[torqueusers] getsize() failed for file in mom_set_limits

Steve Young chemadm at hamilton.edu
Tue Sep 23 09:53:34 MDT 2008


On Sep 23, 2008, at 11:08 AM, Arnau Bria wrote:

> On Tue, 23 Sep 2008 10:46:48 -0400
> Steve Young wrote:
>
> Hi Steve,
>
>> I have used this set up in order for users to request machines with
>> enough scratch disk space for their jobs:
>>
>> On the node in mom_priv/config:
>>
>> size[fs=/scratch]
>>
>> Which correlates to the /scratch partition on the machine.
> Ok, same for me till here but using home dir.
>
>> Then when I do a checknode on the server I see:
>>
>> checking node node0001
>>
>> State:   Running  (in current state for 00:00:00)
>> Configured Resources: PROCS: 32  MEM: 30G  SWAP: 18G  DISK: 1649G
>> Utilized   Resources: DISK: 154G
>> Dedicated  Resources: PROCS: 30  MEM: 30G
>
>
> checking node td033.pic.es
>
> State:   Running  (in current state for 00:00:00)
> Configured Resources: PROCS: 8  MEM: 15G  SWAP: 29G  DISK: 103G
> Utilized   Resources: DISK: 25G
> Dedicated  Resources: PROCS: 3
>
>> Now all my users need to do is specify this in their batch scripts
>> to request the amount of scratch space they need:
>>
>> #PBS -l file=500gb
>
> But what are they saying with this line? That they need a node with at
> least 500gb of free space for their jobs? or their job will use
> 500gb at max, so toque is setting ulimit?
> Cause it's clear for me too, but your same (or very similar) env did  
> not
> work here...
>

I believe it is more like "reserve me 500gb of scratch space on a  
machine that has at least that available"  so more a max than a  
minimum. If they go over that the job fails. The same as for asking  
for cpus or memory.


> The other point is that I cannot force my users to specify that, they
> are grid users, so I have to set the resource at queueu level. But it
> should be the same.

One would think it should be ;-).

>
>
>> I'm not sure if this solves your situation as your talking
>> about /home space as opposed to some other partition (/scratch in my
>> example). But I just wanted to point out that this set up has been
>> working for me. Hope this helps,
>
> Yep, Tom Rudwick told me same as you in maui list.
>
>
>> -Steve
> Thanks for your reply,
> Arnau
> _______________________________________________
> torqueusers mailing list
> torqueusers at supercluster.org
> http://www.supercluster.org/mailman/listinfo/torqueusers



More information about the torqueusers mailing list