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@ceph.io To unsubscribe send an email toceph-users-leave@ceph.io