You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@river.apache.org by "Mark Brouwer (JIRA)" <ji...@apache.org> on 2007/07/26 14:46:40 UTC

[jira] Created: (RIVER-43) Deadlock in group VM creation failure with restartable objects

Deadlock in group VM creation failure with restartable objects
--------------------------------------------------------------

                 Key: RIVER-43
                 URL: https://issues.apache.org/jira/browse/RIVER-43
             Project: River
          Issue Type: Bug
          Components: com_sun_jini_phoenix
    Affects Versions: jtsk_2.1
            Reporter: Mark Brouwer
            Priority: Minor


Bugtraq ID [6512930|http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6512930]

A deadlock condition exists in {{Activation.GroupEntry.getInstantiator}}. If the group VM exits, and there is a restartable  customer, the watchdog will attempt to restart it, which results in another {{getInstantiator}} call. This can "recurse" again and again. Eventually one of those {{getInstantiator}} calls can block in {{Pstartgroup}} if the {{groupThrottle}} is reached, and it blocks with the {{GroupEntry}} locked. If all of that occurs before some/all of the earlier {{getInstantiator}} threads are scheduled and executed, those threads are stuck waiting to reacquire the {{GroupEntry}} lock coming out of the wait, so they won't execute a {{Vstartgroup}}, and a deadlock results.

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