On Tue, 30 Jun 2015, Alexis Huxley wrote:
Hi, is there any news regarding cgroup-based enforcement of h_vmem?
I saw that Mark Dixon posted patches against 8.1.1 (details at
http://gridengine.org/pipermail/users/2012-September/004699.html), but
these are not in 8.1.8 nor the 20150129 snapshot
On Sun, 5 Jul 2015, Dave Love wrote:
...
I don't understand. If you drop the rlimit, the only control of memory
is qmaster killing when the measured usage exceeds the h_vmem. In
particular, the application can't fail gracefully if it tries to use too
much memory, and it can cause OOM by allocat
ties are
calculated where there are multiple pending jobs for a sharetree node. As
the maths I want to use on our cluster might be different than the maths
someone else needs on theirs, such a thing might be useful...
Cheers,
Mark
--
--
On Mon, 7 Dec 2015, Ondrej Valousek wrote:
Yes, it hurts us quite badly because our compute farm is based on shared
storage - we use NFS&auth_sys. This means we are already limited to 16
groups and GE is allocating 1 for itself. Tried investigate Kerberos
support for GE (which would be a perfe
Hi there,
Is there any interest for a meeting in the UK looking at the internals of
gridengine? Potential topics might be:
* Building from source
* How the code is organised
* How to debug or develop gridengine
The principles discussed ought to be applicable to any flavour of
gridengine that
Hi there,
Playing around with CentOS 7 + SoGE 8.1.9, just noticed that attempts to
view the cluster config from a non-admin host fails:
$ qconf -sconf
denied: host "" is not an admin host
True for all the '-s*' switches I tried.
Is this intentional or desirable?
Personally, I quite like
On Thu, 25 Aug 2016, William Hay wrote:
...
I think this has already been reported as a bug and Dave says he'll have
to redo the change that broke it.
https://arc.liv.ac.uk/trac/SGE/ticket/1579
...
Doh - thanks :)
Mark
___
SGE-discuss mailing list
On Sun, 26 Feb 2017, Glenn Johnson wrote:
It seems this is partly a problem with specifying the 'slots' keyword. If I
specify the number of cores in the binding, as opposed to 'slots', then I
see the binding displayed.
qsub -binding linear:28 ...
Hi Glenn,
We set "-binding set linear" in the
On Tue, 27 Jun 2017, juanesteban.jime...@mdc-berlin.de wrote:
Never mind. One of my users submitted a job with 139k subjobs.
...
Hi,
I don't think I have all the messages from this thread for some reason. No
doubt I'm going to repeat things someone else has suggested - apologies in
advance
On Fri, 8 Mar 2019, Reuti wrote:
...
> We got access to a SLURM equipped cluster where one always get complete
> nodes and are asked to avoid single serial jobs or to pack them by
> scripting to fill the nodes. With the additional need for a workflow
> application (kinda DRMAA) and array job dep
Hi all,
I've hung up my cape - giving up my superpowers on the University of Leeds
supercomputers - and before I went, had a bit of fun looking at our gridengine
accounting logs. Thought I'd share it, in case anyone found it interesting:
https://arc.leeds.ac.uk/very-nearly-almost-14-years-of-a-
11 matches
Mail list logo
