Project

General

Profile

Actions
Copy link

Bug #24089

closed

mds: print slow requests to debug log when sending health WRN to monitors (if < ~5)

Added by Patrick Donnelly almost 6 years ago. Updated almost 6 years ago.

Status:
Rejected
Priority:
Urgent
Assignee:
Patrick Donnelly
Category:
Introspection/Control
Target version:
Ceph - v13.2.0
% Done:

0%

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

Description

2018-05-09 16:40:35.802 7f7941e92700 10 mds.0.server find_idle_sessions.  laggy until 0.000000
2018-05-09 16:40:35.802 7f7941e92700 20 mds.0.server laggiest active session is client.4384 172.21.15.31:0/3739868178
2018-05-09 16:40:35.802 7f7941e92700 20 mds.0.server laggiest active session is client.4384 172.21.15.31:0/3739868178 and sufficiently new (2018-05-09 16:40:22.145594)
2018-05-09 16:40:35.802 7f7941e92700 20 mds.0.server not evicting a slow client, because there is only one

From: /ceph/teuthology-archive/pdonnell-2018-05-09_16:08:35-kcephfs-mimic-testing-basic-smithi/2510137/remote/smithi197/log/ceph-mds.a.log.gz

It'd be helpful to actually see what that was. (I can probably figure out what it was by reading the rest of the log --which I'm about to do -- but it'd make things easier to just put it right there.)

Actions
Copy link
 #1

Updated by Patrick Donnelly almost 6 years ago

  • Status changed from New to Rejected

Nevermind, it is actually printed earlier in the log. Sorry for the noise.

Actions
Copy link

Also available in: Atom PDF