On 10/08/2015 04:28 PM, wd_hw...@wistron.com wrote: > Hi Wido: > According to your reply, if I add/remove OSDs from Ceph cluster, I have to > wait all PGs moving action are completed. > Then 'rbd ls' operation may works well. > Is there any way to speed up PGs action of adding/removing OSDs ? >
If completely depends on your cluster. Usually more CPU power improves the peering performance, but it's not a guarantee. Usually a peering operation shouldn't take very long. Wido > Thanks a lot. > > Best Regards, > WD > > -----Original Message----- > From: ceph-users [mailto:ceph-users-boun...@lists.ceph.com] On Behalf Of Wido > den Hollander > Sent: Thursday, October 08, 2015 10:06 PM > To: ceph-users@lists.ceph.com > Subject: Re: [ceph-users] How to improve 'rbd ls [pool]' response time > > On 10/08/2015 10:46 AM, wd_hw...@wistron.com wrote: >> Hi, all: >> If the Ceph cluster health status is HEALTH_OK, the execution time of >> 'sudo rbd ls rbd' is very short, like the following results. >> $ time sudo rbd ls rbd >> real 0m0.096s >> user 0m0.014s >> sys 0m0.028s >> >> But if there are several warnings (eg: 1 pgs degraded; 6 pgs incomplete; >> 1650 pgs peering; 7 pgs stale;), the execution time of 'sudo rbd ls rbd' may >> take a long time. >> >> Is there any way to improve the response time of 'rbd' commands? > > If you have incomplete PGs those can't serve any I/O. Same goes for PGs which > are in peering state. > > That causes all RADOS operations to block. You'd have to resolve that > situation and that will make 'rbd ls' snappy again. > >> Any help would be much appreciated. >> >> Best Regards, >> WD Hwang >> >> ---------------------------------------------------------------------- >> ---------------------------------------------------------------------- >> ------------------- This email contains confidential or legally >> privileged information and is for the sole use of its intended recipient. >> Any unauthorized review, use, copying or distribution of this email or the >> content of this email is strictly prohibited. >> If you are not the intended recipient, you may reply to the sender and >> should delete this e-mail immediately. >> ---------------------------------------------------------------------- >> ---------------------------------------------------------------------- >> ------------------- >> >> >> >> _______________________________________________ >> ceph-users mailing list >> ceph-users@lists.ceph.com >> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com >> > > > -- > Wido den Hollander > 42on B.V. > Ceph trainer and consultant > > Phone: +31 (0)20 700 9902 > Skype: contact42on > _______________________________________________ > ceph-users mailing list > ceph-users@lists.ceph.com > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > _______________________________________________ > ceph-users mailing list > ceph-users@lists.ceph.com > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > -- Wido den Hollander 42on B.V. Ceph trainer and consultant Phone: +31 (0)20 700 9902 Skype: contact42on _______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
