Bug 1328594

Summary:	cluster nodes warn about missing rbd client admin socket
Product:	Red Hat Storage Console	Reporter:	Ken Dreyer (Red Hat) <kdreyer>
Component:	ceph-installer	Assignee:	Andrew Schoen <aschoen>
Status:	CLOSED ERRATA	QA Contact:	Rachana Patel <racpatel>
Severity:	urgent	Docs Contact:
Priority:	urgent
Version:	2	CC:	aschoen, ceph-eng-bugs, hnallurv, johfulto, nthomas, sankarshan, tchandra
Target Milestone:	---
Target Release:	2
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:	ceph-installer-1.0.5-4.el7	Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2016-08-23 19:49:19 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:

Description Ken Dreyer (Red Hat) 2016-04-19 19:32:11 UTC

Description of problem:
"ceph" command-line operations (like "ceph health") print a warning to the console because the RBD client admin socket is not present.

Version-Release number of selected component (if applicable):
ceph-installer-1.0.5-1.el7.noarch
ceph-ansible-1.0.5-3.el7.noarch


Steps to Reproduce:
1. Set up a cluster with ceph-installer.
2. SSH to a cluster node (mon or OSD).
3. Run "sudo ceph health"

Actual results:
2016-04-19 15:20:05.128327 7fc0e4033700 -1 asok(0x7fc0dc001680) AdminSocketConfigObs::init: failed: AdminSocket::bind_and_listen: failed to bind the UNIX domain socket to '/var/run/ceph/rbd-clients/ceph-client.admin.20455.140466301441968.asok': (2) No such file or directory
HEALTH_OK


Expected results:
HEALTH_OK
(with no warnings)

Additional info:
This seems to be a regression from a month ago, but I don't see any blindingly-obvious commits in ceph-ansible that would have caused this.

Comment 2 Ken Dreyer (Red Hat) 2016-04-19 20:28:21 UTC

FWIW I tried this with an old build of ceph (10.0.4-2.el7cp) and it still shows up. That indicates it's likely to be in the installer (ceph-ansible)

Comment 3 Ken Dreyer (Red Hat) 2016-04-19 20:35:08 UTC

Commenting out "admin socket" in the "[client]" section makes `ceph health` print HEALTH_OK without any warnings.

Comment 4 Andrew Schoen 2016-04-21 18:56:34 UTC

A fix for this is in the following PR: https://github.com/ceph/ceph-ansible/pull/721

Comment 11 Tejas 2016-07-29 14:39:53 UTC

This is not seen in any of our recent tests.
Moving to verified.

Comment 13 errata-xmlrpc 2016-08-23 19:49:19 UTC

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHEA-2016:1754