You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@ignite.apache.org by "Andrew Mashenkov (JIRA)" <ji...@apache.org> on 2018/04/17 10:18:00 UTC

[jira] [Created] (IGNITE-8295) Possible deadlock on partition eviction.

Andrew Mashenkov created IGNITE-8295:
----------------------------------------

             Summary: Possible deadlock on partition eviction.
                 Key: IGNITE-8295
                 URL: https://issues.apache.org/jira/browse/IGNITE-8295
             Project: Ignite
          Issue Type: Bug
            Reporter: Andrew Mashenkov
         Attachments: deadlock.stack

GridCacheOffheapManager.recreateCacheDataStore() calls updatePartitionCounter() under partStoreLock which may try to acquire checkpointReadLock.

recreateCacheDataStore() method can be called with checkpointReadLock (on GridDhtPartitionsExchangeFuture.updatePartitionFullMap) 
or without checkpointReadLock (GridDhtPartitionEvictor thread calls evictPartitionAsync),

So, checkpoint can cause a deadlock if it happens in between.


Seems, we should acquire checkpointReadLock before partStoreLock. 
  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)