Orion Poplawski or...@cora.nwra.com writes:
This would be fairly manual, something like:
- Backup /usr/share/gridengine, /var/spool/gridengine/cell
- use yum swap/shell to replace the gridengine-* packages with sge-* packages
- Move the old /usr/share/gridengine/, /var/spool/gridengine/cell
berg...@merctech.com writes:
We've got a job that was suspended via:
qmod -sj $jobid
that's continuing to run. The job consists of a BASH script, which in
turn submits other jobs in a loop, sleeping for 30 seconds after each loop.
When I examine the job status on the node where it
Version 8.1.8 of the Son of Grid Engine distribution is available from
http://arc.liv.ac.uk/downloads/SGE/releases/8.1.8/, with a few bug
fixes and enhancements.
The packaging now should work on any recent Fedora and Debian, and there
are now RPMs for all the current Fedora/EPEL releases
Version 8.1.7 of the Son of Grid Engine distribution is available from
http://arc.liv.ac.uk/downloads/SGE/releases/8.1.7/, with various bug
fixes and enhancements. A notable fix is for the longstanding problem
of occasional major space leaks in qmaster with schedd_job_info=true.
Please report
I discovered too late that this release doesn't build with later (than
in RHEL6 and Debian wheezy) versions of GNU binutils, due to one test
program. You can either just delete the dependence on test_drmaa.1.0 in
libs/japi/Makefile or patch it like
SGE 8.1.6 is available from
http://arc.liv.ac.uk/downloads/SGE/releases/8.1.6/, fixing various bugs.
Please report bugs, patches and suggestions for enhancement
https://arc.liv.ac.uk/trac/SGE#mail.
Release notes:
* Bug fixes
* Man and message fixes
* Fix building of patched openssh for
You can't, in general, mix grid engine versions -- not 6.2u5 with
anything earlier, I'm pretty sure. The qmaster messages file probably
says so. Trying to can crash the qmaster
https://arc.liv.ac.uk/trac/SGE/ticket/1441.
Alan McKay alan.mc...@gmail.com writes:
On Mon, Sep 23, 2013 at 11:00
François-Michel L'Heureux fmlheur...@datacratic.com writes:
Hi
I need help to write to stdin of a qrsh job.
The easiest example I came with is:
echo test | qrsh cat
It prints test as expected but then it hangs.
That was fixed in SGE 8.0something; download link via URL below.
--
Prentice Bisbal prentice.bis...@rutgers.edu writes:
Hi, everyone,
After not being on this list for about 15 months, I'm now supporting
SGE again and need your assistance with a problem. The problem I have
is the same one Alex Chekholko reported just about a year ago:
Lionel SPINELLI spine...@ciml.univ-mrs.fr writes:
Hello all,
I have a question that is not directly linked to SGE but relates to
the same. Which tool administrators that have to install, manage,
configure and ensure coherence between lot of grid nodes use? I mean,
if I have 10 nodes in my
Joseph Farran jfar...@uci.edu writes:
[Please don't post content-type: text/html.]
Hi all.
We have Grid Engine 8.1.4 running on a cluster with CentOS 6.4, using kernel
2.6.32-358.18.1.We are just getting started on setting up job checkpoint.
We got BLCR compiled and are currently
Son of Grid Engine 8.1.5 is now available from
http://arc.liv.ac.uk/downloads/SGE/releases/8.1.5/. The changes are
mainly to fix build problems, particularly on MS Windows:
Version 8.1.5
-
* Bug fixes
* Fix strsignal-related build failure on MS Windows, at least
* Fix MS
Joseph Farran jfar...@uci.edu writes:
Howdy.
We are running Son of Grid Engine 8.1.3.
I compiled 8.1.4 and downloaded and un-tar the gui_installer-8.1.4.tar into
the compiled directory.
Why do you want the GUI installer if you're doing an upgrade? Does it
actually support upgrades?
