[gridengine users] queues behaving differently

[John Young]

On 07/10/2012 04:30 PM, Rayson Ho wrote:
> On Tue, Jul 10, 2012 at 4:23 PM, John Young <j.e.young at larc.nasa.gov> wrote:
>> With this in place, it seems odd that from one of my queues I
>> get a default setting for the number of descriptors of 1024.
>>
>> So I have two questions really:
>>
>> 1. Why am I getting different behavior from the two queues?
>
> Could be OS issue - try upping the limit as root from an interactive
> shell, and see if you can increase the limit?

Hmmm...  If it was an OS issue, I would expect both queues to behave
the same way since both are under the same OS, but they don't.

Curiously, if I try running an interactive job in the same queue
where changing the limits fails in batch mode, it succeeds even
if I am not root.  Very strange...

>
>> 2. Since H_DESCRIPTORS=65536, why can't I request that amount and get it?
>
> I believe you are hitting the OS limit, try something smaller, and/or
> change the limit set by /etc/security/limits.conf &
> /proc/sys/fs/file-max .
>
> Rayson

I had already upped the limit in /etc/security/limits.conf:

#
# Up the number of open files allowed - 02 Jun 2011 - JY
#
*                soft    nofile          65535
*                hard    nofile          65535


Changing this did not seem to have any effect on the problem.
I'm still baffled why it works under one queue but not under the
other.

JY