Project

General

Profile

Bug #2721

Ceph status does not work in 0.48 even if it is still documented

Added by Maciej Galkiewicz about 11 years ago. Updated over 10 years ago.

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

0%

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

Description

# ceph -s
unrecognized subsystem
# ceph --help
usage: ceph [options] [commands]
If no commands are specified, enter interactive mode.
Commands:
   (osd|pg|mds) stat -- get monitor subsystem status
   ...
Options:
   -i infile
   -o outfile
        specify input or output file (for certain commands)
   -s or --status
        print current system status
   -w or --watch
        watch system status changes in real time (push)
--conf/-c        Read configuration from the given configuration file
-d               Run in foreground, log to stderr.
-f               Run in foreground, log to usual location.
--id/-i          set ID portion of my name
--name/-n        set name (TYPE.ID)
--version        show version and quit

# ceph -s osd
unrecognized subsystem
# ceph osd stat
e39299: 2 osds: 2 up, 2 in

History

#1 Updated by Sage Weil about 11 years ago

can you include 'ceph -v' output? this works for me...

root@burnupi30:~# ceph -s
   health HEALTH_OK
   monmap e1: 3 mons at {a=10.214.135.30:6789/0,b=10.214.135.28:6789/0,c=10.214.135.26:6789/0}, election epoch 56, quorum 0,1,2 a,b,c
   osdmap e1521: 191 osds: 191 up, 191 in
    pgmap v1017142: 70472 pgs: 70471 active+clean, 1 active+clean+scrubbing; 10595 GB data, 31725 GB used, 140 TB / 175 TB avail
   mdsmap e1: 0/0/1 up

root@burnupi30:~# ceph -v
ceph version 0.48argonaut-2-g67455c2 (commit:67455c21879c9c117f6402259b5e2da84524e169)

#2 Updated by Maciej Galkiewicz about 11 years ago

  1. ceph -v
    ceph version 0.48argonaut (c2b20ca74249892c8e5e40c12aa14446a2bf2030)

#3 Updated by Sage Weil about 11 years ago

Weird! Two things:

- does 'ceph status' work?  (this is identical to -s now)
 - can you try with commit:b53cdb97d15f9276a9b26bec9f29034149f93358?  this was accessing invalid memory in the arg parsing code.  maybe that is responsible (it's one of the two extra argonaut patches on the cluster i'm testing with)

#4 Updated by Maciej Galkiewicz about 11 years ago 

# ceph status
unrecognized subsystem

I will try to check suggested commit.

#5 Updated by Sage Weil about 11 years ago

Oh! The monitor is still running the old version.

#6 Updated by Sage Weil about 11 years ago

you can just restart, or to verify in the future, 'ceph --admin-daemon /var/run/ceph/ceph-mon.a.asok version' (adjust socket path as needed)

#7 Updated by Maciej Galkiewicz about 11 years ago

I have restarted ceph mon with invoke-rc.d ceph restart mon and now it works! However I am 100% sure that I did it after upgrading ceph with command invoke-rc.d ceph restart.

#8 Updated by Sage Weil over 10 years ago

  • Status changed from New to Resolved

wrong monitor version was running

Also available in: Atom PDF