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-4192) Analyze Threading for Tracing correctness
Date Sat, 16 Apr 2016 21:54:25 GMT

    [ https://issues.apache.org/jira/browse/ACCUMULO-4192?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15244421#comment-15244421
] 

Billie Rinaldi commented on ACCUMULO-4192:
------------------------------------------

I think the removal of that TraceRunnable was fixing a bug: ACCUMULO-3507. It was sort of
related to switching to htrace, in that switching to htrace revealed the bug.

> Analyze Threading for Tracing correctness
> -----------------------------------------
>
>                 Key: ACCUMULO-4192
>                 URL: https://issues.apache.org/jira/browse/ACCUMULO-4192
>             Project: Accumulo
>          Issue Type: Task
>          Components: trace
>            Reporter: Josh Elser
>            Priority: Blocker
>             Fix For: 1.7.2, 1.8.0
>
>
> When we switched from cloudtrace to htrace, we lost an implicit {{TraceRunnable}} wrapper
in NamingThreadFactory (which is used by SimpleThreadPool). This results in the current Trace's
span being lost across threads.
> We should inspect uses of SimpleThreadPool (and maybe TraceRunnable in 1.6) to make sure
that we don't have any other hidden tracing problems before the next 1.7 and 1.8 release.



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

Mime
View raw message