Some more information:
1) I've got 3 instances of the same configuration environment set up:
SIT, UAT and production
2) Production is running on 2 physical server hooked up to the same
switch (next to each other). They're on the same IP segment
3) SIT and UAT each are 2 Solaris zones hosted on the same physical
machine and they're also on the same IP segment
On Fri, Sep 26, 2008 at 7:33 PM, CS Wong <lilwong_at_gmail.com> wrote:
> Hi,
>
> I've had my Glassfish cluster running for some time in production and
> over time I've been adding more modules to it. I have this problem
> that has been steadily getting worse over time.
>
> When I run "asadmin start-cluster <cluster-name>", I can see in both
> my instance server.log the following line:
>
> [#|2008-09-26T19:00:43.680+0800|INFO|sun-appserver9.1|javax.ee.enterprise.system.tools.synchronization|_ThreadID=10;_ThreadName=main;|IdentityManager
> Data: User:admin|#]
>
> After a long time, then only I'd see something like the following:
>
> [#|2008-09-26T19:22:31.521+0800|INFO|sun-appserver9.1|javax.ee.enterprise.system.tools.synchronization|_ThreadID=10;_ThreadName=main;1307.017;|SYNC010:
> Synchronization completed in 1307.017 seconds.|#]
>
> That's like 20 minutes just for the synchronization alone... and there
> doesn't seem to be any logs at all about what is being done.
>
> Eventually, the startup would still proceed, but I'm worried about
> this. Is this considered normal behavior for a clustered environment?
> Is there anything I can do to speed things up?
>
> Thanks,
> Wong
>