accumulo-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Dave Marion (JIRA)" <>
Subject [jira] [Commented] (ACCUMULO-1321) Dynamic Classloader lost jars
Date Mon, 29 Apr 2013 23:18:19 GMT


Dave Marion commented on ACCUMULO-1321:

bq. When loading out of HDFS, the files never get cleaned up, even when shutting down. So
this is something that we're going to have to deal with. The only other potential issue is
that the hdfs loader drops the jars in with an arbitrary identifier included in the name,
so if things aren't cleaned up I wonder if we could potentially hit a collision causing an
error similar to what we're seeing now.

  Looking at [1], the cache should be cleaned up on close and the files are created with a
one-up number that is initialized with a Random.

> Dynamic Classloader lost jars
> -----------------------------
>                 Key: ACCUMULO-1321
>                 URL:
>             Project: Accumulo
>          Issue Type: Sub-task
>            Reporter: John Vines
>            Assignee: Keith Turner
>            Priority: Blocker
>             Fix For: 1.5.0
>         Attachments: ACCUMULO-1321.patch
> We have a table setup that uses some custom iterators. We ran an MR job against it without
issues. We then ran the job immediately after the first one wrapped and 2 of my tservers errored
with ClassNotFoundException, even though it ran just fine before.
> Unfortunately we don't have a stack trace (to see if it was breaking differently in the
VFSClassLoader), nor a convenient way to recreate currently. We're working on reproducing
it in order to get more information.

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see:

View raw message