[rhos-list] unable to delete an instance

Vogel Nicolas nicolas.vogel at heig-vd.ch
Fri Mar 22 09:29:14 UTC 2013


Here are some logs which could help:

Compute.log :
2013-03-22 09:32:08 2343 WARNING nova.compute.manager [-] Found 1 in the database and 0 on the hypervisor.
2013-03-22 09:32:08 2343 INFO nova.compute.manager [-] [instance: 273c7a2b-a27c-42ef-9cdb-dbe5d592b194] During sync_power_state the instance has a pending task. Skip.
...
2013-03-22 09:34:19 2343 INFO nova.compute.manager [-] Updating host status
2013-03-22 09:34:55 AUDIT nova.compute.manager [req-8c181208-125d-487b-8933-edf44c8088bf fdfcf45f412a46e4880e6d8bd8143b61 bc308f566c2b46aea7812a033de1f21c] [instance: 273c7a2b-a27c-42ef-9cdb-dbe5d592b194] Rebooting instance
2013-03-22 09:35:00 2343 ERROR nova.virt.libvirt.driver [-] [instance: 273c7a2b-a27c-42ef-9cdb-dbe5d592b194] During wait destroy, instance disappeared.
2013-03-22 09:35:00 INFO nova.virt.libvirt.firewall [req-8c181208-125d-487b-8933-edf44c8088bf fdfcf45f412a46e4880e6d8bd8143b61 bc308f566c2b46aea7812a033de1f21c] [instance: 273c7a2b-a27c-42ef-9cdb-dbe5d592b194] Called setup_basic_filtering in nwfilter
2013-03-22 09:35:00 INFO nova.virt.libvirt.firewall [req-8c181208-125d-487b-8933-edf44c8088bf fdfcf45f412a46e4880e6d8bd8143b61 bc308f566c2b46aea7812a033de1f21c] [instance: 273c7a2b-a27c-42ef-9cdb-dbe5d592b194] Ensuring static filters
2013-03-22 09:35:04 ERROR nova.compute.manager [req-8c181208-125d-487b-8933-edf44c8088bf fdfcf45f412a46e4880e6d8bd8143b61 bc308f566c2b46aea7812a033de1f21c] [instance: 273c7a2b-a27c-42ef-9cdb-dbe5d592b194] Cannot reboot instance: Unable to pre-create chardev file '/var/lib/nova/instances/instance-0000001f/console.log': Aucun fichier ou dossier de ce type

I also have the ERROR "nova [-] need more than 0 values to unpack" or "cinder [-] need more than 0 values to unpack".

And the cinder service is unavailable:
[root at IICT-SV1259 ~(keystone_admin)]$ cinder list
ERROR: n/a (HTTP 400)

Is that a problem from keystone?

-----Original Message-----
From: Russell Bryant [mailto:rbryant at redhat.com] 
Sent: jeudi 21 mars 2013 17:45
To: Vogel Nicolas
Cc: 'rhos-list at redhat.com'
Subject: Re: [rhos-list] unable to delete an instance

On 03/21/2013 12:00 PM, Vogel Nicolas wrote:
> Hi,
> 
>  
> 
> I had an instance who couldn’t terminate properly so I used the 
> command “nova reset-state –active <id>” to modify her status to 
> Active. It works but after that, using nova delete <id>, my instance 
> is still standing in the “Deleting” Task.
> 
> Can I delete this instance direct in the database? What is the way to 
> manually terminate an instance?

We need to go back to figuring out why the instance couldn't be deleted originally.  Do you know how you got it into that state?

If all else fails, the brute force is manually removing the instance from the database and the hypervisor, but you should never have to do that.

> This instance had a volume attached to her so I think I will have to 
> manually detach and delete this volume too. But command nova 
> volume-list is giving me “ERROR: ConnectionRefused: '[Errno 111] Connection refused'”.

Do other nova commands give you that error?  Can you provide the full command and output?

--
Russell Bryant




More information about the rhos-list mailing list