ode-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Alex Boisvert (JIRA)" <j...@apache.org>
Subject [jira] Commented: (ODE-686) ODE on Derby cannot recover from out of space error
Date Tue, 27 Oct 2009 15:43:59 GMT

    [ https://issues.apache.org/jira/browse/ODE-686?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12770542#action_12770542

Alex Boisvert commented on ODE-686:

I was under the impression we had decided to not support <bpel:wait> but apparently
we do it in a somewhat broken way.  If we support it, the schedule created should be in-memory
instead of persistent.  We should open a separate issue for this.   The two dangers of long-running
in-memory processes are 1) memory consumption since we don't do hydration/dehydration of in-memory
instances and 2) increased exposure to a node failure.  Nothing inherently wrong here except
that those are important design considerations.

> ODE on Derby cannot recover from out of space error
> ---------------------------------------------------
>                 Key: ODE-686
>                 URL: https://issues.apache.org/jira/browse/ODE-686
>             Project: ODE
>          Issue Type: Bug
>          Components: BPEL Runtime, JBI Integration
>    Affects Versions: 1.2
>         Environment: Linux, Fuse ODE 1.2 JBI
>            Reporter: Mateusz Nowakowski
>            Assignee: Rafal Rusin
>             Fix For: 1.3.4
>         Attachments: stacktraces.txt
> When the free space on disk is running out, ODE fails on SQLs, but when the free space
on disk increases, ODE cannot recover.
> (attached stack traces)
> PS. All the processes are "in memory", why is database used for saving "persistent jobs"?

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

View raw message