Adam Brenner aebre...@uci.edu writes:
FYI: The fix for this was to include the IzPack jar files into CLASSPATH:
export CLASSPATH=/path/to/IzPack/lib/*:$CLASSPATH;
./start_gui_installer
You don't need an installed IzPack, just the installer .jar. Perhaps
the problem is that you have one, and
William Hay w@ucl.ac.uk writes:
my second problem, i have to find out the progress of reserving.
may be a job requests 4 slots, i have to find out the time when it
starts reserving, when it has got 2 slots, when it has got the 3rd
slot and so on.
This appears to embody a slight
Reuti re...@staff.uni-marburg.de writes:
Please have a look at your /tmp. The starting execd will write the cause of
not being able to start in a file therein.
For what it's worth, that depends on the version. sge-8.0.0e+ writes to
syslog, as you'd expect a daemon to. (The previous
Txema Heredia Genestar txema.llis...@gmail.com writes:
Hi all,
I have a cluster in production running rocks-cluster 6.0 using SGE6.2u5.
SGE6.2u5 has a bug that kills the qmaster when an amount of jobs using
both -pe and -hold_jid are used. OGE (theoretically) has this bug
fixed.
OGE is
Tina Friedrich tina.friedr...@diamond.ac.uk writes:
Hi Txema,
I recently upgrades our Grid Engine from SGE6.2u4 (I think it was) to
OGE8.1.3.
I think that means SGE 8.1.3.
SGE6.2 and OGE8.1.3 execd's happily coexist on the same nodes;
however, I found that the qmaster processes don't,
Brendan Moloney molo...@ohsu.edu writes:
Hello,
I use multiple queues to divide up available resources based on job
run times. Large parallel jobs will typically span multiple queues and
this has generally been working fine thus far.
I'd strongly recommend avoiding that. Another reason
Guillermo Marco Puche guillermo.ma...@sistemasgenomicos.com writes:
Hello,
I'm very interested in a web application like xml-qstat:
http://xml-qstat.org/.
Is there any other alternative to this project (I would like to know
the available options before start installing stuff on my
Reuti re...@staff.uni-marburg.de writes:
You can switch on params MONITOR=1 in `qconf -msconf` (`man
sched_conf`),
And perhaps use qsched(1) from recent SoGE to analyze the results.
--
Community Grid Engine: http://arc.liv.ac.uk/SGE/
___
users
Black, Chris chris.bl...@roche.com writes:
The qsub cmdline I am using is:
qsub -q rnd.q -t 1-50 -tc 2 -N cbTA -j y -cwd /path/to/task.sh
task.sh just contains some echos and a sleep 600.
On our 6.2u4 cluster, the scheduler properly only runs two of the 50 tasks
at once. On SoGE 8.1.x all
ChrisDag d...@sonsorol.org writes:
Kind of funny as I have an active consulting gig right now to move a
financial services group away from Torque/Moab due to poor performance.
I heard poor reports of Moab a while ago from a site which was supposed
to be a flagship for it.
Anyone else get
Joe Borġ m...@jdborg.com writes:
Yep, removing that from the sudoers file has fixed the issue!
Thanks guys.
Regards,
Joseph David Borġ
http://www.jdborg.com
On 5 September 2013 13:54, Joe Borġ m...@jdborg.com wrote:
I'm about to try hashing out `requiretty` in the sudoers file. I
François-Michel L'Heureux fmlheur...@datacratic.com writes:
While investigating jobs that have been running for way too long, I've
found out that qhost shows nodes that are dead with alive stats such
as load, memuse and swapus. qstat also shows them processing jobs with
state r, as if the
Reuti re...@staff.uni-marburg.de writes:
I don't know whether it will be different in old versions, but that's
what I'd expect.
Argh, true - I missed the 6 and tried with 1. - Reuti
I know the feeling!
--
Community Grid Engine: http://arc.liv.ac.uk/SGE/
Reuti re...@staff.uni-marburg.de writes:
If you want to keep a dedicated CPU and its cores out of the business:
you can try to set an affinity mask on these CPUs before you start the
sge_execd. I would expect that all forks of this process will inherit
this.
