Re: [one-users] VMs still reported as running although host is down

[Carlos Martín Sánchez]

Hi Steffen,

There are two reasons I can think of that can cause that behaviour:

1) The next monitoring cycle hasn't been triggered.
2) The monitorization was triggered, but the command crashed and it didn't
return neither FAILURE or SUCCESS to the core.

I've made a couple of tests and found that, indeed, there is a bug :/
You can its ticket here [1].

The problem is that the core sends the monitoring request to the ssh driver,
and it is supposed to return FAILURE or SUCCESS. Instead, the driver is
crashing, and the Host and VM keep the last monitoring data (and status).

This could also cause trouble if the motorization probe gets stuck, and
doesn't end its execution.
We are aware of this problem, and we intend to add timeouts and checks to
the drivers to ensure that the actions are always answered to the core.

Thanks for the feedback!
Carlos.

[1] http://dev.opennebula.org/issues/630

--
Carlos Martín, MSc
Project Major Contributor
OpenNebula - The Open Source Toolkit for Cloud Computing
www.OpenNebula.org  | cmar...@opennebula.org


On Wed, May 11, 2011 at 5:10 PM, Steffen Neumann wrote:

> Hi,
>
> I shut down some hosts (had to, there were NFS issues)
> and didn't delete the VMs beforehand.
>
> I am a bit surprised that onehost list still shows
> the node as "on":
> node02 default2   1600   1280   1200 63G   58.1G   on
>
> and that onevm list also reports the VMs as running:
> 386 oneadmin  suse113 runn  22  8G  node02 08 06:51:45
>
> 1) Why doesn't onehost report "err" if it can't execute
>   the probes on the shutdown node ?
>
> 2) Why doesn't onevm report "unknown" or whatever if it
>   can't reach the not-running VM on the not-running node ?
>
> Should I open (one or two ?) tickets ?
>
> Yours,
> Steffen
>
>
> --
> IPB HalleAG Massenspektrometrie & Bioinformatik
> Dr. Steffen Neumann  http://www.IPB-Halle.DE
> Weinberg 3   http://msbi.bic-gh.de
> 06120 Halle  Tel. +49 (0) 345 5582 - 1470
>  +49 (0) 345 5582 - 0
> sneumann(at)IPB-Halle.DE Fax. +49 (0) 345 5582 - 1409
>
>
> ___
> Users mailing list
> Users@lists.opennebula.org
> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>
___
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org

[one-users] VMs still reported as running although host is down

[Steffen Neumann]

Hi,

I shut down some hosts (had to, there were NFS issues)
and didn't delete the VMs beforehand. 

I am a bit surprised that onehost list still shows
the node as "on":
node02 default2   1600   1280   1200 63G   58.1G   on

and that onevm list also reports the VMs as running:
386 oneadmin  suse113 runn  22  8G  node02 08 06:51:45

1) Why doesn't onehost report "err" if it can't execute 
   the probes on the shutdown node ?

2) Why doesn't onevm report "unknown" or whatever if it 
   can't reach the not-running VM on the not-running node ?

Should I open (one or two ?) tickets ? 

Yours,
Steffen


-- 
IPB HalleAG Massenspektrometrie & Bioinformatik
Dr. Steffen Neumann  http://www.IPB-Halle.DE
Weinberg 3   http://msbi.bic-gh.de
06120 Halle  Tel. +49 (0) 345 5582 - 1470
  +49 (0) 345 5582 - 0
sneumann(at)IPB-Halle.DE Fax. +49 (0) 345 5582 - 1409


___
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org