Hi Ronny On 5/12/22 12:47, ronny.lippold wrote: > hi to all here > we tried a lot and now, we need your help ... > > we are using 5 proxmox 7.2-3 server with kernel 5.15.30-2-pve and ceph > 16.2.7.. > per server, we use 9 osd (8x 2tb, 1x8tb both sas ssd, connected via sas > hba) > the second cluster for replication is the same hardware. > > > at first, we tried was journal mode. > everything was fine, but the used disk space in the cluster was growing > (nearly 700gb per day). > entries_behind_primary were looking good, some images with a value of > 300, most 1-3. Yes according to my test we were not able to have a good replication speed on a single image (I think it was 30Mb/s per image something like that). So you have probably a few image that write a lot and thus are much behind in term of replication... > after that, we switched to snapshot based. > also ... everything was working fine (50 days), up to last week. > since last week, we see a high i/o load in our virtual machines (kvm). > > we tried a lot, but did not find anything. > the hardware, proxmox and kvms were not modified. > after disabling the snapshot feature on the rbd images, the load goes > down. > disabling the rbd mirror processes did not help, load stays up. If you have no rbd-mirror running while snapshot mirroring is enabled, for me it means me that the load come from taking/deleting snapshots... At what interval did you configure for mirror snapshots? Cheers, -- Arthur Outhenin-Chalandre _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |