hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "stack (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HBASE-532) Odd interaction between HRegion.get, HRegion.deleteAll and compactions
Date Thu, 10 Apr 2008 17:34:08 GMT

    [ https://issues.apache.org/jira/browse/HBASE-532?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12587716#action_12587716

stack commented on HBASE-532:

I was going to reinstate history array and then have the memcache scanner open iterators on
all history array members.  Flusher would clear the tail of the array on occasion.  Scanners
would still have hold references to history array elements so it could keep going.

Bryan D suggested that we just not open iterators at all.  He suggested that we just refactor
Memcache Scanners so they use the get and getFull Memcache primitives.   We'd add a getNext
kind of thing that would do like MapFile.getClosest only it would only return the next, not
the asked-for row OR the next closest but the next closest only.

Reviewing, it looks like this approach would simplify things, use less memory but be slower
-- but because the slowness is all up in memory, it won't show in general scan numbers.  I'm
having a go at this latter suggestion.  Patch soon.

> Odd interaction between HRegion.get, HRegion.deleteAll and compactions
> ----------------------------------------------------------------------
>                 Key: HBASE-532
>                 URL: https://issues.apache.org/jira/browse/HBASE-532
>             Project: Hadoop HBase
>          Issue Type: Bug
>    Affects Versions: 0.2.0, 0.1.1
>            Reporter: Jim Kellerman
>            Assignee: stack
>             Fix For: 0.2.0, 0.1.2
> If you apply the patch for HBASE-483 to the 0.1 branch and comment out lines 309 and
315 of MetaUtils.java (which force compactions of the root and meta regions respectively),
TestMergeTool fails. Why forcing compactions makes the test succeed is a mystery to me.

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message