how about also increasing osd_recovery_threads?
On Wed, Sep 4, 2019 at 10:47 AM Guilherme Geronimo < [email protected]> wrote: > Hey hey, > > First of all: 10GBps connection. > > Then, some magic commands: > > # ceph tell 'osd.*' injectargs '--osd-max-backfills 32' > # ceph tell 'osd.*' injectargs '--osd-recovery-max-active 12' > # ceph tell 'osd.*' injectargs '--osd-recovery-op-priority 63' > > =D > > []'s > Arthur (aKa Guilherme Geronimo) > > On 04/09/2019 06:44, Amudhan P wrote: > > Hi, > > I am using ceph version 13.2.6 (mimic) on test setup trying with cephfs. > my ceph health status showing warning. > > My current setup: > 3 OSD node each with a single disk, recently I added one more disk in one > of the node and ceph cluster status showing warning. > I can see the progress but it was more than 12 hours but still its moving > objects. > > How to increase the speed of moving objects? > > output from "ceph -s" > > cluster: > id: 7c138e13-7b98-4309-b591-d4091a1742b4 > health: HEALTH_WARN > 834820/7943361 objects misplaced (10.510%) > > services: > mon: 1 daemons, quorum mon01 > mgr: mon01(active) > mds: cephfs-tst-1/1/1 up {0=mon01=up:active} > osd: 4 osds: 4 up, 4 in; 12 remapped pgs > > data: > pools: 2 pools, 64 pgs > objects: 2.65 M objects, 178 GiB > usage: 548 GiB used, 6.7 TiB / 7.3 TiB avail > pgs: 834820/7943361 objects misplaced (10.510%) > 52 active+clean > 11 active+remapped+backfill_wait > 1 active+remapped+backfilling > > io: > recovery: 0 B/s, 6 objects/s > > output from "ceph osd df " > > ID CLASS WEIGHT REWEIGHT SIZE USE AVAIL %USE VAR PGS > 0 hdd 1.81940 1.00000 1.8 TiB 88 GiB 1.7 TiB 4.71 0.64 40 > 3 hdd 1.81940 1.00000 1.8 TiB 96 GiB 1.7 TiB 5.15 0.70 24 > 1 hdd 1.81940 1.00000 1.8 TiB 182 GiB 1.6 TiB 9.79 1.33 64 > 2 hdd 1.81940 1.00000 1.8 TiB 182 GiB 1.6 TiB 9.79 1.33 64 > TOTAL 7.3 TiB 548 GiB 6.7 TiB 7.36 > MIN/MAX VAR: 0.64/1.33 STDDEV: 2.43 > > regards > Amudhan P > > _______________________________________________ > ceph-users mailing list -- [email protected] > To unsubscribe send an email to [email protected] > > _______________________________________________ > ceph-users mailing list -- [email protected] > To unsubscribe send an email to [email protected] >
_______________________________________________ ceph-users mailing list -- [email protected] To unsubscribe send an email to [email protected]
