The only "client" running at the time is the TSM scheduler service, which isn't 
running a backup, it's just waiting like the scheduler service usually does.
No backups or restores going on through the plug-in.
And javaw just appears to take off running for no reason.
We can even reboot the plugin server, and it cranks up again.



-----Original Message-----
From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU] On Behalf Of 
apaschal5 AT frontier DOT com
Sent: Monday, June 17, 2013 12:21 PM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: [ADSM-L] TSM VE plug-in server and java32w.e xe loop

Hi, Wanda.  Out of curiosity, do you keep having to kill javaw multiple times 
if you kill the client first?

Oh, and no, I haven't this problem myself.

Sent from my anonymous smartphone because they don't pay me to advertise.

----- Reply message -----
From: "Prather, Wanda" <Wanda.Prather AT ICFI DOT COM>
To: <ADSM-L AT VM.MARIST DOT EDU>
Subject: [ADSM-L] TSM VE plug-in server and java32w.exe loop
Date: Fri, Jun 14, 2013 14:55


I've seen this twice, at different sites, and hope somebody can tell me what to 
pursue.

TSM VE 6.4.0.0 is installed in a Win2K8-64 VM.
The plug-in server is running there, also the TSM Scheduler service that acts 
as the data mover, and the VE recovery agent.
For the purposes of this discussion, assume that the VSphere client is also 
running there, although the same thing occurs if the VSphere client is running 
elsewhere.
The VM has no other apps installed on it, it's only purpose in life is to serve 
up TSM/VE.

There are no backups or restores running in the plug-in or the scheduler 
service at the time this happens.

The plug-in works fine for days or weeks.
Then CPU utilization goes to 100%, so that starting the plug-in (via the 
VSphere client) takes > 10 minutes.  (Same is true if VSphere client is started 
elsewhere, it takes forever to get the plug-in displayed because the VE server 
is choked.)

Task Manager shows that the culprit is javaw.exe*32, hogging 99% of the CPU..
And it just runs and runs and runs.
I don't know what is doing it.  It cranks up, stops for a while sometimes, then 
starts again.

Rebooting the plug-in server doesn't change things.  javaw.exe*32 cranks up 
again, hogs the machine.

The circumvention is to use task manager to kill off the javaw.exe*32 process.  
It spawns another, you kill it, it spawns another, you kill it, etc. etc.
Eventually, it stops spawning more and you get an html error "cannot display" 
on the plug-in.
Then you restart the plug-in, then everything is fine.  javaw.exe*32 hangs 
around, but is well-behaved thereafter.  Until the next time.

Since there is nothing else on this VM, I have to assume it's something in 
derby or the other plug-in infrastructure that is turning java into a demon 
daemon.   What's weird is that a reboot of the VM doesn't usually cure it, it 
has some recollection of something it thinks it needs to do.

Anybody seen this before, or can give me a clue where to look?

Thanks
W





Wanda Prather  |  Senior Technical Specialist  | Wanda.Prather AT icfi DOT com  
|  www.icfi.com ICF International  | 401 E. Pratt St, Suite 2214, Baltimore, MD 
21202 | 410..539.1135 (o)