-----Original Message-----
From: Christopher Schultz <ch...@christopherschultz.net>
Sent: Tuesday, August 28, 2018 3:35 PM
To: users@tomcat.apache.org
Subject: Re: [OT] Tomcat Clustering Support
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256
Scott,
I'm no Tomcat-clustering expert, but...
On 8/28/18 13:59, Scott Evans wrote:
> <Cluster className="org.apache.catalina.ha.tcp.SimpleTcpCluster"
> channelSendOptions="10" channelStartOptions="3">
>
> <Manager className="org.apache.catalina.ha.session.BackupManager"
> expireSessionsOnShutdown="false"
> notifyListenersOnReplication="true" mapSendOptions="10" />
>
> <Channel
> className="org.apache.catalina.tribes.group.GroupChannel">
> <Membership
> className="org.apache.catalina.tribes.membership.McastService"
> address="228.1.0.13" port="45522" frequency="500" dropTime="15000"
> soTimeout="10000" domain="mercer"/> <Receiver
> className="org.apache.catalina.tribes.transport.nio.NioReceiver"
> address="10.255.250.34" port="4003" selectorTimeout="8000"
> maxThreads="25" />
>
> <Sender
> className="org.apache.catalina.tribes.transport.ReplicationTransmitter
">
>
>
<Transport
> className="org.apache.catalina.tribes.transport.nio.PooledParallelSend
er"
>
>
timeout="8000" poolSize="25"
> /> </Sender> <Interceptor
> className="org.apache.catalina.tribes.group.interceptors.TcpPingInterc
eptor"/>
>
>
<Interceptor
> className="org.apache.catalina.tribes.group.interceptors.TcpFailureDet
ector"/>
>
>
<Interceptor
> className="org.apache.catalina.tribes.group.interceptors.MessageDispat
ch15Interceptor"/>
>
>
>
> <Interceptor
> className="org.apache.catalina.tribes.group.interceptors.StaticMembers
hipInterceptor">
>
> <LocalMember
> className="org.apache.catalina.tribes.membership.StaticMember"
> domain="mercer" uniqueId="{5,6,7,0,1,2,3,4,0,0,0,0,0,0,1,0}"/>
>
> <Member
> className="org.apache.catalina.tribes.membership.StaticMember"
> port="4001" host="10.255.250.35" domain="mercer"
> uniqueId="{5,6,7,0,1,2,3,4,0,0,0,0,0,0,0,0}" />
>
> <Member
> className="org.apache.catalina.tribes.membership.StaticMember"
> port="4002" host="10.255.250.35" domain="mercer"
> uniqueId="{5,6,7,0,1,2,3,4,0,0,0,0,0,0,0,1}" />
>
> <Member
> className="org.apache.catalina.tribes.membership.StaticMember"
> port="4004" host="10.255.250.34" domain="mercer"
> uniqueId="{5,6,7,0,1,2,3,4,0,0,0,0,0,0,1,1}" />
>
> </Interceptor>
>
>
> </Channel>
>
> <Valve className="org.apache.catalina.ha.tcp.ForceReplicationValve"
> /> <Valve className="org.apache.catalina.ha.tcp.ReplicationValve"
> filter=".*javax\.faces\.resource.*"/> <Valve
> className="org.apache.catalina.ha.session.JvmRouteBinderValve"/>
>
> <Deployer
> className="org.apache.catalina.ha.deploy.FarmWarDeployer"
> tempDir="/tmp/war-temp/" deployDir="/tmp/war-deploy/"
> watchDir="/tmp/war-listen/" watchEnabled="false"/>
>
> <!--<ClusterListener
> className="org.apache.catalina.ha.session.ClusterSessionListener"/>-->
>
>
</Cluster>
It looks like you have both multicast AND static membership enabled.
Keichi's presentation on Clustering at ApacheCon Miami (2017) has a slide
(it's slide #38 here:
https://events.static.linuxfound.org/sites/events/files/slides/TomcatClu
ster_3.pdf)
that says that using static-membership requires that you disable multicast.
Also, just confirming that you have two Tomcat nodes on one IP address
(10.255.250.35, ports 4001 and 4002).
Can you post a thread dump of a deadlock situation? Only the deadlocked
threads should really be necessary to post. Can you replicate the deadlock
without using your own full application? That is, can you create a simple
application that can be used to reproduce this on a similarly-configured
test instance (cluster) of Tomcat nodes?
- -chris
-----BEGIN PGP SIGNATURE-----
Comment: Using GnuPG with Thunderbird - https://www.enigmail.net/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=eKZv
-----END PGP SIGNATURE-----
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org
Hi Chris,
Just getting back to this. Thanks for the info on static membership and
multicast, we may give that a try. We've also finally been able to get a
thread dump of a recent deadlock situation that occurred which is good news.
To answer your other questions, you are correct that two Nodes are on one IP
and we do not have a way to create a simple program to reproduce the issue.
It is just randomly occurring, though very infrequently.
For next steps, due to the sensitive nature and voluminous amount of the
data contained in the thread dump, we would ask that we somehow take this
offline so it's not shared on the forums. Please let me know what our
options would be for that.
Thanks
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org
