Project

General

Profile

Bug #21225

ceph-mgr: dashboard and zabbix plugin report wrong values

Added by Tobias Rehn about 2 years ago. Updated over 1 year ago.

Status:
Can't reproduce
Priority:
Normal
Assignee:
-
Category:
-
Target version:
Start date:
09/04/2017
Due date:
% Done:

0%

Source:
Tags:
Backport:
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Pull request ID:

Description

I have installed a ceph cluster using the latest stable (luminous 12.2.0). I enabled the dashboard and zabbix plugin. Both report wrong values for ceph health. The cluster is in HEALTH_WARN state whereas both zabbix and the dashboard show HEALTH_OK.

I have added a screenshot from the new dashboard which show a degraded pool and the overall status "HEALTH_OK".

Bildschirmfoto 2017-09-04 um 15.48.12.jpg View (908 KB) Tobias Rehn, 09/04/2017 02:03 PM

History

#1 Updated by Josh Durgin about 2 years ago

  • Project changed from Ceph to mgr

#2 Updated by John Spray about 2 years ago

How odd...

Please could you add these settings on mon and mgr nodes:

debug mon = 10
debug mgr = 10
debug ms = 1

Then restart mon+mgr, get it into this state again (system unhealthy but mgr says HEALTH_OK), and attach the mon/mgr logs since the restart.

#3 Updated by Tobias Rehn about 2 years ago

It is getting strange now. I restarted my mon/mgr nodes and the problem disappeared. It seems to work now and I cannot reproduce the problem currently. Maybe the active mgr had an issue.

I will keep my eyes on the problem and get back to you once I have further information. From my side the ticket can be closed.

#4 Updated by John Spray about 2 years ago

  • Status changed from New to Can't reproduce

#5 Updated by Hans van den Bogert over 1 year ago

Has this been addressed in another ticket? This is exactly what I'm experiencing all the time.

#6 Updated by Martin Emrich over 1 year ago

Same issue here with 12.2.2. I just restarted all mgr and mon with debuglevel 10. After restart, the dashboard correctly displays the HEALTH_WARN state.

#7 Updated by John Spray over 1 year ago

This seems likely to be the same issue as http://tracker.ceph.com/issues/22142, the fix for which will be in 12.2.3

#8 Updated by Martin Emrich over 1 year ago

Thanks John... :)

Time to upgrade...

Also available in: Atom PDF