The shepherd won't take any
William Hay w@ucl.ac.uk writes:
You could have a look at Gold
http://www.clusterresources.com/products/gold-allocation-manager.php
We've been integrating it into our SGE cluster/JSV recently for a
similar but slightly different usage but have yet to use it in anger.
The Gold docs
Son of Grid Engine 8.1.4 is now available from
http://arc.liv.ac.uk/downloads/SGE/releases/8.1.4/. The changes are
mainly bug fixes plus some minor enhancements. See below for the
significant ones. There are various binaries available as well as the
source, including Debian ARM packages in
berg...@merctech.com writes:
In a continuing effort to resolve the problem where the queue gets put into an
error state with the message (in qstat):
=
= can't open file
/opt/sge/6.2u5/default/spool/r820-1/active_jobs/93629.1/pe_hostfile:
Permission denied
=
I've enabled
Reuti re...@staff.uni-marburg.de writes:
• Job comes back to R status.
Do you use any checkpointing interface, to restart the job? If so, it
should output Rr in `qstat` instead of a plain R for the SGE job state.
No, I don't use any checkpointing interface.
Then the state should
Reuti re...@staff.uni-marburg.de writes:
Am 26.08.2013 um 15:57 schrieb Julien Nicoulaud:
I meant non-interactive qrsh, eg:
$ qrsh -cwd -now no -binding env linear:6 -V -b yes /bin/env
= No SGE_BINDING env
I don't know whether it will be different in old versions, but that's
what I'd
Jewell, Chris c.p.jew...@massey.ac.nz writes:
The message is from a failure of setuid(2) or similar. I don't know if
it's a libc bug that errno seems no to be set (Success) as it should
be.
The two possible cases are:
EAGAIN The uid does not match the current uid and uid brings
Sorry, I apparently missed this before.
William Hay w@ucl.ac.uk writes:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 01/08/13 04:09, Jewell, Chris wrote:
Hello all,
A while since I posted here, so good to be back!
My installation of GE 8.1.3 from the Scientific Linux 6.3 RPM
William Hay w@ucl.ac.uk writes:
You could just add a starter_method to your queue configuration. This
gets run instead of your command and passed enough info to recreate
the command that would have been run. We do something similar with
cpusets rather than cgroups.
I'd expect the
Ed Lauzier elauzi...@perlstar.com writes:
Hi Dave,
I found the section where the static buffer is defined.
I'm thinking on the best way to handle this for the patch.
Having a static buffer is fine especially for the size it is.
Properly handling variables that exceed this limit is where
Jean-Baptiste Denis jbde...@pasteur.fr writes:
Hello everybody,
I'm using OGS/GE 2011.11p1 on Linux and I want to play with cgroups.
What do you want to do with them? (They're not without problems.)
My idea was simply to dynamically create a cgroup on an execution node
per JOBID using a
Reuti re...@staff.uni-marburg.de writes:
Hi,
Am 22.08.2013 um 10:12 schrieb Jean-Baptiste Denis:
Hello everybody,
I'm using OGS/GE 2011.11p1 on Linux and I want to play with cgroups.
My idea was simply to dynamically create a cgroup on an execution node
per JOBID using a prolog (and
Stephen Spencer spen...@cs.washington.edu writes:
Dave,
[...] don't let users use direct ssh to compute nodes. isn't an option.
How might I configure SGE to take into account the current node load - it
sounds like you may know.
Reuti pointed at the documentation, but you seem to have a
Jesse Becker becker...@mail.nih.gov writes:
That rather depends on how unsafe it was to start with...
Nah, 5 halflives is 5 halflives. It's just a question of *time* more
than initial danger. ;-)
Doubtless. I've learned that we physicists don't understand such
things.
This is especially
Tina Friedrich tina.friedr...@diamond.ac.uk writes:
The real and effective user is not root, and never was. Never caused us any
problems. The NFS share is exported with root_squash.
If that's the spool area, it means you get world-writable files in it.
This is quite interesting. And all
Reuti re...@staff.uni-marburg.de writes:
That isn't getting to my question though. The man page says the number of
slots reserved for the job is the number of slots requested with the -pe
switch + 1. But I don't see any evidence of that, unless I don't
understand what slots reserved
Reuti re...@staff.uni-marburg.de writes:
The file was created on Windows and the lines end in CR/LF, the ? represents
the CR.
