> 24 feb 2015 kl. 17:55 skrev Simon Andrews :
>
>
>
>> On 24/02/2015 16:34, "Reuti" wrote:
>>
>>
>>>
>>> Thanks for the reply. I kind of guessed this might be the answer. I
>>> think the issue is that we're assuming that the allocated memory will be
>>> available as a single block, wherea
21 apr 2014 kl. 19:59 skrev Prentice Bisbal :
> After one of these qrsh jobs fails, I get the following e-mail:
>
> Job 5326173 caused action: Job 5326173 set to ERROR
> User=
> Queue =pow1...@.
> Start Time =
> End Time=
> failed assumedly before job:can't get p
Hi,
I was about to suggest oneSIS when I saw that Dave was suggesting it. I would
still put my two cents on it, given that we run a small, but very heterogenous
cluster using oneSIS for node management. One of the aspects I like most with
it, is that adding a new node is just a matter of settin
Another option would be to make h_rt a forced complex in the complex
configuration with no default. We have the following line:
#name shortcut typerelop requestable consumable
default urgency
#--
Try setting your locale to C or en_US.UTF-8 (eg export LC_ALL=C) before running
qstat. qhost failed on me when using sv_SE.UTF-8.
Mikael
13 mar 2013 kl. 23:31 skrev "Stefano Bridi" :
> The setup is a fresh installed: Centos6.4 with
> gridengine-qmaster-8.1.3-1.el6.x86_64
> gridengine-execd-8.
ssion denied
Mikael
---
Sent from a crippled computer (a.k.a a phone)
13 mar 2013 kl. 23:10 skrev "Mikael Brandström Durling" :
> Hi sge users,
> I have been testing the USE_CGROUPS option that is available to execd. When
> USE_CGROUPS is enabled it works fine to submit job
Hi sge users,
I have been testing the USE_CGROUPS option that is available to execd. When
USE_CGROUPS is enabled it works fine to submit jobs one by one. But when I
submitted 70 serial jobs, all queues on all hosts were set to error state. It
happens after 2 or more jobs have started on the host
12 mar 2013 kl. 15:33 skrev Reuti
:
>
>> The problem we would like to solve is when users submit a job with software
>> that defaults to start a number of worker threads equal to the number of
>> cores, and thus parasitising on other jobs allocations within the node.
>
> Often this comes due
12 mar 2013 kl. 12:12 skrev Reuti
:
> Hi,
>
> Am 12.03.2013 um 10:41 schrieb Mikael Brandström Durling:
>
>> I noticed in the man page for sge_conf that there is an experimental option
>> for enabling cpusets in age. I tried to search the mail list archives and
>
Hi,
I noticed in the man page for sge_conf that there is an experimental option for
enabling cpusets in age. I tried to search the mail list archives and google
for documentation on how to enable it, which led to the
util/resources/scripts/setup-cgroups-etc. Calling it from the sgeexecd init
8 mar 2013 kl. 23:38 skrev Dave Love
:
>
> I'm using builtin remote startup.
Ok, I switched to that and now the h_vmem limits behave as expected.
>
>> I'm using the rshd-wrapper and pam_sge-qrsh-setup.so. However, now
>> after the upgrade I noticed that the rshd-wrapper will just create an
>>
6 mar 2013 kl. 23:08 skrev Mikael Brandström Durling :
>
>
> 6 mar 2013 kl. 19:33 skrev Dave Love
> :
>
>> Reuti writes:
>>
>>>> I can't reproduce that (with openmpi tight integration). Doing this
>>>> (which gets three four-co
6 mar 2013 kl. 19:33 skrev Dave Love
:
> Reuti writes:
>
>>> I can't reproduce that (with openmpi tight integration). Doing this
>>> (which gets three four-core nodes):
>>>
>>> qsub -pe openmpi 12 -l h_vmem=256M
>>> echo "Script $(hostname): $TMPDIR $NSLOTS"
>>> ulimit -v
>>> for HOST in $(
27 feb 2013 kl. 20:38 skrev Reuti
:
> Am 27.02.2013 um 16:22 schrieb Mikael Brandström Durling:
>
>> Ok, seems somewhat hard to patch without deep knowledge of the inner
>> workings of GE. Interstingly, if I manually start a qrsh -pe openmpi_span N,
>> and then qrsh
sonable request.
Thanks for your rapid reply,
Mikael
26 feb 2013 kl. 21:32 skrev Reuti
:
> Am 26.02.2013 um 19:45 schrieb Mikael Brandström Durling:
>
>> I have recently been trying to run OpenMPI jobs spanning several nodes on
>> our small cluster. However, it seems to
Hi,
I have recently been trying to run OpenMPI jobs spanning several nodes on our
small cluster. However, it seems to me as sub-jobs launched with qsub -inherit
(by openmpi) gets killed at a memory limit of h_vmem, instead of h_vmem times
the number of slots allocated to the sub-node. Is there
16 matches
Mail list logo
