accumulo-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Billie Rinaldi (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (ACCUMULO-3725) Majc trace tacked onto minc trace
Date Tue, 14 Apr 2015 18:14:12 GMT

    [ https://issues.apache.org/jira/browse/ACCUMULO-3725?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14494522#comment-14494522
] 

Billie Rinaldi commented on ACCUMULO-3725:
------------------------------------------

I'm going to commit a patch that stops the minc root span before the major compaction is initiated.
 This will give the major compaction a new parent instead of having the minc root span be
its parent.  The minc span still needs to be stopped in the finally block, in case an error
occurs before it is stopped in the try block.  Stopping more than once should have no effect.

> Majc trace tacked onto minc trace
> ---------------------------------
>
>                 Key: ACCUMULO-3725
>                 URL: https://issues.apache.org/jira/browse/ACCUMULO-3725
>             Project: Accumulo
>          Issue Type: Bug
>          Components: trace
>            Reporter: Billie Rinaldi
>            Assignee: Billie Rinaldi
>             Fix For: 1.7.0
>
>
> [~elserj] noticed an issue where a minorCompaction trace of length 10295 ms also contained
spans for a major compaction starting at time offset 46336 ms.  Looking into this, it seems
like MinorCompactionTask.run does the following:
> {noformat}
> start minc root span
> try {
>   minor compaction
>   maybe major compaction
> } finally {
>   stop minc root span
> }
> {noformat}
> The major compaction is async, so it gets initiated with the minor compaction span as
its parent, then the minor compaction span is stopped, and at some point later the major compaction
occurs.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message