There is a tool dos2unix (or use an `sed` command) to remove these characters
from the script file.
I use JSV to catch that:
# Check for the common case of
Ray Pete rp...@broadinstitute.org writes:
Udo,
*Assuming you are running a RH like distro and your service cgconfig is
running*
It was quite some time ago I set this up but If I remember I had trouble
with 2011.11p1 for some reason. ( probably user config error on my part :-)
)
I don't
Tina Friedrich tina.friedr...@diamond.ac.uk writes:
Hello list,
have finally decided to look into upgrading our SGE6.2 installation -
mainly to see if it helps with my job scheduling problem.
From which version? It should be essentially trivial if you're
upgrading from 6.2u5 (after doing
Kevin Buckley kevin.buckley.ecs.vuw.ac...@gmail.com writes:
FLAC ? Ogg ? PulseAudio ?
So then I read the way the dependencies have developed and it seems that it's
down to the requirement for Java, as satisfied in the CentOS 6.4 distro.
-- Processing Dependency: java = 1.6.0 for package:
Txema Heredia Genestar txema.llis...@gmail.com writes:
Hi all,
I was wondering if there is any way to allow a user to choose how many
jobs they want to have running concurrently in the cluster. I am aware
that I, as an administrator, can specify limits in the slot usage for
each user whith
Orion Poplawski or...@cora.nwra.com writes:
On 05/13/2013 09:40 AM, Orion Poplawski wrote:
Would it be possible for qhost -xml output to include the number of slots
used
by a job on that host?
Okay, I see how it is done - there are multiple job entries for each
slot.
Indeed, like the
Esztermann, Ansgar ansgar.eszterm...@mpibpc.mpg.de writes:
Hello List,
the JATASK problem is back. It seems to be limited to jobs in E state
(but does not disappear upon qmod -c). Is anyone willing to check
this?
I can't reproduce it, and I've fixed all the cases of malformed XML I
know of
Farid Chabane farid.chab...@ymail.com writes:
Hi Stephen,
Yes, SGE take into account the current load of nodes even if the load was
caused with a non-SGE-job.
That depends on how you configure it.
I'd say don't let users use direct ssh to compute nodes.
--
Community Grid Engine:
Brian McNally bmcna...@uw.edu writes:
It seems that using qacct to
display overall usage per user (-o), for example, might be a little
misleading if the actual accounting information is stored
internally. Users might draw conclusions about their usage and how
that'll impact their job
Brian McNally bmcna...@uw.edu writes:
See my comment to Reuti about halflife and why that seems
confusing. For GE to penalize users beyond one halflife, it'd have to
retain job data for longer than one halflife.
?? All it has to do is to decay/accumulate over the relevant interval.
You can
Mark Dixon m.c.di...@leeds.ac.uk writes:
We use the share tree here, rather than the functional policy, so this
might not be applicable.
By default, the usage of a job is wholly based on slots*seconds.
I think it's only (effectively) slots if you use
sharetree_reserved_usage, as we do. I'm
Reuti re...@staff.uni-marburg.de writes:
Does anyone know a way to define this option in the gridengine directly ?
I'm using ogs 2011.11p1 on Intel 6core processors.
With 6 cores per machine I would simply request full nodes.
[I guess they're two-socket; we don't insist on filling those,
Reuti re...@staff.uni-marburg.de writes:
Do you mean it takes resource requests and their reservation into
account properly?
No, not directly. It scans the output of `qstat` of any given queue
for waiting jobs (hence: which may run in this queue) and bases the
decision on this.
I'm sure
Quy NGUYEN DAI clus...@gdtech.eu writes:
Hi all,
I just setup Son GE on a small Linux cluster. It works well.
Now I would like to setup this GE with LS-DYNA which uses a Network License
server on LAN.
Already look for that on Internet but I can not found any example to do
that.
