Re: Migrate OSDs to different backend

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



I think option 2 should be reliable.

On Wed, Jul 29, 2015 at 9:00 PM, Kenneth Waegeman
<kenneth.waegeman@xxxxxxxx> wrote:
> Hi all,
>
> We are considering to migrate all our OSDs of our EC pool from KeyValue to
> Filestore. Does someone has experience with this? What would be a good
> procedure?
>
> We have Erasure Code using k+m: 10+3, with host-level failure domain on 14
> servers. Our pool is 30% filled.
>
> I was thinking:
> We set the weight of 1/2 of the OSDS on each host to 0 and let the cluster
> migrate the data
> We then remove these OSDS, and re-add them
> We do then the same with the other OSDS
> (Or we do it in 3 times with 1/3 of the OSDS)
>
> Another option:
> We repeatedly;
> Remove all KV OSDS of 2 servers (m=3)
> Re-add all those OSDS with filestore
> Wait for data to rebalance
>
> Does someone know what would be the best way? Are there things we should not
> forget or be careful with ?
>
> Thank you very much!
>
> Kenneth
> _______________________________________________
> ceph-users mailing list
> ceph-users@xxxxxxxxxxxxxx
> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com



-- 
Best Regards,

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



[Index of Archives]     [Information on CEPH]     [Linux Filesystem Development]     [Ceph Development]     [Ceph Large]     [Ceph Dev]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [xfs]


  Powered by Linux