On 10/29/2012 11:33 AM, James Vincent wrote:
Thanks very much. Sorry for the double post - my first was rejected
due to wroneg email, but I guess it made it through.
Do you happen to know if one can query the number of jobs queued and
running? I'd like to put status monitor somewhere for regular users. I
think this will cut down on the number of emails back to me asking why
jobs are not running. :)
use the 'report tool', see: reports_wsgi.ini and run_reports.sh
Hans-Rudolf
Jim
On Mon, Oct 29, 2012 at 4:09 AM, Hans-Rudolf Hotz wrote:
Hi Jim
If using a standalone server, Galaxy is using its own queuing system, which
you can modify by changing (in universe_wsgi.ini):
local_job_queue_workers
user_job_limit
but the system is not aware of the memory size and number of cores. In your
"specific example", running out of memory is probably the bigger issue.
Hope this helps,
Hans-Rudolf
On 10/26/2012 01:41 PM, James Vincent wrote:
Hello,
I cannot seem to find the details on how jobs are managed when Galaxy
is run on a standalone server. Are jobs managed in a queue of some
sort? Does Galaxy determine resources available in some way and
moderate jobs that are run?
As a specific example, we have a 40 core machine with 1TB of memory.
If 10 users each submit 10 jobs each, say for a bowtie job that has
been set to use 8 cores, will Galaxy sort this all out and only run
jobs to fill the 40 cores? Are there settings to control this behavior
without installing an external job manager like SGE?
Thanks,
Jim
___
Please keep all replies on the list by using "reply all"
in your mail client. To manage your subscriptions to this
and other Galaxy lists, please use the interface at:
http://lists.bx.psu.edu/
___
Please keep all replies on the list by using "reply all"
in your mail client. To manage your subscriptions to this
and other Galaxy lists, please use the interface at:
http://lists.bx.psu.edu/
