[torqueusers] Java job getting killed in MOM as soon as it gets started

[Prakash Velayutham]

Hi All,

I am noticing a problem with java jobs. The nodes are Opterons (x86_64).

1) When the VM is started with no default -Xms and -Xmx values, jobs get 
killed when they get to about 84 MB of RES memory limit (this is very 
repeatable as I have checked this several times), but the manpage on 
java says that the default value is 64 MB.
2) When I specify a -Xms200m and -Xmx2500m, job is correctly killed when 
it exceeds 2.5GB of RES memory.
3) When I set -Xms200m and -Xmx3500m or -Xmx4000m, job gets killed in 
less than 10 seconds of getting started.

Anyone has a clue on why this is happening?

TIA,
Prakash