hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Lars Hofhansl (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-17937) Memstore size becomes negative in case of expensive postPut/Delete Coprocessor call
Date Tue, 18 Apr 2017 17:52:41 GMT

    [ https://issues.apache.org/jira/browse/HBASE-17937?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15973163#comment-15973163

Lars Hofhansl commented on HBASE-17937:

So we looked at the code here. I think this is only a temporary condition like this:
1. doMiniBatchMutation starts and writes things into the memstore
2. the background flusher flushes the memstore. Decrementing the memstore size.
3. doMiniBatchMutation finishes. Incrementing the memstore size.

So we only have an issue between #2 and #3. When doMiniBatchMutation finishes the size will
be correct.
I do _not_ think it's a big problem, just a spurious log warning.

If we did not want that, we should move the incrementing of the memstore size right after
the sync of the wal in doMiniBatchMutation (inside the MVCC transaction).

> Memstore size becomes negative in case of expensive postPut/Delete Coprocessor call
> -----------------------------------------------------------------------------------
>                 Key: HBASE-17937
>                 URL: https://issues.apache.org/jira/browse/HBASE-17937
>             Project: HBase
>          Issue Type: Bug
>    Affects Versions: 2.0.0, 1.3.1, 0.98.24
>            Reporter: Abhishek Singh Chouhan
>            Assignee: Abhishek Singh Chouhan
>         Attachments: HBASE-17937.master.001.patch
> We ran into a situation where the memstore size became negative due to expensive postPut/Delete
Coprocessor calls in doMiniBatchMutate. We update the memstore size in the finally block of
doMiniBatchMutate, however a queued flush can be triggered during the coprocessor calls(if
they are taking time eg. index updates) since we have released the locks and advanced mvcc
at this point. The flush will turn the memstore size negative since the value subtracted is
the actual value flushed from stores. The negative value impacts the future flushes amongst
others that depend on memstore size.

This message was sent by Atlassian JIRA

View raw message