[prev in list] [next in list] [prev in thread] [next in thread] 

List:       jboss-development
Subject:    Re: [JBoss-dev] Farming on JBoss
From:       Scott Marlow <scott.marlow.opensource () gmail ! com>
Date:       2005-07-25 20:12:34
Message-ID: 1122322355.7076.25.camel () smarlow ! dnsdhcp ! wal ! novell ! com
[Download RAW message or body]

I posted some farm deploymment documentation on the wiki which I hope
helps resolve some issues. See
http://wiki.jboss.org/wiki/Wiki.jsp?page=JBossFarmDeployment .  I also
added some FAQ info at
http://wiki.jboss.org/wiki/Wiki.jsp?page=JoinTheClusterBeforeUpdatingTheFarmDirectory \
(linked from Cluster FAQ).

Also of interest, might be
http://www.jboss.org/index.html?module=bb&op=viewtopic&p=3876167#3876167
which mentions some of the JBoss5 plans.  

1.  The startup order problem seems to be resolved (couldn't reproduce
it in AS 4.0.3).  I wanted to leave JBAS-1238 open in case it reoccurs. 

2.  I like this idea, see reference to JBoss5 plans above (there are
similar ideas posted there).

3.  Is there a Jira task for this problem?

4.  This is currently true (in 4.0.2+ and probably earlier.)  When files
are pushed out to the cluster farm, they aren't recursively
re-replicated unless they are updated.  



On Mon, 2005-07-25 at 14:54 -0400, acoliver@jboss.org wrote:
> So I think there are enough bugs open on farming ATM to say "it needs
> some help".  I wanted to suggest what I think it needs.
> 
> Thoughts:
> 
> 1. the startup order ought not to matter (bug opened but not fixed --
> novell looked at it)
> 
> 2. there ought to be a controlled deploy pattern,
> preferrably configurable (meaning first server A, then server B, then
> server C...wait n minutes, then server D, then server E, then server F)
> -- suggest "seconds between node deployments", "node groups (in either
> fixed ip ranges or order in cluster indicies)", "seconds between node
> groups".
> 
> 3. if the master who has the farmed thing dies it ought not under any
> circumstances undeploy (not sure why this happens sometimes enough to
> document).
> 
> 4. The file ought to be copied to the farm director on each node.  The
> deployer ought to mark it as "old" so that it isn't recursively
> re-replicated.
> 
> -Andy
> 
> 
> 
> -------------------------------------------------------
> SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
> from IBM. Find simple to follow Roadmaps, straightforward articles,
> informative Webcasts and more! Get everything you need to get up to
> speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
> _______________________________________________
> JBoss-Development mailing list
> JBoss-Development@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/jboss-development



-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
_______________________________________________
JBoss-Development mailing list
JBoss-Development@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jboss-development


[prev in list] [next in list] [prev in thread] [next in thread]