[PVE-User] Ceph: PANIC or DON'T PANIC? ;-)

[Alwin Antreich]

Hi Marco,

On 11/29/2016 03:05 PM, Marco Gaiarin wrote:
> Mandi! Alwin Antreich
>   In chel di` si favelave...
> 
>> What does the following command give you?
>> ceph osd pool get <POOLNAME> min_size
> 
> root at capitanamerica:~# ceph osd pool get DATA min_size
> min_size: 1
> root at capitanamerica:~# ceph osd pool get VM min_size
> min_size: 1
> root at capitanamerica:~# ceph osd pool get LXC min_size
> min_size: 1

The min_size 1 means in a degraded state, ceph serves the data as long as one copy is available.

> 
> 
>> As a general thought, a HA cluster would be always running, so the event that you shutdown all nodes is unlikely to
>> happen.
> 
> Ahem, not so unlikely... we have UPSes but not diesel generators... ;-(

If they shutdown cleanly, then it shouldn't be a problem, as far as I have tested it myself.

> 
> 
>> And if you decide to shutdown all nodes, then a couple of minutes should be ok to get everything running again
>> and ceph needs its time, till all services are running.
> 
> This is not the case. I've started the nodes one by one...
> 

I don't see this behavior on our test cluster, when we shutdown all hosts and start them up at a later time.

-- 
Cheers,
Alwin