[one-users] Automatic recover from hardware failure

[Johan Kooijman]

Sander,

See
https://github.com/dignus/one/commit/02a5a65eae48d58edaa414af94802bc74e5b1e24
for the fix you can apply yourself before 4.8.1 is released.


On Thu, Aug 14, 2014 at 8:36 PM, Sander Klein <roedie at roedie.nl> wrote:

> Hi,
>
>
> On 14.08.2014 16:20, Carlos Martín Sánchez wrote:
>
>> Hi,
>>
>> On Thu, Aug 14, 2014 at 1:07 PM, Johan Kooijman
>> <mail at johankooijman.com> wrote:
>>
>>  Hi all,
>>>
>>> I'm almost finished setting up the POC for ONE so far. One last
>>> piece I can't really figure out is how to recreate VM's on a
>>> different host in case a host fails, like hardware off the grid.
>>> How would I do such a thing?
>>>
>>
>> You can enable the fault tolerance hook in oned.conf [1]. This hook
>> will perform the onevm delete --recreate action on the VMs of the
>> failed host. To avoid false positives because of network connectivity
>> issues, use the -p flag of the hook.
>>
>> The delete-recreate action will create a new VM from the original
>> template. This means that the VM will have a new ID, IP, and clean
>> disks.
>> We are working on a new mechanism for hosts with shared storage, to
>> "migrate" the failed VM to a new host keeping the current IPs, disk
>> state, etc.
>>
>
> Is there an ETA on this new mechanism? I just ran into the same problem
> today in my test setup.
>
> Greets,
>
> Sander
>



-- 
Met vriendelijke groeten / With kind regards,
Johan Kooijman
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opennebula.org/pipermail/users-opennebula.org/attachments/20140822/af7c9e26/attachment.htm>