[Users] unable to stop/suspend/kill CT

Kir Kolyshkin kir at virtuozzo.com
Thu Feb 4 11:02:26 PST 2016


Hi Bogdan,

This looks very much like a cpu scheduler lockup, as many of the processes
belonging to the container are in R state but not running.

Can you try resetting the cpulimit for the container in question, 
something like

vzctl set $CTID --cpulimit 0

and see if anything changes?

Also, take a look at cpu.stat for some of the processes that is in such 
state?
Say, this one:
root      107398  0.0  0.0  25460   396 ?        Rs   12:19   0:00 vzctl 
exec 111 ps

cat /proc/vz/fairsched/107398/cpu.stat

If throttled_time is big, it means my hypothesis makes sense.

I am also ccing Vladimir, who knows a thing or two about our fair cpu 
scheduler.

Kir.

On 02/04/2016 05:48 AM, Bogdan-Stefan Rotariu wrote:
> Hi there,
>
> We are having issues with one container that cannot be 
> stopped/suspended or killed, all commands remain in Sleep or Running 
> Sleep.
> Any ideea how to stop this container withour rebooting the main machine?
> We did try to kill all proceeses, they do not die.
>
>       CTID      NPROC STATUS    IP_ADDR         HOSTNAME
>        111        100 running   a.b.c.d server.name
>
>
> [3839648.976835] CPT ERR: ffff8803dd109000,111 :foreign process 
> 14243/9892(bash) inside CT (e.g. vzctl enter or vzctl exec).
> [3839648.976842] CPT ERR: ffff8803dd109000,111 :suspend is impossible 
> now.
> [3839649.977756] CPT ERR: ffff8803dd109000,111 :foreign process 
> 14243/9892(bash) inside CT (e.g. vzctl enter or vzctl exec).
> [3839649.977764] CPT ERR: ffff8803dd109000,111 :suspend is impossible 
> now.
> [3839650.978718] CPT ERR: ffff8803dd109000,111 :foreign process 
> 14243/9892(bash) inside CT (e.g. vzctl enter or vzctl exec).
> [3839650.978726] CPT ERR: ffff8803dd109000,111 :suspend is impossible 
> now.
> [3839665.639557] CPT ERR: ffff880839216000,111 :foreign process 
> 14243/9892(bash) inside CT (e.g. vzctl enter or vzctl exec).
> [3839665.639564] CPT ERR: ffff880839216000,111 :suspend is impossible 
> now.
> [3839666.640019] CPT ERR: ffff880839216000,111 :foreign process 
> 14243/9892(bash) inside CT (e.g. vzctl enter or vzctl exec).
>
> root       19890  0.0  0.0  25460   376 ?        Rs   03:34   0:00 
> /usr/sbin/vzctl exec 111 cat /proc/meminfo
> root       39626  0.0  0.0  25460   376 ?        Rs   03:44   0:00 
> /usr/sbin/vzctl exec 111 cat /proc/meminfo
> root       65503  0.0  0.0  27560   412 ?        Rs   11:59   0:00 
> vzctl enter 111
> root       65508  0.0  0.0  27560   416 ?        Rs   11:59   0:00 
> vzctl enter 111
> root       65522  0.0  0.0  27560   416 ?        Rs   11:59   0:00 
> vzctl enter 111
> root       73329  0.0  0.0  25460   372 ?        Rs   12:00   0:00 
> /usr/sbin/vzctl exec 111 cat /proc/meminfo
> root       73371  0.0  0.0  25460   380 ?        Rs   12:00   0:00 
> /usr/sbin/vzctl exec 111 cat /proc/meminfo
> root       74865  0.0  0.0  25464   408 ?        Rs   12:00   0:00 
> vzctl stop 111
> root       75864  0.0  0.0  25464   412 ?        Rs   12:04   0:00 
> vzctl stop 111
> root       85384  0.0  0.0  25464   404 ?        Rs   12:08   0:00 
> vzctl stop 111
> root       96674  0.0  0.0  25464   412 ?        Rs   12:12   0:00 
> vzctl stop 111
> root       96787  0.0  0.0  25464   408 ?        Rs   12:13   0:00 
> vzctl stop 111 --fast
> root      107300  0.0  0.0  27560   412 ?        Rs   12:18   0:00 
> vzctl enter 111
> root      107398  0.0  0.0  25460   396 ?        Rs   12:19   0:00 
> vzctl exec 111 ps
