On Sat, May 05, 2012 at 09:44:35AM +0200, Juerg Haefliger wrote: > >> But better would be to use the device number of the relevant > >> filesystem to filter the events. The device is 252:2, which means in > >> kernel terms is it: > >> > >> dev = (major << 20) | minor > >> = 0xfc00002 > >> > >> So you should be able to get just the xfs-hang events via: > >> > >> # trace-cmd record -e xfs\* -d 'dev == 0xfc00002' > >> > >> and as long as you don't host log files on /xfs-hang, it'll only > >> record the workload running on the xfs-hang filesystem. > > > > Third try: https://region-a.geo-1.objects.hpcloudsvc.com:443/v1.0/AUTH_9630ead2-6194-40df-afd3-7395448d4536/xfs-hang/report-2012-04-27-180secs.tgz > > Filtered by device, trace events go to a different filesystem. > > Did anybody have a chance to look at the data? I've had a quick look, but I need to write scripts to visualise it (i.e. graph it) to determine if there's any pattern to the issue. Cheers, Dave. -- Dave Chinner david@xxxxxxxxxxxxx _______________________________________________ xfs mailing list xfs@xxxxxxxxxxx http://oss.sgi.com/mailman/listinfo/xfs
