Hi Konstantin, thanks for your suggestions. > Lars, you have too much PG's for this OSD's. I suggest to disable PG > autoscaler and: > > - reduce number of PG's for cephfs_metada pool to something like 16 PG's. Done. > > - reduce number of PG's for cephfs_data to something like 512. Done. > - update crush rule for cephfs_metadata pool - set domain to 'rack' > instead 'host'. Done. > Also please paste your `ceph osd tree`. > $ ceph osd tree ID CLASS WEIGHT TYPE NAME STATUS REWEIGHT PRI-AFF -1 195.40730 root default -25 195.40730 room PRZ -26 195.40730 row rechts -27 83.74599 rack 1-eins -3 27.91533 host onode1 0 hdd 5.51459 osd.0 up 1.00000 1.00000 1 hdd 5.51459 osd.1 up 1.00000 1.00000 2 hdd 5.51459 osd.2 up 1.00000 1.00000 3 hdd 5.51459 osd.3 up 1.00000 1.00000 37 hdd 5.51459 osd.37 up 1.00000 1.00000 4 ssd 0.34239 osd.4 up 1.00000 1.00000 -13 27.91533 host onode4 13 hdd 5.51459 osd.13 up 1.00000 1.00000 14 hdd 5.51459 osd.14 up 1.00000 1.00000 15 hdd 5.51459 osd.15 up 1.00000 1.00000 16 hdd 5.51459 osd.16 up 1.00000 1.00000 40 hdd 5.51459 osd.40 up 1.00000 1.00000 33 ssd 0.34239 osd.33 up 1.00000 1.00000 -22 27.91533 host onode7 25 hdd 5.51459 osd.25 up 1.00000 1.00000 26 hdd 5.51459 osd.26 up 1.00000 1.00000 27 hdd 5.51459 osd.27 up 1.00000 1.00000 28 hdd 5.51459 osd.28 up 1.00000 1.00000 30 hdd 5.51459 osd.30 up 1.00000 1.00000 36 ssd 0.34239 osd.36 up 1.00000 1.00000 -28 55.83066 rack 2-zwei -7 27.91533 host onode2 5 hdd 5.51459 osd.5 up 1.00000 1.00000 6 hdd 5.51459 osd.6 up 1.00000 1.00000 7 hdd 5.51459 osd.7 up 1.00000 1.00000 8 hdd 5.51459 osd.8 up 1.00000 1.00000 38 hdd 5.51459 osd.38 up 1.00000 1.00000 31 ssd 0.34239 osd.31 up 1.00000 1.00000 -16 27.91533 host onode5 17 hdd 5.51459 osd.17 up 1.00000 1.00000 18 hdd 5.51459 osd.18 up 1.00000 1.00000 19 hdd 5.51459 osd.19 up 1.00000 1.00000 20 hdd 5.51459 osd.20 up 1.00000 1.00000 41 hdd 5.51459 osd.41 up 1.00000 1.00000 34 ssd 0.34239 osd.34 up 1.00000 1.00000 -29 55.83066 rack 3-drei -10 27.91533 host onode3 9 hdd 5.51459 osd.9 up 1.00000 1.00000 10 hdd 5.51459 osd.10 up 1.00000 1.00000 11 hdd 5.51459 osd.11 up 1.00000 1.00000 12 hdd 5.51459 osd.12 up 1.00000 1.00000 39 hdd 5.51459 osd.39 up 1.00000 1.00000 32 ssd 0.34239 osd.32 up 1.00000 1.00000 -19 27.91533 host onode6 21 hdd 5.51459 osd.21 up 1.00000 1.00000 22 hdd 5.51459 osd.22 up 1.00000 1.00000 23 hdd 5.51459 osd.23 up 1.00000 1.00000 24 hdd 5.51459 osd.24 up 1.00000 1.00000 29 hdd 5.51459 osd.29 up 1.00000 1.00000 35 ssd 0.34239 osd.35 up 1.00000 1.00000 So I just wait for the remapping and merging being done and see what happens. Thanks so far! Best regards, Lars _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |