cherry pick https://review.opendev.org/c/starlingx/rook-ceph/+/783584/ build and deploy failed

Bug #1925285 reported by Yong Fu
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
StarlingX
Invalid
Low
chen haochuan

Bug Description

Brief Description
-----------------
After cherry pick https://review.opendev.org/c/starlingx/rook-ceph/+/783584/, build angd deploying mutil-node(2+2+2 virtual) failed.

engineer build issue.

stx-openstack apply failed.

Reproducibility
---------------
100% reproducible

Tags: stx.4.0
Revision history for this message
Yong Fu (fuyong) wrote :
description: updated
Austin Sun (sunausti)
Changed in starlingx:
importance: Undecided → Low
assignee: nobody → chen haochuan (martin1982)
Revision history for this message
chen haochuan (martin1982) wrote :
Revision history for this message
chen haochuan (martin1982) wrote :

In the attachment, there is no lablel openstack-control-plane on controller-0, which makes openstack apply fail.

Please ensure the host deployment correct. And use "fm alarm-list" to ensure there is no alarm, and host-list to ensure hosts are all available, before apply openstack

Revision history for this message
Yong Fu (fuyong) wrote :
Revision history for this message
chen haochuan (martin1982) wrote :

Please check the snapshot in attachment

Revision history for this message
Weiyuan.Wang-Neu (weiyuan-wang) wrote :

I tried to deploy using the base code from master, and found a problem when config data network for work nodes:

The the names of devices for data network are generally eth1000 and eth1001, but they changed to eth0 and eth1 on work-0 and eth1 and eth2 on work-1 when using the iso built by the base.

So I modified the auto-deploy script and deployed the cherry-pick version ( https://review.opendev.org/c/starlingx/rook-ceph/+/783584/) again.

The result is OK.

I think the reason for the previous failure was that the network device name was incorrect and failed to config the data network.

Changed in starlingx:
status: New → Invalid
Revision history for this message
zhao.shuai (zhao.shuai.neusoft) wrote :

Maybe this is the test method issue, If not see this problem anymore.
Please close this LP ticket. @FuYong. @WangWeiyuan

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.