ironic_deploy_ceph test failed after deployment on getting ironic-deploy-linux image
Affects | Status | Importance | Assigned to | Milestone | ||
---|---|---|---|---|---|---|
| Fuel for OpenStack |
High
|
Andrey Shestakov | |||
| Mitaka |
High
|
Andrey Shestakov | |||
| Newton |
High
|
Andrey Shestakov | |||
Mirantis OpenStack | Status tracked in 10.0.x | |||||
| 10.0.x |
High
|
Andrey Shestakov | |||
| 8.0.x |
High
|
Vitaly Sedelnik | |||
| 9.x |
High
|
Andrey Shestakov |
Bug Description
iso - 8.0-566
Steps:
1. Create cluster
2. Add 1 node with controller+ceph-osd role
3. Add 2 nodes with controller+
4. Add 1 node with compute role
5. Add 1 nodes with ironic role
6. Deploy the cluster
7. Upload image to glance - http://
8. Get 'ironic-
Actual result - image is not present that caused test failure
http://
Andrey Sledzinskiy (asledzinskiy) wrote : | #1 |
tags: | added: area-qa |
Changed in fuel: | |
status: | New → Confirmed |
Changed in fuel: | |
assignee: | Kyrylo Romanenko (kromanenko) → MOS Ironic (mos-ironic) |
Changed in fuel: | |
assignee: | MOS Ironic (mos-ironic) → Andrey Shestakov (ashestakov) |
Andrey Shestakov (ashestakov) wrote : | #2 |
tags: | removed: area-qa |
affects: | fuel → mos |
Changed in mos: | |
milestone: | 9.0 → none |
no longer affects: | fuel/8.0.x |
tags: | added: ironic |
Timur Nurlygayanov (tnurlygayanov) wrote : | #3 |
Does the issue reproduced in 100 % of cases or it is a question of timeouts? If it is possible random fails we need to move it to MOS 8.0 MU and MOS 9.0, if it is reproducing in 100% of cases we need to fix it as soon as possible.
Andrey Shestakov (ashestakov) wrote : | #4 |
We never got this situation. looks like very low chance
Serge Kovaleff (serge-kovaleff) wrote : | #5 |
Should we close it as Cannot reproduce?
tags: |
added: area-ironic removed: ironic |
Andrey Shestakov (ashestakov) wrote : | #6 |
Addressed by https:/
Changed in fuel: | |
assignee: | nobody → Andrey Shestakov (ashestakov) |
importance: | Undecided → High |
milestone: | none → 10.0 |
status: | New → Fix Committed |
Reviewed: https:/
Committed: https:/
Submitter: Jenkins
Branch: stable/mitaka
commit 2ab7c95205af43a
Author: Andrey Shestakov <email address hidden>
Date: Fri Mar 25 13:28:29 2016 +0200
Add RadosGW readiness check
After radosgw service start it requires some time to get operational state
This change introduces check script which waits until radosgw start to serve
http requests.
Change-Id: Iaeb3f60c0da114
Closes-bug: #1545651
(cherry picked from commit 3d159974341211f
Verified
[root@nailgun ~]# shotgun2 short-report
cat /etc/fuel_build_id:
193
cat /etc/fuel_
193
cat /etc/fuel_release:
9.0
cat /etc/fuel_
mitaka-9.0
rpm -qa | egrep 'fuel|astute|
fuel-release-
fuel-misc-
python-
fuel-openstack
fuel-provision
python-
fuel-9.
fuel-nailgun-
rubygem-
fuel-library9.
fuel-agent-
fuel-ui-
fuel-setup-
nailgun-
shotgun-
network-
fuel-bootstrap
fuel-migrate-
fuelmenu-
fuel-notify-
fuel-ostf-
fuel-mirror-
fuel-utils-
Serge Kovaleff (serge-kovaleff) wrote : | #9 |
@Vitaly, do we need the bug fix in MOS 8.updates?
To successful images upload rados should be available.
But as i see in logs, rados was not ready when task upload_images started.
puppet: main]/Main/ Service[ radosgw] /ensure (notice): ensure changed 'stopped' to 'running' 10.109. 54.3:8080/ swift/v1/ glance 503 Service Unavailable
2016-02-15 06:08:00 +0000 /Stage[
glance:
<147>Feb 15 06:08:40 node-4 glance-api: 2016-02-15 06:08:40.911 180754 ERROR swiftclient [-] Container HEAD failed: http://
last error in apache:
[Mon Feb 15 06:08:43.702711 2016] [proxy:error] [pid 64189:tid 139853966554880] AH00940: FCGI: disabled connection for (127.0.0.1)
So rados got available in 3 seconds later then upload task started.
Some validation should be added to task enable_rados.