In environments with ~200 osds and ~6000 pgs /metrics take ~100 seconds to return

Bug #1924592 reported by Giuseppe Petralia
18
This bug affects 4 people
Affects Status Importance Assigned to Milestone
Prometheus Ceph Exporter Charm
Won't Fix
Medium
Unassigned
prometheus-ceph-exporter snap
Won't Fix
Undecided
Unassigned

Bug Description

Ceph exporter takes ~100 seconds to return

root@juju-733f2f-5-lxd-13:~# /usr/lib/nagios/plugins/check_http -I 127.0.0.1 -p 9128 -u /metrics -t 100
HTTP OK: HTTP/1.0 200 OK - 416838 bytes in 87.556 second response time |time=87.555859s;;;0.000000;100.000000 size=416838B;;;0

Ceph cluster size:

root@juju-733f2f-3-lxd-1:~# ceph -s
  cluster:
    id: 3d196f64-760e-11eb-b486-e7aafffac620
    health: HEALTH_OK

  services:
    mon: 3 daemons, quorum juju-733f2f-3-lxd-1,juju-733f2f-5-lxd-1,juju-733f2f-4-lxd-1 (age 26h)
    mgr: juju-733f2f-3-lxd-1(active, since 2w), standbys: juju-733f2f-4-lxd-1, juju-733f2f-5-lxd-1
    osd: 189 osds: 189 up (since 24h), 189 in (since 2w)
    rgw: 3 daemons active (juju-733f2f-3-lxd-12, juju-733f2f-4-lxd-12, juju-733f2f-5-lxd-12)

  task status:

  data:
    pools: 20 pools, 6177 pgs
    objects: 15.92M objects, 54 TiB
    usage: 162 TiB used, 1.7 PiB / 1.9 PiB avail
    pgs: 6170 active+clean
             7 active+clean+scrubbing+deep

  io:
    client: 57 KiB/s rd, 3.3 MiB/s wr, 19 op/s rd, 470 op/s wr

Also syslog is flooded with:

Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"4.6c2","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"3.6c4","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"4.6c3","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"3.6c3","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"4.6c4","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"3.6c2","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"4.6c5","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"3.6c1","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"4.6c6","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"3.6c0","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"4.6c7","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"3.6bf","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"4.6b8","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"3.6be","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"4.6b9","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"3.6bd","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"4.6ba","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"3.6bc","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"4.6bb","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"3.6bb","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"4.6bc","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"3.6ba","prefix":"query"}: rados: Permission denied
Apr 15 14:32:10 juju-733f2f-5-lxd-13 prometheus-ceph-exporter.ceph-exporter[832270]: 2021/04/15 14:32:10 failed sending PG command {"format":"json","pgid":"4.6bd","prefix":"query"}: rados: Permission denied

We should understand if there is a better configuration to run ceph-exporter

i.e. setting CEPH_RADOS_OP_TIMEOUT when runnign ceph-exporter
https://github.com/digitalocean/ceph_exporter

Edin S (exsdev)
Changed in charm-prometheus-ceph-exporter:
importance: Undecided → Medium
Revision history for this message
Eric Chen (eric-chen) wrote :

This charm is no longer being actively maintained. Please consider using the new Canonical Observability Stack instead. (https://charmhub.io/topics/canonical-observability-stack)

Changed in charm-prometheus-ceph-exporter:
status: New → Won't Fix
Changed in snap-prometheus-ceph-exporter:
status: New → Won't Fix
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.