cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Brandon Williams (JIRA)" <>
Subject [jira] Commented: (CASSANDRA-2073) Streaming occasionally makes gossip back up
Date Fri, 28 Jan 2011 22:10:43 GMT


Brandon Williams commented on CASSANDRA-2073:

It looks like in my case, both nodes are stuck in the same spot:


"GossipStage:1" prio=10 tid=0x0000000041644000 nid=0xece waiting on condition [0x00007fdbe861f000]
   java.lang.Thread.State: WAITING (parking)
    at sun.misc.Unsafe.park(Native Method)
    - parking to wait for  <0x00007fdc5ecf8828> (a java.util.concurrent.FutureTask$Sync)
    at java.util.concurrent.locks.LockSupport.park(
    at java.util.concurrent.locks.AbstractQueuedSynchronizer.parkAndCheckInterrupt(
    at java.util.concurrent.locks.AbstractQueuedSynchronizer.doAcquireSharedInterruptibly(
    at java.util.concurrent.locks.AbstractQueuedSynchronizer.acquireSharedInterruptibly(
    at java.util.concurrent.FutureTask$Sync.innerGet(
    at java.util.concurrent.FutureTask.get(
    at org.apache.cassandra.db.HintedHandOffManager.deleteHintsForEndPoint(
    at org.apache.cassandra.service.StorageService.excise(
    at org.apache.cassandra.service.StorageService.handleStateLeft(
    at org.apache.cassandra.service.StorageService.onChange(
    at org.apache.cassandra.gms.Gossiper.doNotifications(
    at org.apache.cassandra.gms.Gossiper.applyApplicationStateLocally(
    at org.apache.cassandra.gms.Gossiper.applyStateLocally(
    at org.apache.cassandra.gms.GossipDigestAckVerbHandler.doVerb(
    at java.util.concurrent.ThreadPoolExecutor$Worker.runTask(
    at java.util.concurrent.ThreadPoolExecutor$

They are waiting on deleting hints, and then likely for the hints CF to compact.

> Streaming occasionally makes gossip back up
> -------------------------------------------
>                 Key: CASSANDRA-2073
>                 URL:
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 0.7.0
>            Reporter: Brandon Williams
>            Priority: Minor
> Streaming occasionally makes gossip back up, causing nodes to mark each other as down
even though the network is ok.  This appears to happen just after streaming has finished.
 I noticed this in the course of working on CASSANDRA-2072, so decommission is one way to
reproduce.  It seems to happen maybe one of fifteen or twenty tries, so it's fairly rare.

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

View raw message