Hi,
Confirmed that this happens to me as well.
After upgrading from 18.2.0 to 18.2.1 OSD metrics like: ceph_osd_op_*
are missing from ceph-mgr.
The Grafana dashboard also doesn't display all graphs correctly.
ceph-dashboard/Ceph - Cluster : Capacity used, Cluster I/O, OSD Capacity
Utilization, PGs per OSD....
curl http://localhost:9283/metrics | grep -i ceph_osd_op
% Total % Received % Xferd Average Speed Time Time Time
Current
Dload Upload Total Spent Left Speed
100 38317 100 38317 0 0 9.8M 0 --:--:-- --:--:-- --:--:--
12.1M
Before the upgrading to reef 18.2.1 I could get all the metrics.
Martin
On 18/01/2024 12:32, Jose Vicente wrote:
Hi,
After upgrading from Quincy to Reef the ceph-mgr daemon is not
throwing some throughput OSD metrics like: ceph_osd_op_*
curl http://localhost:9283/metrics | grep -i ceph_osd_op
% Total % Received % Xferd Average Speed Time Time Time
Current
Dload Upload Total Spent Left
Speed
100 295k 100 295k 0 0 144M 0 --:--:-- --:--:--
--:--:-- 144M
However I can get other metrics like:
# curl http://localhost:9283/metrics | grep -i ceph_osd_apply
# HELP ceph_osd_apply_latency_ms OSD stat apply_latency_ms
# TYPE ceph_osd_apply_latency_ms gauge
ceph_osd_apply_latency_ms{ceph_daemon="osd.275"} 152.0
ceph_osd_apply_latency_ms{ceph_daemon="osd.274"} 102.0
...
Before the upgrading to reef (from quincy) I I could get all the
metrics. MGR module prometheus is enabled.
Rocky Linux release 8.8 (Green Obsidian)
ceph version 18.2.1 (7fe91d5d5842e04be3b4f514d6dd990c54b29c76) reef
(stable)
# netstat -nap | grep 9283
tcp 0 0 127.0.0.1:53834 127.0.0.1:9283
ESTABLISHED 3561/prometheus
tcp6 0 0 :::9283 :::* LISTEN
804985/ceph-mgr
Thanks,
Jose C.
_______________________________________________
ceph-users mailing list --ceph-users@xxxxxxx
To unsubscribe send an email toceph-users-leave@xxxxxxx
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
