[
https://issues.apache.org/jira/browse/HBASE-20411?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16462969#comment-16462969
]
huaxiang sun commented on HBASE-20411:
--------------------------------------
Seems to me that blockSignal in MemStoreFlusher seems is still needed, but the
sync block can just include
{code:java}
wakeupFlushThread();
try {
// we should be able to wait forever, but we've seen a bug where
// we miss a notify, so put a 5 second bound on it at least.
blockSignal.wait(5 * 1000);
} catch (InterruptedException ie) {
LOG.warn("Interrupted while waiting");
interrupted = true;
}
{code}
This is to avoid the case that reclaimMemStoreMemory() does not wake up flush
thread and it misses the blockSignal.notifyAll(). In this case, it needs to
wait for 5 seconds to wake up, which is bad. I will read the code again to
confirm and will put a patch, thanks.
> Ameliorate MutableSegment synchronize
> -------------------------------------
>
> Key: HBASE-20411
> URL: https://issues.apache.org/jira/browse/HBASE-20411
> Project: HBase
> Issue Type: Bug
> Reporter: stack
> Priority: Major
> Attachments: 2.load.patched.17704.lock.svg,
> 2.load.patched.2.17704.lock.svg, 2.more.patch.12010.lock.svg, 41901.lock.svg,
> HBASE-20411-atomiclong-longadder.patch, HBASE-20411.branch-2.0.001.patch,
> HBASE-20411.branch-2.0.002.patch, HBASE-20411.branch-2.0.003.patch,
> HBASE-20411.branch-2.0.004.patch, HBASE-20411.branch-2.0.005.patch,
> HBASE-20411.branch-2.0.006.patch, HBASE-20411.branch-2.0.007.patch
>
>
> This item is migrated from HBASE-20236 so it gets dedicated issue.
> Let me upload evidence that has this synchronize as a stake in our write-time
> perf. I'll migrate the patch I posted with updates that come of comments
> posted by [~mdrob] on the HBASE-20236 issue.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
