On Tue, Feb 14, 2023 at 12:05 AM 郑亮 <zhengliang0901(a)gmail.com> wrote:
Hi all,
Does cephfs subvolume have commands similar to rbd perf to query iops,
bandwidth, and latency of rbd image? `ceph fs perf stats` shows metrics of
the client side, not the metrics of the cephfs subvolume. What I want to
get is the metrics at the subvolume level like below.
You'd need to mount the subvolume to get stats for it via the `ceph fs
perf stats' command you mention and/or (preferably) using
cephfs-top[0] tool.
[0]:
https://docs.ceph.com/en/latest/cephfs/cephfs-top/
[root@smd-exporter-5f87dcb946-wt7tl /]# rbd perf image iostat
rbd: waiting for initial image stats
NAME WR RD
WR_BYTES RD_BYTES WR_LAT RD_LAT
pool/csi-vol-5c2d22e3-9195-11ed-aab9-a6732a88c7dd 8/s 0/s 215
KiB/s 0 B/s 14.62 ms 0.00 ns
pool/csi-vol-10e31971-9196-11ed-aab9-a6732a88c7dd 5/s 0/s 79
KiB/s 0 B/s 13.18 ms 0.00 ns
pool/csi-vol-0f6e0dbe-9196-11ed-aab9-a6732a88c7dd 1/s 0/s 23
KiB/s 0 B/s 20.60 ms 0.00 ns
pool/csi-vol-88704e7a-919c-11ed-aab9-a6732a88c7dd 1/s 0/s 12
KiB/s 0 B/s 12.38 ms 0.00 ns
pool/csi-vol-a87c21e9-919c-11ed-aab9-a6732a88c7dd 1/s 0/s 14
KiB/s 0 B/s 13.21 ms 0.00 ns
pool/csi-vol-b6d040a2-919c-11ed-aab9-a6732a88c7dd 0/s 0/s 8
KiB/s 0 B/s 9.88 ms 0.00 ns
pool/csi-vol-efb7c236-6fc9-11ed-aab9-a6732a88c7dd 0/s 0/s 3.2
KiB/s 0 B/s 2.96 ms 0.00 ns
NAME WR RD
WR_BYTES RD_BYTES WR_LAT RD_LAT
pool/csi-vol-5c2d22e3-9195-11ed-aab9-a6732a88c7dd 12/s 0/s 165
KiB/s 0 B/s 26.07 ms 0.00 ns
pool/csi-vol-10e31971-9196-11ed-aab9-a6732a88c7dd 4/s 0/s 91
KiB/s 0 B/s 9.83 ms 0.00 ns
pool/csi-vol-b6d040a2-919c-11ed-aab9-a6732a88c7dd 1/s 0/s 22
KiB/s 0 B/s 15.25 ms 0.00 ns
pool/csi-vol-efb7c236-6fc9-11ed-aab9-a6732a88c7dd 1/s 0/s 38
KiB/s 0 B/s 20.40 ms 0.00 ns
pool/csi-vol-88704e7a-919c-11ed-aab9-a6732a88c7dd 0/s 0/s 9.6
KiB/s 0 B/s 12.96 ms 0.00 ns
pool/csi-vol-a87c21e9-919c-11ed-aab9-a6732a88c7dd 0/s 0/s 10
KiB/s 0 B/s 3.99 ms 0.00 ns
pool/csi-vol-0f6e0dbe-9196-11ed-aab9-a6732a88c7dd 0/s 0/s 2.4
KiB/s 0 B/s 7.33 ms 0.00 ns
Best Regards,
Liang Zheng
_______________________________________________
ceph-users mailing list -- ceph-users(a)ceph.io
To unsubscribe send an email to ceph-users-leave(a)ceph.io
--
Cheers,
Venky