Did you check how many caps the clients are using? The thread you
refer to contains some instructions.
Zitat von Ben :
checked that disk utilization had been normal during the incident. cephfs
slow performance could not seemingly be attributed to osd.
All HDD, no ssd in fact.
I found this t
checked that disk utilization had been normal during the incident. cephfs
slow performance could not seemingly be attributed to osd.
All HDD, no ssd in fact.
I found this thread related:
https://lists.ceph.io/hyperkitty/list/ceph-users@ceph.io/thread/B7K6B5VXM3I7TODM4GRF3N7S254O5ETY/
Does it ha
Do you see a high disk utilization? Are those OSDs hdd-only or at
least have their db on SSDs? I'd say the HDDs are the bottleneck.
There was a recent thread [1] where Zakhar explained nicely how many
IOPS you can expect from a hdd-only cluster. Maybe that helps.
[1]
https://lists.ceph.io
