[multiple_cluster_networks] Ceph can't start on nodes from non-default nodegroup

Bug #1395048 reported by Artem Panchenko
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Fuel for OpenStack
Fix Committed
High
Vladimir Kuklin

Bug Description

api: '1.0'
astute_sha: 65eb911c38afc0e23d187772f9a05f703c685896
auth_required: true
build_id: 2014-11-19_11-01-11
build_number: '115'
feature_groups:
- mirantis
fuellib_sha: 74f7d1923d0b450be69850a82f3be8b17fc707c9
fuelmain_sha: fc2c7b9deb0dad12482410d4750b264c57450c0c
nailgun_sha: ffb507b8222f1529bca113489413ed536d2b517b
ostf_sha: 85272980659a977582e5b2dc67479fe9d7c1bbe7
production: docker
release: '6.0'

Deployment of HA cluster with few nodegroups and Ceph for volumes/images fails, because Ceph can't start on nodes from non-default net group:

http://jenkins-product.srt.mirantis.net:8080/job/6.0_multiple-cluster-networks_temp/22/testReport/(root)/deploy_ceph_ha_nodegroups/deploy_ceph_ha_nodegroups/

Steps to reproduce:

1. Create new cluster: CentOS, HA, NeutronGre, Ceph for volumes & images
2. Add 3 controller+ceph nodes from default nodegroup and 2 compute+ceph from non-default nodegroup
3. Deploy changes

Expected behavior:

 - environment successfully deployed

Actual result:

 - deployment of computes+ceph nodes failed. Here is the part of puppet logs (node-4): http://paste.openstack.org/show/135970/

I think that Ceph was configured incorrectly by puppet on nodes from non-default nodegroup:

http://paste.openstack.org/show/135969/

After I manually changed networks addresses in ceph.conf, service has been started:

http://paste.openstack.org/show/135968/

Diagnostic snapshot is attached.

Revision history for this message
Artem Panchenko (apanchenko-8) wrote :
Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Fix proposed to fuel-library (master)

Fix proposed to branch: master
Review: https://review.openstack.org/136516

Changed in fuel:
status: New → In Progress
Changed in fuel:
assignee: Ryan Moe (rmoe) → Vladimir Kuklin (vkuklin)
Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Fix merged to fuel-library (master)

Reviewed: https://review.openstack.org/136516
Committed: https://git.openstack.org/cgit/stackforge/fuel-library/commit/?id=00af1d323306f8f5a5602e12c2701bbc355cae7b
Submitter: Jenkins
Branch: master

commit 00af1d323306f8f5a5602e12c2701bbc355cae7b
Author: Vladimir Kuklin <email address hidden>
Date: Wed Dec 3 19:57:48 2014 +0300

    Use get_network_role_property for Ceph network settings

    Ceph nodes outside of the default nodegroup need to use
    their group's management and storage networks.

    Change-Id: I5c52874d06d2107f523e999d931c501e3da50bc2
    Closes-bug: #1395048

Changed in fuel:
status: In Progress → Fix Committed
Andrew Woodward (xarses)
tags: added: feature-nodegroup
removed: multiple-cluster-networks
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.