tomcat-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Filip Hanik" <m...@filip.net>
Subject RE: tomct-replication Exception: java.util.ConcurrentModificationException
Date Tue, 15 Apr 2003 19:35:16 GMT
> The entry happends fairly regularly.  Does this mean no mcast traffic
> has occurred from that ip address for 5 seconds?

and what is your message frequency set to? it should be set to 500, or
something way smaller than 5000 at least.
If all your servers are up, then this should never happen. Because when it
does happen, it occurs a lot of overhead since the state has to be
replicated again.

not sure when the session listener is executed. The session is replicated by
the thread that handled the request, but it doesn't do so until the request
is completed.

btw, are you running tomcat 4 or 5?

Filip

> -----Original Message-----
> From: Dennis Muhlestein [mailto:dennis@zserve.com]
> Sent: Tuesday, April 15, 2003 12:22 PM
> To: Filip Hanik
> Subject: Re: tomct-replication Exception:
> java.util.ConcurrentModificationException
>
>
> Ours is set to 5000 (5 seconds right?)
>
> The entry happends fairly regularly.  Does this mean no mcast traffic
> has occurred from that ip address for 5 seconds?
>
> What happends to the tcp socket when this occurs?  Does it establish a
> new one?  Do you think that this could have been the cause of some of
> the session problems?
>
> Here is an added tidbit:  We have a SessionListener that executes a
> delete statement on a db when an attribute is removed.  Sometimes, there
> is a db deadlock problem in that code.  The effect is that whichever
> thread was trying to call into the session listener is going to be
> blocked for about 5 seconds.  I'm wondering if there is any correlation
> to the thread that calls into the session listener and the one that
> standardmanager/ReplicationManager is using.
>
> When our app works rock solid, then they can release tomcat 5 right :-)
>
> Thanks - Dennis
>
>
> Filip Hanik wrote:
> > this log entry means that a nodewas added to the cluster, in this case
> > 10.1.1.13 listening on port 4001.
> >
> > if this happens during the middle of your run, I suggest that
> you increase
> > the drop time,
> > hence the member has been considered dead and is now added again
> >
> > Filip
> >
> >
> >>-----Original Message-----
> >>From: Dennis Muhlestein [mailto:dennis@zserve.com]
> >>Sent: Tuesday, April 15, 2003 11:30 AM
> >>To: tomcat-user@jakarta.apache.org
> >>Subject: Re: tomct-replication Exception:
> >>java.util.ConcurrentModificationException
> >>
> >>
> >>In addition to the original exception.  We also get this:
> >>
> >>==> logs/localhost_log.2003-04-15.txt <==
> >>2003-04-15 10:49:08 StandardManager[]: [InMemoryReplicationManager]
> >>Replication member
> >>added:org.apache.catalina.cluster.mcast.McastMember[tcp://10.1.1.1
> >>3:4001,10.1.1.13,4001]
> >>
> >>I think the replication unit is suffering when this excpetion occurs.
> >>We have had some users complaining of their session being timed out.
> >>
> >>-Thanks
> >>
> >>
> >>Andy Eastham wrote:
> >>
> >>>Dennis,
> >>>
> >>>This is a feature of the collections framework.  If you create
> >>
> >>an iterator
> >>
> >>>on a collection, then modify the underlying collection object,
> >>
> >>you get this
> >>
> >>>exception.
> >>>
> >>>I've had it when I've mistakenly used an iterator to loop over
> >>
> >>a collection
> >>
> >>>and tried to delete an object from the underling  collection.
> >>
> >>Sometimes you
> >>
> >>>can perform the operation on the iterator itself to avoid this.
> >>>
> >>>You should be able to redesign the code to avoid it, once you
> >>
> >>know why it's
> >>
> >>>occurring.
> >>>
> >>>Andy
> >>>
> >>>
> >>>
> >>>>-----Original Message-----
> >>>>From: Dennis Muhlestein [mailto:dennis@zserve.com]
> >>>>Sent: 15 April 2003 15:34
> >>>>To: tomcat-user@jakarta.apache.org
> >>>>Subject: tomct-replication Exception:
> >>>>java.util.ConcurrentModificationException
> >>>>
> >>>>
> >>>>We get this exception in our catalina.out file fairly often.  Any
> >>>>suggestions?
> >>>>
> >>>>java.util.ConcurrentModificationException
> >>>>        at java.util.HashMap$HashIterator.nextEntry(HashMap.java:762)
> >>>>        at java.util.HashMap$KeyIterator.next(HashMap.java:798)
> >>>>        at
> >>>>org.apache.catalina.cluster.mcast.McastMembership.expire(McastMemb
> >>>>ership.java:52)
> >>>>        at
> >>>>org.apache.catalina.cluster.mcast.McastServiceImpl.receive(McastSe
> >>>>rviceImpl.java:116)
> >>>>        at
> >>>>org.apache.catalina.cluster.mcast.McastServiceImpl$ReceiverThread.
> >>>>run(McastServiceImpl.java:140)
> >>>>
> >>>>
> >>>>Using Filip's tomcat-replication.jar/tomcat 4.1.18/blackdown-jdk-1.4.1
> >>>>
> >>>>-Thanks
> >>>>Dennis
> >>>>
> >>>>
> >>>>---------------------------------------------------------------------
> >>>>To unsubscribe, e-mail: tomcat-user-unsubscribe@jakarta.apache.org
> >>>>For additional commands, e-mail: tomcat-user-help@jakarta.apache.org
> >>>>
> >>>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>
> >>
> >>---------------------------------------------------------------------
> >>To unsubscribe, e-mail: tomcat-user-unsubscribe@jakarta.apache.org
> >>For additional commands, e-mail: tomcat-user-help@jakarta.apache.org
> >>
> >>
> >
> >
>
>


---------------------------------------------------------------------
To unsubscribe, e-mail: tomcat-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tomcat-user-help@jakarta.apache.org


Mime
View raw message