Re: OSD down after PG increase

[Dan Van Der Ster <daniel.vanderster@xxxxxxx>]

On 13 Mar 2014 at 11:41:30, Gandalf Corvotempesta (gandalf.corvotempesta@xxxxxxxxx) wrote:

2014-03-13 11:32 GMT+01:00 Dan Van Der Ster <daniel.vanderster@xxxxxxx>: 
> Do you have any other pools? Remember that you need to include _all_ pools 
> in the PG calculation, not just a single pool. 

Actually I have only standard pools (that should be 3) 
In production i'll also have RGW. 

So, which is the exact equation to do including X pools ? Should I 
multiply the number of replica with number of pools ? 

(168*100)/(3*4) in case of 4 pools with replica 3 ? 

Yes, that’s would be correct. But only increase the # PGs for pools that you will actually put data into. For example, with RGW probably only need to add PGs to .rgw.buckets.

Cheers, Dan

_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com