[Linux-cluster] rgmanager vm.sh using virsh under RHEL5.4

[Alain RICHARD]

Le 3 nov. 2009 à 18:56, Lon Hohberger a écrit :

> Federico kicked be in to finishing this... Want to give it a shot?
> Effectively, it makes 'path' work with virsh - making virsh mode  
> upgrade
> compatible with 'xm' mode.  This means that we can really (actually)
> deprecate 'xm' support.
>
> Pay no attention to the fact that it uses a global var called  
> 'xmlfile';
> it doesn't do any config file parsing, so it still works with Xen and
> libvirt (xml) style config files:
>
> http://git.fedorahosted.org/git/?p=resource-agents.git;a=commit;h=1e42add4f2ec63d4c796d544118dfcccbd8a042b
>
> It just splits up a OCF_RESKEY_path on ':' and looks for a file in  
> each
> directory named NAME or NAME.xml.  The requirement is obviously that
> your config files need to be named after your VM.  If this is a  
> problem,
> you must use the full 'xmlfile' option.
>
> -- Lon
>


I am currently testing this version of vm.sh that is handling xmlfile  
and path differently :

- if use_virth=0, use xm as before
- if path and xmlfile, ignore path and issue a warning (you should use  
xmlfile)
- if xmlfile only, use it
- if path, search for a file name under path or name.xml under path  
and set xmfile to this file

in the case of virsh, creation is handled using 'virsh create xmlfile'  
if xmlfile is not empty, or 'virsh create name' if there is no xmlfile/ 
path configured.

The effect of this is that the config file must be :

- an xml file, with or without .xml extension, if xmlfile or path  
attribute is set
- else, a classic xen config file under /etc/xen

In order to stay compatible with current rgmanager configuration, we  
must ensure that use_virsh is set to 0 for vm that use classical xen  
conf files and path directive, else the vm fails to lauch because  
virsh create is not able to handle xen config file and virsh start,  
that is able to handle xen conf files, is not able to get the file  
from an other location than /etc/xen.

An other point is that if libvirtd is not running, the status returned  
by this vm.sh for a vm is always "indeterminate", so I have to launch  
it and to disable the default libvirt network because I really don't  
need it.

The last problem so far is that clusvcadm -M always end-up with an  
error although the migration is working correctly :

#clusvcadm -M vm:ns2 -m titan2
Trying to migrate vm:ns2 to titan2...Failure

# clustat
Cluster Status for titan-cluster @ Thu Nov  5 10:07:24 2009
Member Status: Quorate

  Member Name                                                     ID    
Status
  ------ ----                                                     ----  
------
  titan1                                                               
1 Online, rgmanager
  titan2                                                               
2 Online, Local, rgmanager
  qdisk1                                                               
0 Online, Quorum Disk

  Service Name                                                      
Owner (Last)                                                     State
  ------- ----                                                      
----- ------                                                     -----
  vm:ns2                                                            
titan2                                                           started

Looking at rgmanager log, I found this :

Nov  5 10:05:19 titan1 clurgmgrd[11231]: <notice> Migrating vm:ns2 to  
titan2
Nov  5 10:05:19 titan1 clurgmgrd: [11231]: <debug> Using /data/vmconf/ 
ns2.xml
Nov  5 10:05:19 titan1 clurgmgrd: [11231]: <debug> virsh migrate -- 
live ns2 xen:/// xenmigr://titan2/
Nov  5 10:05:27 titan1 clurgmgrd: [11231]: <err> Migrate ns2 to titan2  
failed:
Nov  5 10:05:27 titan1 clurgmgrd: [11231]: <err> error: Domain not  
found: xenUnifiedDomainLookupByName
Nov  5 10:05:27 titan1 clurgmgrd[11231]: <notice> migrate on vm "ns2"  
returned 7 (unspecified)
Nov  5 10:05:27 titan1 clurgmgrd[11231]: <err> Migration of vm:ns2 to  
titan2 failed; return code 7

Nov  5 10:05:19 titan2 clurgmgrd[7484]: <debug> FW: Forwarding migrate  
request to 1
Nov  5 10:05:24 titan2 clurgmgrd: [7484]: <debug> Using /data/vmconf/ 
ns2.xml
Nov  5 10:05:25 titan2 clurgmgrd[7484]: <notice> vm:ns2 is now running  
locally
Nov  5 10:05:34 titan2 clurgmgrd: [7484]: <debug> Using /data/vmconf/ 
ns2.xml

This error is generated by libvirtd :

# virsh migrate --live ns2 xen:/// xenmigr://titan1/
error: Domain not found: xenUnifiedDomainLookupByName

# echo $?
1

Is it normal ? My initial goal to use virsh instead of xm was for  
better error handling during migrations...

Regards,

-- 
Alain RICHARD <mailto:alain.richard at equation.fr>
EQUATION SA <http://www.equation.fr/>
Tel : +33 477 79 48 00     Fax : +33 477 79 48 01
E-Liance, Opérateur des entreprises et collectivités,
Liaisons Fibre optique, SDSL et ADSL <http://www.e-liance.fr>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/linux-cluster/attachments/20091105/34c71645/attachment.htm>