hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Lefty Leverenz (JIRA)" <>
Subject [jira] [Commented] (HIVE-7353) HiveServer2 using embedded MetaStore leaks JDOPersistanceManager
Date Wed, 12 Nov 2014 07:00:47 GMT


Lefty Leverenz commented on HIVE-7353:

bq.  ... we are removing TODOC label after doc but before you review ...

That's fine because my reviews generally turn up trivial edits that don't affect doc usability.
 (On the other hand when I write doc, it makes sense to keep the label pending expert review.)

Today's flood of documentation is very encouraging -- thanks all!  I'll get to the reviews
as soon as I can, and tackle the backlist too.

> HiveServer2 using embedded MetaStore leaks JDOPersistanceManager
> ----------------------------------------------------------------
>                 Key: HIVE-7353
>                 URL:
>             Project: Hive
>          Issue Type: Bug
>          Components: HiveServer2
>    Affects Versions: 0.13.0
>            Reporter: Vaibhav Gumashta
>            Assignee: Vaibhav Gumashta
>             Fix For: 0.14.0
>         Attachments: HIVE-7353.1.patch, HIVE-7353.2.patch, HIVE-7353.3.patch, HIVE-7353.4.patch,
HIVE-7353.5.patch, HIVE-7353.6.patch, HIVE-7353.7.patch, HIVE-7353.8.patch, HIVE-7353.9.patch
> While using embedded metastore, while creating background threads to run async operations,
HiveServer2 ends up creating new instances of JDOPersistanceManager which are cached in JDOPersistanceManagerFactory.
Even when the background thread is killed by the thread pool manager, the JDOPersistanceManager
are never GCed because they are cached by JDOPersistanceManagerFactory.

This message was sent by Atlassian JIRA

View raw message