Bug #23017: mgr log spamming about down osds - mgr - Ceph

Actions

Copy link

Bug #23017

closed

mgr log spamming about down osds

Added by Dan van der Ster about 6 years ago. Updated about 6 years ago.

Status:

Resolved

Priority:

Normal

Assignee:

Category:

ceph-mgr

Target version:

% Done:

Source:

Community (user)

Tags:

Backport:

luminous

Regression:

Severity:

3 - minor

Reviewed:

Affected Versions:

Ceph - v12.2.2

ceph-qa-suite:

Pull request ID:

Crash signature (v1):

Crash signature (v2):

Description

When you have down osds in a cluster the ceph-mgr log is spammed with one line per down osd per second:

2018-02-15 17:46:46.419099 7f200d2bd700  0 Cannot get stat of OSD 1012
2018-02-15 17:46:48.425872 7f200d2bd700  0 Cannot get stat of OSD 1012
2018-02-15 17:46:48.956493 7f1fedd80700  0 Cannot get stat of OSD 1012
2018-02-15 17:46:50.509597 7f200d2bd700  0 Cannot get stat of OSD 1012

The osd is down and out:

osd.1012 down out weight 0 up_from 93002 up_thru 93002 down_at 93005 last_clean_interval [88366,93001) 128.142.23.74:6800/1586568 128.142.23.74:6801/4586568 128.142.23.74:6810/4586568 128.142.23.74:6812/4586568 autoout,exists 46de0a55-62c8-44b2-b1fe-5a0d81f7b530

Related issues 1 (0 open — 1 closed)