Project

General

Profile

Actions
Copy link

Bug #44074

open

new "daemons have recently crashed" is confusing

Added by Sébastien Han about 4 years ago. Updated about 4 years ago.

Status:
New
Priority:
Normal
Assignee:
-
Category:
-
Target version:
-
% Done:

0%

Source:
Tags:
Backport:
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

My cluster is showing the following:

[WRN] RECENT_CRASH: 1 daemons have recently crashed
    client.admin crashed on host rook-ceph-osd-0-9b68cb58f-fq84f at 2020-02-10 14:15:27.822992Z

It's happening on Rook, rook-ceph-osd-0-9b68cb58f-fq84f is the previous OSD pod which is now gone.
I upgraded from 14.2.7 to "ceph version 15.1.0-336-g3f4065c (3f4065c44305b20b698eea84e0440e698aab9280) octopus (rc)" and then got that message.

Not reproducible when restarting the container.

But still, since the error seems related to the client, it seems confusing to treat the client as a daemon.

Actions
Copy link
 #1

Updated by Neha Ojha about 4 years ago

  • Project changed from RADOS to mgr
Actions
Copy link

Also available in: Atom PDF