[ale] lsof and a hung system

Jim Kinney jim.kinney at gmail.com
Mon Oct 19 22:58:10 EDT 2015


On Oct 19, 2015 6:33 PM, "DJ-Pfulio" <DJPfulio at jdpfu.com> wrote:
>
> On 10/19/2015 06:14 PM, Jim Kinney wrote:
> > So the user notifies me they can do a cd to an nfs mounted directory. I
> > get on and can't do ls -la on /. there's a trio of cat commands that
> > are hung and anything that reads data from the drives basically does
> > nothing. Nothing is running but the load average is 20 and not
> > changing. So I try to kill the cat's and unmount the nfs folder.  No
> > work. Dang! what else is open? run lsof. Nothing happens.
> > I'm expecting all sorts of file system errors when after I press the
> > reset switch. A sync command just hangs. dmesg shows user space
> > applications segfaults all over the the place. So the (very new) user
> > would just run it again. I've got a zombie collection that The Walking
> > Dead would call a "herd".
> >
>
> sync?
Yep
> hard?
Yep
> nfs v3/v4?
4
> mount options?
Just read/write sizes

Other system with same nfs mounted storage is fine. Storage server is
connected to both number crunchers by dedicated, unswitched 10Gbps fiber
ethernet.
>
>
>
> Zombies?!!?
A total of 10 zombie processes. All were running code from and data writes
to the nfs mounted space.
>
> *
>
http://abcnews.go.com/Technology/zombie-bees-found-northeast/story?id=22290433
>
> * http://www.cdc.gov/phpr/zombies.htm - gotta be prepared
>
> _______________________________________________
> Ale mailing list
> Ale at ale.org
> http://mail.ale.org/mailman/listinfo/ale
> See JOBS, ANNOUNCE and SCHOOLS lists at
> http://mail.ale.org/mailman/listinfo
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ale.org/pipermail/ale/attachments/20151019/0cb53362/attachment.html>


More information about the Ale mailing list