> root      116638  0.0  0.0 108168  1368 ?        S    12:21   0:00 sh 
> -c /usr/sbin/vzctl exec 111 cat /proc/meminfo | grep --max-count=1 
> 'MemFree' | awk '{print $2}'
> root      116639  0.0  0.0  25460  1024 ?        S    12:21   0:00 
> /usr/sbin/vzctl exec 111 cat /proc/meminfo
> root      116642  0.0  0.0  25460   364 ?        S    12:21   0:00 
> /usr/sbin/vzctl exec 111 cat /proc/meminfo
> root      116643  0.0  0.0  25460   384 ?        Rs   12:21   0:00 
> /usr/sbin/vzctl exec 111 cat /proc/meminfo
> root      116650  0.0  0.0  25460   384 ?        Rs   12:21   0:00 
> /usr/sbin/vzctl exec 111 cat /proc/meminfo
> root      117653  0.0  0.0  25460   380 ?        Rs   12:22   0:00 
> /usr/sbin/vzctl exec 111 cat /proc/meminfo
> root      117746  0.0  0.0 108168  1368 ?        S    12:22   0:00 sh 
> -c /usr/sbin/vzctl exec 111 cat /proc/meminfo | grep --max-count=1 
> 'MemFree' | awk '{print $2}'
> root      117747  0.0  0.0 108168  1368 ?        S    12:22   0:00 sh 
> -c /usr/sbin/vzctl exec 111 cat /proc/meminfo | grep --max-count=1 
> 'MemFree' | awk '{print $2}'
> root      117748  0.0  0.0  25460  1016 ?        S    12:22   0:00 
> /usr/sbin/vzctl exec 111 cat /proc/meminfo
> root      117749  0.0  0.0  25460  1020 ?        S    12:22   0:00 
> /usr/sbin/vzctl exec 111 cat /proc/meminfo
> root      117754  0.0  0.0  25460   360 ?        S    12:22   0:00 
> /usr/sbin/vzctl exec 111 cat /proc/meminfo
> root      117755  0.0  0.0  25460   356 ?        S    12:22   0:00 
> /usr/sbin/vzctl exec 111 cat /proc/meminfo
> root      117756  0.0  0.0  25460   380 ?        Rs   12:22   0:00 
> /usr/sbin/vzctl exec 111 cat /proc/meminfo
> root      117757  0.0  0.0  25460   376 ?        Rs   12:22   0:00 
> /usr/sbin/vzctl exec 111 cat /proc/meminfo
> root      118191  0.0  0.0 108168  1372 ?        S    12:22   0:00 sh 
> -c /usr/sbin/vzctl exec 111 cat /proc/meminfo | grep --max-count=1 
> 'MemTotal' | awk '{print $2}'
> root      118192  0.0  0.0  25460  1020 ?        S    12:22   0:00 
> /usr/sbin/vzctl exec 111 cat /proc/meminfo
> root      118195  0.0  0.0  25460   360 ?        S    12:22   0:00 
> /usr/sbin/vzctl exec 111 cat /proc/meminfo
> root      118196  0.0  0.0  25460   380 ?        Rs   12:22   0:00 
> /usr/sbin/vzctl exec 111 cat /proc/meminfo
> root      126585  0.0  0.0  25464   408 ?        Rs   12:25   0:00 
> vzctl stop 111
> root      129412  0.0  0.0  25464   352 ?        Rs   12:26   0:00 
> vzctl stop 111
> root      138146  0.0  0.0  25464   404 ?        Rs   12:28   0:00 
> vzctl stop 111
> root      147844  0.0  0.0  25464   408 ?        Rs   12:33   0:00 
> vzctl stop 111
> root      157178  0.0  0.0  25464   412 ?        Rs   12:36   0:00 
> vzctl stop 111
> root      158300  0.0  0.0  25464   400 ?        Rs   12:39   0:00 
> vzctl stop 111
> root      179962  0.0  0.0  25464   408 ?        Rs   12:49   0:00 
> vzctl stop 111
> root      180039  0.0  0.0  25464   408 ?        Rs   12:49   0:00 
> vzctl stop 111
> root      220918  0.0  0.0  25464   412 ?        Rs   13:04   0:00 
> vzctl stop 111
> root      240631  0.0  0.0  25464   408 ?        Rs   13:14   0:00 
> vzctl stop 111
> root      247169  0.0  0.0  25464   412 ?        Rs   13:15   0:00 
> vzctl stop 111
> root      250371  0.0  0.0  25464   400 ?        Rs   13:19   0:00 
> vzctl stop 111 --fast
>
>
> _______________________________________________
> Users mailing list
> Users at openvz.org
> https://lists.openvz.org/mailman/listinfo/users



More information about the Users mailing list