Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 1199266

Summary: OSP compute nodes should not rely on ceph command
Product: Red Hat OpenStack Reporter: Keith Schincke <kschinck>
Component: openstack-foreman-installerAssignee: Jason Guiditta <jguiditt>
Status: CLOSED ERRATA QA Contact: Yogev Rabl <yrabl>
Severity: high Docs Contact:
Priority: high    
Version: 6.0 (Juno)CC: aberezin, arkady_kanevsky, cwolfe, dmacpher, ebarrera, egolov, mburns, morazi, nlevinki, ohochman, rajini.karthik, rhos-maint, sclewis, yeylon
Target Milestone: z2Keywords: ZStream
Target Release: Installer   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: openstack-foreman-installer-3.0.17-1.el7ost Doc Type: Enhancement
Doc Text:
This enhancement removes Ceph-specific configuration files installed on Compute nodes since Ceph is neither available nor required on Compute nodes.
Story Points: ---
Clone Of: Environment:
Last Closed: 2015-04-07 15:08:48 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:
Embargoed:

Description Keith Schincke 2015-03-05 18:58:00 UTC
Description of problem:

Puppet deploys a complete ceph.client.volumes.keyring file on the OSP
compute nodes. Only the key value is needed to complete the
configuration of libvirt. The ceph command is not available on the OSP
compute nodes.

Recommend changing puppet to call virsh and feed it the value of the
generated volumes key.

/usr/bin/virsh secret-set-value --secret ${rbd_secret_uuid} --base64
${volumes_key} 

Version-Release number of selected component (if applicable):
foreman-installer-1.6.0-0.3.RC1.el7ost.noarch


How reproducible:
100%

Steps to Reproduce:
1.
2.
3.

Actual results:


Expected results:


Additional info:

Comment 5 Jason Guiditta 2015-03-10 18:41:50 UTC
Patch posted:
https://github.com/redhat-openstack/astapor/pull/487

Comment 6 Rajini Karthik 2015-03-11 02:41:31 UTC
+1

Comment 7 Crag Wolfe 2015-03-11 05:25:50 UTC
I'm certain the pre-patch behaviour was such that a puppet run would never display an error (esp. if the ceph cluster wasn't set up yet).  I'm not certain this is the case with patch (do we care?).  The advantage to the patch (as everybody is probably aware) is that ceph admin is not required on the compute node.

Comment 8 Crag Wolfe 2015-03-11 16:43:11 UTC
*** Bug 1197603 has been marked as a duplicate of this bug. ***

Comment 9 Keith Schincke 2015-03-11 16:50:45 UTC
Does this patch stop the deployment of the /etc/ceph/ceph.volumes.keyring file?

Comment 10 Crag Wolfe 2015-03-11 16:54:53 UTC
No, it does not.  /etc/ceph/ceph.client.volumes.keyring will still be there.

Comment 11 Crag Wolfe 2015-03-11 18:53:40 UTC
To correct my earlier Comment 7, a puppet run on a compute node with this patch will not cause a puppet error if the ceph cluster is not yet set up (good!).

Comment 12 Evgeni Golov 2015-03-12 13:13:35 UTC
RE Comment 7: the old code will produce Puppet errors in some cases too:
Mar 10 15:44:31 compute01 puppet-agent[12959]: Could not set 'file' on ensure: No such file or directory - /etc/nova/secret.xml20150310-12959-1wqjr2a at 127:/etc/puppet/environments/production/modules/quickstack/manifests/compute_common.pp
Mar 10 15:44:31 compute01 puppet-agent[12959]: Could not set 'file' on ensure: No such file or directory - /etc/nova/secret.xml20150310-12959-1wqjr2a at 127:/etc/puppet/environments/production/modules/quickstack/manifests/compute_common.pp
Mar 10 15:44:31 compute01 puppet-agent[12959]: Wrapped exception:
Mar 10 15:44:31 compute01 puppet-agent[12959]: No such file or directory - /etc/nova/secret.xml20150310-12959-1wqjr2a
Mar 10 15:44:31 compute01 puppet-agent[12959]: (/Stage[main]/Quickstack::Compute_common/File[/etc/nova/secret.xml]/ensure) change from absent to file failed: Could not set 'file' on ensure: No such file or directory - /etc/nova/secret.xml20150310-12959-1wqjr2a at 127:/etc/puppet/environments/production/modules/quickstack/manifests/compute_common.pp

(This being a special case somewhat, as the CEPH cluster used was not set-up by the installer but by-hand before the OSP deployment)

Comment 13 Jason Guiditta 2015-03-12 20:10:49 UTC
Merged

Comment 15 Yogev Rabl 2015-03-30 13:39:22 UTC
verified in version ruby193-rubygem-staypuft-0.5.21-1.el7ost.noarch

Comment 17 errata-xmlrpc 2015-04-07 15:08:48 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://rhn.redhat.com/errata/RHSA-2015-0791.html