On 11/26/2014 04:05 AM, Yujian Peng wrote:
Hi all, I have a ceph cluster in production. Most of the write requests are small. I found that iops is a bottleneck. I want to move all of the journal datas to partitions on SSDs. Here is the procedures: 1.Set noout flag. ceph osd set noout 1.Stop osd 0 2.Copy the journal datas to new partitions. cp /dev/sda1 /dev/sdb1 3.Delete the journal link. rm /var/lib/ceph/osd/ceph-0/journal 4.Create journal link. ln -s /dev/sdb1 /var/lib/ceph/osd/ceph-0/journal 5.Start osd 0 6.Do step 1 to step 5 for all osds 7.Unset noout flag. ceph osd unset noout Is this Ok?
I don't know if things have changed, but I don't think you want to outright move the journal like that. Instead, something like:
ceph-osd -i N --flush-journal <delete old journal> ceph-osd -N --mkjournal
Since the size of jornal partitions on SSDs is 10G, I want to set filestore max sync interval to 30 minutes. Is 30 minutes resonable? How to set filestore max sync interval on the fly?
The effect of this is that you'll be able to write the journal uninterrupted for a long time, but syncs could be very expensive. I'd suggest giving it a try at a couple of values (default, 60s, 300s) under say 20 minutes of continuous heavy workload to observe the effects and see how you like it.
Any help, is highly appreciated. Thanks! _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