Can GE
Reuti re...@staff.uni-marburg.de writes:
I don't use calendars, and I didn't know there was that interaction with
reservation, but I suppose it makes sense in terms of implementing
look-ahead. Is it documented somewhere already?
It seems to be an effect of the enabled backfilling by this
Jake Carroll jake.carr...@uq.edu.au writes:
Hi Grid-Engine gurus of the grid engine list.
I'll answer anyway.
I have a question about the fair-share policy and the subsequent
algorithm/ratios it uses to manage user workload.
Currently, we've set our grid engine up for fair-share in a very
Mahbube Rustaee rust...@gmail.com writes:
Hi all,
How can run GAMESS under GE?
You need to be specific about which GAMESS, and how it was built, at
least.
--
Community Grid Engine: http://arc.liv.ac.uk/SGE/
___
users mailing list
Jesse Becker becker...@mail.nih.gov writes:
This is a problem that has plagued me, and various people time and
again, and every time it gets fixed, the method and cause seems to get
lost in the aether. I've hit it several times over the years, and each
time the problem and solution see to
Sangamesh Banappa sangamesh.bana...@locuz.com writes:
Hi,
I need some details on the accounting data captured by GridEngine. The
qacct output for the last 90 days is:
Please make a bug report with suggestions for improvement if it's not
adequately explained under
Adam Brenner aebre...@uci.edu writes:
In our case, it is very helpful for our users to directly SSH into the
nodes to determine what is wrong with their qsub scripts, etc. This is
a follow up the following thread by Joseph and Harry:
Joe Borġ m...@jdborg.com writes:
Hi Guys,
I was wondering if anyone has any guidance with installing execd onto
Windows 7 using Microsoft's SFU.
In what respect? If you want to build it, see source/README.windows in
SGE 8.1.3. Otherwise you probably need to look at the Oracle docs for
more
Bernard Massot bernard.mas...@u-psud.fr writes:
I'm using gridengine 6.2u5 on Debian Squeeze.
I recommend not using that for various reasons. You can build Debian
packages from the SGE 8.1.3 distribution (installing into /opt).
Unfortunately it looks as if SGE in Debian is dead -- no-one seems
Reuti re...@staff.uni-marburg.de writes:
You mean one level above - in /tmp or alike? The $TMPDIR name is
usuall $JOB_ID.${SGE_TASK_ID/undefined/1}.$QUEUE.
I've changed that to use the cell name, not the queue for SGE 8.1.4. If
anyone thinks that will break more than it fixes, please speak
Reuti re...@staff.uni-marburg.de writes:
On the exechost? I don't do it at all on a per job basis. In case your users
fight for the disk space you can implement a consumable for the disk space in
combination with a load sensor:
Reuti re...@staff.uni-marburg.de writes:
I guess a calendar is the simplest option then? Will SGE refrain from
scheduling jobs if I create a maintenance calendar and connect it to
all queues?
If max_reservation is set to a value different from zero: yes. Then
it will take the set calendar
Reuti re...@staff.uni-marburg.de writes:
For serial and parallel SMP jobs there is:
http://wiki.gridengine.info/wiki/index.php/StephansBlog as an
option. Maybe instead of using slots a custom complex is necessary
which is called medium and all jobs of this type have to request
it. This is
Reuti re...@staff.uni-marburg.de writes:
Some time ago Fritz mentioned that SCM to control it is abandoned and
Univa integrated something new to control it.
SDM is still available if you want it, and the others I'm aware of are
on http://arc.liv.ac.uk/SGE/tools.html, apart from one funded with
Nicolás Serrano Martínez-Santos nserr...@dsic.upv.es writes:
Hi,
We are currently using SGE6.2u5 in our little cluster (~150 cores) and I am
trying to configure it to manage GPU correct usage. I have been able to define
multiple slots for each GPU card and also to reserve memory using
Stephen Willey step...@esstec.co.uk writes:
You could use a load sensor to do this. We use one to detect if
people are logged in and suspend/requeue the jobs if someone logs in
while a job's on their workstation.
I don't understand how that addresses the question (as I understand it).
