedEntries) {
> LOG.debug("Undeploy region " + rse.hri + " from " + rse.hsa);
> try {
> HBaseFsckRepair.closeRegionSilentlyAndWait(admin, rse.hsa, rse.hri);
> offline(rse.hri.getRegionName());
> } catch (IOException ioe) {
> LOG.warn("Got exception when attempting to offline region "
> + Bytes.toString(rse.hri.getRegionName()), ioe);
> }
> }
> }
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
