On Fri, 1 Apr 2011 10:53:09 +0400 Степан Фёдоров <stepan.fedorov@xxxxxxxx> wrote: > Each target has one real LUN: > > USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND > 0 targets ~1Gb: > root 19675 0.1 9.7 1247120 1198808 ? SLs Mar30 4:51 /usr/sbin/tgtd > > 1 target: > root 19675 0.1 9.7 1247120 1198808 ? SLsl Mar30 4:51 /usr/sbin/tgtd > > 10 targets: > root 19675 0.1 9.7 1542176 1199096 ? SLsl Mar30 4:51 /usr/sbin/tgtd > > 20 targets: > root 19675 0.1 9.7 1870016 1199416 ? SLsl Mar30 4:51 /usr/sbin/tgtd > > 30 targets: > root 19675 0.1 9.7 2197856 1199736 ? SLsl Mar30 4:51 /usr/sbin/tgtd > > 40 targets: > root 19675 0.1 9.7 2525696 1200056 ? SLsl Mar30 4:51 /usr/sbin/tgtd > > 50 targets: > root 19675 0.1 9.7 2853536 1200376 ? SLsl Mar30 4:51 /usr/sbin/tgtd > > 100 targets: > root 19675 0.1 9.7 4492736 1201976 ? SLsl Mar30 4:52 /usr/sbin/tgtd > > 200 targets: > root 19675 0.1 9.7 7771136 1205176 ? SLsl Mar30 4:52 /usr/sbin/tgtd > > 400 targets: > root 19675 0.1 9.8 14327936 1211576 ? SLsl Mar30 4:56 /usr/sbin/tgtd > > This 14327936, is more than memory physically installed on this node > (12Gb), and i don't know what will happen, if tgtd will try to use all > of this memory. If user system has swap memory, tgtd can continue to work, I think. > So, we can see increasing VSZ field by about 30Mb per one LUN. per one target with one lun? Anyway, still need to investigate why we use so much memory. -- To unsubscribe from this list: send the line "unsubscribe stgt" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