Thomas Mainka t.mai...@science-computing.de writes:
Hi everyone,
currently I am trying to upgrade an old SGE cluster from an 6.2 release to SGE
8.1.3. The upgrade itself is no problem, however the sge_execd of 8.1.3
consumes a
lot of CPU on the exec host when a job is running--sometimes
Mikael Brandström Durling mikael.durl...@slu.se writes:
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
Stefano Bridi stefano.br...@gmail.com writes:
Many thanks, in this way it works.. strange behaviour anyway..
stef
It was the result of trying to accommodate the system distributed over
different locales with the floating point data exchange in text format
(ugh) -- from long experience of that
Campbell McLeay campbell.mcl...@primefocusworld.com writes:
Thanks for all the suggestions. In the end the easiest was to just
monitor the qmaster log and run a qmod -sq on the execution host to
suspend it, then a qmod -cj to resubmit the job. This will do until
sssd gets fixed
Off topic,
Reuti re...@staff.uni-marburg.de writes:
Hi,
Am 20.03.2013 um 15:58 schrieb Lars van der bijl:
hey everyone,
a few weeks ago we where having issue with users submitting jobs the
did massive IO to the file system. each task was writing out about 4
GB of data and reading in about 2GB. as
and I'm surprised the problem
hasn't come to light before. Thanks.
* Don't pass any user environment to remote startup daemons -- better
fix for half of CVE-2012-0208
I think this change should fix it, but I can't test it immediately.
Fri Mar 8 22:04:13 GMT 2013 Dave Love d.l
Jim Phillips j...@ks.uiuc.edu writes:
The slots feature is only needed to support pe size ranges, e.g.,
qsub -pe smp 10-20.
If so, I wouldn't have bothered with the small effort to do it.
Otherwise you can set the binding size to match
the pe size in a jsv script. Multi-host jobs should
Reuti re...@staff.uni-marburg.de writes:
You'll be lucky if RPMs work on a Debian system.
Well, `rpm2cpio` or `alien` could help.
I meant because of shared library names and versions rather than the
packaging.
--
Community Grid Engine: http://arc.liv.ac.uk/SGE/
Erik Soyez e.so...@science-computing.de writes:
Thanks Reuti for your quick-as-always answer!
I have added an additional express-PE for each existing PE which
should ensure that express jobs really stick to the express queue;
users submit with wildcard-PEs anyway, so they don't need to
Ian Johnson ian.john...@capita-ti.com writes:
Reuti,
Problem solved. It was a hostname lookup problem. Once all the hosts
had correct hosts files qrsh can now connect to any slave from all
submit hosts. Thank you very much for your help over the last few
days.
Below is the trace from
Reuti re...@staff.uni-marburg.de 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 $(tail -n +2 $PE_HOSTFILE|cut -f1 -d' ');
Jim Phillips j...@ks.uiuc.edu writes:
Just enable core binding in the scheduler and use a jsv script to
force binding for all jobs. Jobs run only on their assigned set of
cores regardless of how many threads they launch.
I don't know what platform that refers to, but it is simply false on
Reuti re...@staff.uni-marburg.de writes:
Am 01.03.2013 um 15:17 schrieb Jacques Foucry:
It looks like qmaster is not running
# ps aux | grep qmaster
root 2218 0.0 0.0 7548 832 pts/0S+ 17:32 0:00 grep qmaster
So I try to launch it:
# /etc/init.d/gridengine-master start
Kshitiz B kshiti...@tcs.com writes:
How to distinguish between the following scenarios which leads to job
deletion :
1. Slave Node Failure
2. Master/Shepherd Node Failure
3. Job deleted by User
4. Job deleted by Admin
I just answered on the SGE list; please don't post separately to both
Raymond Wan rwan.w...@gmail.com writes:
RQS sounds like what I'm looking for (for #1) -- the problem I'm
having is that users are being a bit greedy when I'm not watching.
