[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Condor-users] nonsensical ImageSize values

Checking the health of my pool, I sometimes look for excessive ImageSizes,
and this time I got

# condor_status -f "%s " 'substr(Name,0,10)' -f "%s " 'substr(RemoteOwner,0,2)' -f "%s " 'Cmd' -f "%gM " 'ImageSize/1024' -f "\n" test | sort -rn -k+3 | head -n10
slot1@o319 be 11336M
slot1@o389 be 11245M
slot2@o246 be 11201M
slot1@o444 be 11178M
slot1@o055 be 11098M
slot2@o615 be 571M
slot2@o611 be 571M
slot2@o609 be 571M
slot2@o608 be 571M
slot2@o607 be 571M

These slots are about 1GB in size, and the total swap space for two of
them is 2GB, so anything beyond 4GB is clearly nonsensical.
Since I intend to use ImageSize to preempt users' jobs which are exceeding
their memory limits, I want to make sure that jobs don't get removed due
to wrong figures. 
What's the detail I have overlooked???


Steffen Grunewald * MPI Grav.Phys.(AEI) * Am Mühlenberg 1, D-14476 Potsdam
Cluster Admin * http://pandora.aei.mpg.de/merlin/ * http://www.aei.mpg.de/
* e-mail: steffen.grunewald(*)aei.mpg.de * +49-331-567-{fon:7233,fax:7298}
No Word/PPT mails - http://www.gnu.org/philosophy/no-word-attachments.html