[CRIU] Dumping lxc container failed - CRIU 1.6

[Pavel Emelyanov]

On 06/25/2015 04:29 PM, Thouraya TH wrote:
> 's' means stop.
> 
> I have used this option because without it, the restart of a container from a snapshot didn't work.


Of course, but since your dump fails, the container should remain running. So can you
check whether dump fails any time you try it, or just sometimes?

> Test:
> lxc-checkpoint -D /tmp/dire -n worker1  (without -s option)
> lxc-stop -n worker1 ( i have stopped the container to test if a container restarts from a snapshot if i have a local crash of the container)
> lxc-checkpoint -r -D  /tmp/dire -n worker1
> 
> lxc-ls -f
> ^CTraceback (most recent call last):
>   File "/usr/bin/lxc-ls", line 432, in <module>
>     containers = get_containers(root=True)
>   File "/usr/bin/lxc-ls", line 261, in get_containers
>     if container.controllable:
> KeyboardInterrupt
> 
> bloqued.
> 
> restore.log:
> 
> Warn  (cr-restore.c:1029): Set CLONE_PARENT | CLONE_NEWPID but it might cause restore problem,because not all kernels support such clone flags combinations!
> RTNETLINK answers: File exists
> RTNETLINK answers: File exists
> RTNETLINK answers: File exists
>    610: Error (files-reg.c:1097): Can't open file tmp/libnetty-transport-native-epoll4589659792274548507.so on restore: No such file or directory
>    610: Error (files-reg.c:1040): Can't open file tmp/libnetty-transport-native-epoll4589659792274548507.so: No such file or directory
>    610: Error (cr-restore.c:253): Can't fixup VMA's fd
> 
> 
> 
> 
> 
> Bests.
> 
> 
> 2015-06-25 13:27 GMT+01:00 Pavel Emelyanov <xemul at parallels.com <mailto:xemul at parallels.com>>:
> 
>     On 06/25/2015 03:14 PM, Thouraya TH wrote:
>     > No, i have used -s option so it is not RUNNING.
> 
>     What does this option mean for lxc? And what if you don't use one
>     and just lxc-checkpoint the container instead?
> 
>     > Kind regards.
>     >
>     > 2015-06-25 12:01 GMT+01:00 Pavel Emelyanov <xemul at parallels.com <mailto:xemul at parallels.com> <mailto:xemul at parallels.com <mailto:xemul at parallels.com>>>:
>     >
>     >     On 06/25/2015 01:15 PM, Thouraya TH wrote:
>     >     > Hi all,
>     >     >
>     >     > May be, cloning containeris the source of the problem...
>     >     >
>     >     > /Does it happen always? What if you try to dump this container again without restart?
>     >     > /
>     >     > Did you mean:
>     >     >
>     >     > lxc-checkpoint -s -D /tmp/dire -n worker1
>     >     > lxc-start -n worker1  ( i have to restart all process needed inside the container)
>     >     > lxc-checkpoint -s -D /tmp/dire1 -n worker1
>     >
>     >     Yes, but without the lxc-start in between (the container remains running after
>     >     first dump, doesn't it?)
>     >
>     >
> 
>