It doesn't seem to have been said that RQS only affects scheduling the
job, not what happens at run time. It specifically
Gaya Nadarajan gay...@ed.ac.uk writes:
Hi,
I'm from Edinburgh but I'm not using the cluster here, I have a
private cluster set up for a project elsewhere. Anyhow, the problem is
associated with the execution daemon failing to acknowledge jobs sent
from the qmaster, causing some jobs to be
Jim Phillips j...@ks.uiuc.edu writes:
That error means that the process launched by qrsh on node09 exited
before the rest of the slots so qmaster killed everything for you.
It's not just that it exited, but that it failed for some reason.
I'd set the log level to info and maybe check the
Reuti re...@staff.uni-marburg.de writes:
Am 27.02.2013 um 20:56 schrieb Mikael Brandström Durling:
snip
In case you look deeper into the issue, it's also worth to note that there
is no option to specify the target queue for `qrsh -inherit` in case you
get slots from different queues on
rk mag hastyt...@gmail.com writes:
hi guys.. i am really just new in using a mailing lists..
anyways, I just wanna ask if Open grid scheduler works on openSUSE?
I don't know, but you can expect
http://arc.liv.ac.uk/downloads/SGE/releases/8.1.3/ to work, although I
haven't tried the latest
Son of Grid Engine 8.1.3 is available, not before time, from
http://arc.liv.ac.uk/downloads/SGE/releases/8.1.3/. The changes are
mainly bug fixes, plus support for packaging as a Debian add-on and a
few enhancements pulled in by dependencies. Building on MS Windows is
now tested and documented
William Hay w@ucl.ac.uk writes:
On 25 January 2013 09:56, Semion Chernin s...@bgu.ac.il wrote:
How I can help to this user?
The trouble is that for some reason the program I am running will output
to the console while run directly on a node, but while submitted as a job -
will not
Stefano Bridi stefano.br...@gmail.com writes:
Hi all, is there a way to use the scratch area (local disk) on the
compute node in a transparent way from the submitted script point of
view?
What I want to do is to copy to and from the compute node scratch area
the job data using the
Reuti re...@staff.uni-marburg.de writes:
Am 25.01.2013 um 18:37 schrieb Tina Friedrich:
Don't know if that helps you, but remember that SGE created a temporary
directory (something like /tmp/#JOB_NO.#TASK_NO.QUEUE). Available to
submitted scripts as $TMPDIR. I believe it even cleans it up
Reuti re...@staff.uni-marburg.de writes:
I copied the `qmake` out of
http://arc.liv.ac.uk/downloads/SGE/releases/8.1.2/gridengine-8.1.2-1.el5.x86_64.rpm
and used it under SGE 6.2u5 and the error is gone.
There's then https://arc.liv.ac.uk/trac/SGE/changeset/4265/sge, though,
which needs a
Jake Carroll jake.carr...@uq.edu.au writes:
We figured it out!
Specific user binary was not respecting vf memory complex and decided to
use all the RAM on random nodes it landed on!
There's nothing to respect. If you use vf, it's only relevant to
scheduling jobs, not memory usage while
Reuti re...@staff.uni-marburg.de writes:
This is unusual and looks like someone tried to cleanup the location
of configuration files by good intention, but didn't make it in a nice
way. Maybe the best would be to ask on the crosswords mailing
list. The tutorials I could point you to are not
Txema Heredia Genestar txema.here...@upf.edu writes:
Thanks for the answer Dave.
I have tried your old rpms and it didn't work. The problem still
persists and the qmaster crashes.
Then it's a different problem.
I'll see if I can schedule a maintenance stop to upgrade the system to
a newer
[Excuse any duplicates -- I'm not sure if gridengine.org is tits-up
again as well as our mail hub sulking at my laptop.]
Reuti re...@staff.uni-marburg.de writes:
I think that's an old version. Suggestions are welcome for any
improvements to the current one, which I tried to tidy up (from
Reuti re...@staff.uni-marburg.de writes:
You can only reset it in `qmon` completely.
What is missing on qconf compared with qmon? (I think I've only used
qmon the few times I've hanged it, but it's a bug if you can't do the
same with qconf.)
I missed the -clearusage for `qconf`. So it's
1 - 100 of 521 matches
Mail list logo