Multiple Remote servers on one machine

[Joshua Szmajda <js...@biotest.net>]

Hey all,

I'm trying to maximize the utility of this server I bought for cacheing. 
It has 32Gb of RAM, so due to Java's 2Gb max-memory limit on 32-bit 
linux architectures, I'm trying to run a fleet of cache servers in 
parallel on the same machine.

When I start them up, I start one rmiregistry, then I start a number of 
java processes that all use nearly the same configuration for the 
remote.cache.ccf, except the ClusterServers lists are different 
(depending on which process this is, it won't list itself in the 
config). For example, one remote aux is configured with

registry.port=10021
and
jcs.auxiliary.RCluster1.attributes.ClusterServers=localhost:10022,localhost:10023,localhost:10024,localhost:10025

Now the problem is, in the logs I see lots of "Read timed out" errors. 
Do I need a separate rmiregistry for each virtual machine? How would I 
even configure which rmiregistry to use per process? I can connect to 
each cache individually without trouble, so I know they're up and running.

Is there an easier way?

Thanks!
-Josh

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

Re: Multiple Remote servers on one machine

[Niall Gallagher <ni...@switchfire.com>]

Hi Josh,

Well one obvious question - can't you install a 64-bit Linux
architecture on the server instead?! Surely the OS is 64-bit if it has
that much memory. 64-bit Java is more or less the same as 32-bit Java as
far as I've found, except that it supports more memory.

Niall

-----Original Message-----
From: Joshua Szmajda <js...@biotest.net>
Reply-To: JCS Users List <jc...@jakarta.apache.org>
To: JCS Users List <jc...@jakarta.apache.org>
Subject: Multiple Remote servers on one machine
Date: Tue, 28 Oct 2008 13:49:37 -0400


Hey all,

I'm trying to maximize the utility of this server I bought for cacheing. 
It has 32Gb of RAM, so due to Java's 2Gb max-memory limit on 32-bit 
linux architectures, I'm trying to run a fleet of cache servers in 
parallel on the same machine.

When I start them up, I start one rmiregistry, then I start a number of 
java processes that all use nearly the same configuration for the 
remote.cache.ccf, except the ClusterServers lists are different 
(depending on which process this is, it won't list itself in the 
config). For example, one remote aux is configured with

registry.port=10021
and
jcs.auxiliary.RCluster1.attributes.ClusterServers=localhost:10022,localhost:10023,localhost:10024,localhost:10025

Now the problem is, in the logs I see lots of "Read timed out" errors. 
Do I need a separate rmiregistry for each virtual machine? How would I 
even configure which rmiregistry to use per process? I can connect to 
each cache individually without trouble, so I know they're up and running.

Is there an easier way?

Thanks!
-Josh

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