[GE users] Qmaster is running out of memory

introx introx at gmail.com
Wed Feb 24 15:44:58 GMT 2010


    [ The following text is in the "iso-8859-1" character set. ]
    [ Your display is set for the "ISO-8859-10" character set.  ]
    [ Some special characters may be displayed incorrectly. ]

Hi,

When I try to submit 1000 jobs (one after the other using a submit script) the Qmaster daemon consuming all the phisical memory on the Master hosts (12 GB) and then start to swap memory. after a while it crashes...
Since I am not the one who has installed the system (and I am also a sun grid newbee....)I can't tell the configuration of the installation.

Seems to me that there is something wrong with the installation/configuration since 1000 jobs shouldn't be a problem for the SGE to handle... at least not for a 12GB Linux with 2 cpu's.
The problem is that I don't really know how to debug it.

We are using sge 6.2_u2 version.
I would highly appreciate any help since I am quite in the dark here :-)

Thanks

Erez



More information about the gridengine-users mailing list