Hi all,

I did a Live Migration today using two hardware nodes that were running the
OpenVZ Kernel 2.6.18-028stab035. Before live migrating, I started a screen
session inside the VE and in the screen sessions' shell I launched the
following command:

      while [ true ]; do date; sleep 1; done

After the migration I entered the VE and the screen session but the while
loop wasn't running anymore. It wasn't printing the date every second as
expected. When I do chechkpointing and restoring on the same machine the
loop is running fine after restoring.

Is there someone experiencing  the same problem?

Mit freundlichen Grüßen / Best Regards
Christian Kaiser
--
IBM Deutschland Entwicklung GmbH
Open Systems Firmware Development
mail: ckaiser2@de.ibm.com

IBM Deutschland Entwicklung GmbH
Vorsitzender des Aufsichtsrats: Martin Jetter
Geschäftsführung: Herbert Kircher
Sitz der Gesellschaft: Böblingen
Registergericht: Amtsgericht Stuttgart, HRB 243294

Christian,

Most likely you have too much unsychronized time on these machines.
Please synchronize it with NTP or by hands and retry.

Kirill

Christian Kaiser2 wrote:
> Hi all,
> 
> I did a Live Migration today using two hardware nodes that were running the
> OpenVZ Kernel 2.6.18-028stab035. Before live migrating, I started a screen
> session inside the VE and in the screen sessions' shell I launched the
> following command:
> 
>       while [ true ]; do date; sleep 1; done
> 
> After the migration I entered the VE and the screen session but the while
> loop wasn't running anymore. It wasn't printing the date every second as
> expected. When I do chechkpointing and restoring on the same machine the
> loop is running fine after restoring.
> 
> Is there someone experiencing  the same problem?
> 
> Mit freundlichen Grüßen / Best Regards
> Christian Kaiser
> --
> IBM Deutschland Entwicklung GmbH
> Open Systems Firmware Development
> mail: ckaiser2@de.ibm.com
> 
> IBM Deutschland Entwicklung GmbH
> Vorsitzender des Aufsichtsrats: Martin Jetter
> Geschäftsführung: Herbert Kircher
> Sitz der Gesellschaft: Böblingen
> Registergericht: Amtsgericht Stuttgart, HRB 243294
> 
>

You're right!
Different time zones and an unsynchronized clock must have caused the
error. I think the "sleep 1" caused the error because the SIG_ALARM was on
the other host not after one second but after a few hours ;-)

Mit freundlichen Grüßen / Best Regards
Christian Kaiser
--
IBM Deutschland Entwicklung GmbH
Open Systems Firmware Development
mail: ckaiser2@de.ibm.com

IBM Deutschland Entwicklung GmbH
Vorsitzender des Aufsichtsrats: Martin Jetter
Geschäftsführung: Herbert Kircher
Sitz der Gesellschaft: Böblingen
Registergericht: Amtsgericht Stuttgart, HRB 243294

Kirill Korotaev <dev@sw.ru> wrote on 09/18/2007 10:15:11 AM:

> Christian,
>
> Most likely you have too much unsychronized time on these machines.
> Please synchronize it with NTP or by hands and retry.
>
> Kirill
>
> Christian Kaiser2 wrote:
> > Hi all,
> >
> > I did a Live Migration today using two hardware nodes that were running
the
> > OpenVZ Kernel 2.6.18-028stab035. Before live migrating, I started a
screen
> > session inside the VE and in the screen sessions' shell I launched the
> > following command:
> >
> >       while [ true ]; do date; sleep 1; done
> >
> > After the migration I entered the VE and the screen session but the
while
> > loop wasn't running anymore. It wasn't printing the date every second
as
> > expected. When I do chechkpointing and restoring on the same machine
the
> > loop is running fine after restoring.
> >
> > Is there someone experiencing  the same problem?
> >
> > Mit freundlichen Grüßen / Best Regards
> > Christian Kaiser
> > --
> > IBM Deutschland Entwicklung GmbH
> > Open Systems Firmware Development
> > mail: ckaiser2@de.ibm.com
> >
> > IBM Deutschland Entwicklung GmbH
> > Vorsitzender des Aufsichtsrats: Martin Jetter
> > Geschäftsführung: Herbert Kircher
> > Sitz der Gesellschaft: Böblingen
> > Registergericht: Amtsgericht Stuttgart, HRB 243294
> >
> >