[Linux-cluster] gfs2_quotad:2498 blocked
ajb2 at mssl.ucl.ac.uk
Fri Mar 25 17:48:56 UTC 2011
Nicolas Ross wrote:
> It was a large, very large directory, with somewhere neer one million
> small files, so the rsync took something like 3 to 4 hours. At some
> point, all nodes' consoles dispalyed this :
> gfs2_quotad:2498 blocked for more that 120 seconds.
> "echo 0 > /proc/sys/kernel/hang_task_timeout_secs" disables this message.
> and then some debuging info dump.
> What does this mean ?
It means you're writing data faster than GFS can handle it, so it's
pausing on you. We see this semi-regularly here on our clusters.
I suspect Bob Petersen's test kernel or patches from ~2-3 weeks ago will
help you as these contain a bunch of fixes for slow writes.
More information about the Linux-cluster