tomcat-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Peter Rossbach>
Subject Re: session replication errors
Date Thu, 17 Nov 2005 09:54:55 GMT
Hey Olve,

I found a the NPE shutdown problem, but I must setup my Linux System
to look inside the strange timeout exception. I want remove the 
waitForAck flag
at the next release, but my time is currently a little bit limited. 
Please look and test
my next cluster checkin .-)

Thanks for reporting the bug.

Olve Hansen schrieb:

>Hey, thanks for your fixes, sadly it didn't fix my problem. That is if I
>am doing things right, which there are no guarantees for. I will
>continue to hunt down the problem(s). 
>If I configure my cluster as you suggested in your first mail, the
>Exceptions disappears, if I use my original config it is as it were. I
>would believe that a valid configuration shouldn't make tomcat throw
>severe exceptions... I also started experiencing shutdown exceptions
>from the svn version. 5.5.12 didn't have these. (stacktrace can be found
>in bottom of mail).
>Are there any gotchas for using waitForAck="true" and
>keepAliveTimeout="-1"? For example less performance? I guess that
>waiting for acks from the whole cluster can be lengthy, especially if we
>want to use pooled replication?
>You also said in an earlier mail that I didn't need both the
>ClusterSessionListener, and the JvmRouteSessionIDBinderListener, why is
>that? And which do I need? And why are both listed in the example set-up
>in clusterdocs, if only one is needed?
>Now for the problem (sorry for the length of this mail).

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message