Deployment is failed on 'upload_cirros' task if ceph selected for objects but not for images
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Fuel for OpenStack |
Invalid
|
High
|
Sergii Rizvan |
Bug Description
Detailed bug description:
Deployment is failed if ceph selected for objects but not for images.
Steps to reproduce:
1. Deploy cluster where ceph selected for objects but not for images.
Issue is observed for any combination of Storage Backends:
Ceph RadosGW for objects (Swift API)
Ceph RBD for ephemeral volumes (Nova)
or
Ceph RadosGW for objects (Swift API)
Ceph RBD for volumes (Cinder)
or
Ceph RadosGW for objects (Swift API)
Ceph RBD for ephemeral volumes (Nova)
Ceph RBD for volumes (Cinder)
Expected results:
Deployment is finished successfully.
Actual results:
"Deployment has failed. Method granular_deploy. Failed to execute hook 'upload_cirros' Failed to run command cd / && ruby /etc/puppet/
uids:
- '8'
parameters:
retries: 3
cmd: ruby /etc/puppet/
cwd: /
timeout: 180
interval: 20
priority: 1300
fail_on_error: true
type: shell
id: upload_cirros
Errors from glance-api log: http://
If try to run it manually:
root@node-8:~# ruby /etc/puppet/
Auth URL is http://
/usr/bin/glance --os-image-
500 Internal Server Error: Failed to upload image 616cff55-
Image 'TestVM' upload from '/usr/share/
Reproducibility:
Always
Workaround:
Not Found
Description of the environment:
FUEL/MOS 8.0, 8.0 + MU4 updates
1 controller with Ceph OSD + 2 compute nodes with Ceph OSD
3 controllers with Ceph OSD + 1 compute node
Changed in fuel: | |
milestone: | none → 8.0-updates |
Changed in fuel: | |
assignee: | nobody → Sergii Rizvan (srizvan) |
Changed in fuel: | |
importance: | Undecided → High |
Diagnostic snapshots: /drive. google. com/file/ d/0Bw14Lq82YyPj bXdGcHl2cnlCaDA /drive. google. com/file/ d/0Bw14Lq82YyPj MXhhZGVQLXh3ZVk
https:/
https:/