Backup & Restore after platfoprm upgrade failing
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
StarlingX |
Fix Released
|
Low
|
Erickson Silva de Oliveira |
Bug Description
Brief Description
-----------------
After upgrade successufuly from stx.7.0 to stx.8.0, the backup & restore didn't work properly.
Severity
--------
Minor
Steps to Reproduce
------------------
Upgrade stx.7.0 to stx.8.0
Run backup & restore after finishing the upgrade and the system is operational.
Expected Behavior
------------------
Restore works properly.
Actual Behavior
----------------
Ansible returns an error.
Reproducibility
---------------
Reproduced two times.
System Configuration
-------
AIO-DX DC with multipath and subcloud 2+2+2
Timestamp/Logs
--------------
TASK [recover-ceph-data : Add controller monitor into monmap file] *******
Thursday 07 December 2023 20:10:55 +0000 (0:00:00.333) 0:46:13.219 *****
fatal: [localhost]: FAILED! => changed=true
cmd:
- monmaptool
- --add
- controller
- /tmp/monmap
delta: '0:00:00.014435'
end: '2023-12-07 20:10:55.464669'
msg: non-zero return code
rc: 255
start: '2023-12-07 20:10:55.450234'
stderr: 'monmaptool: invalid ip:port ''/tmp/monmap'''
stderr_lines: <omitted>
stdout: ''
stdout_lines: <omitted>
Changed in starlingx: | |
status: | New → In Progress |
Changed in starlingx: | |
importance: | Undecided → Low |
tags: | added: stx.9.0 stx.storage |
Changed in starlingx: | |
assignee: | nobody → Erickson Silva de Oliveira (esilvade) |
Reviewed: https:/ /review. opendev. org/c/starlingx /stx-puppet/ +/905266 /opendev. org/starlingx/ stx-puppet/ commit/ b17f59ebcb913e1 625390637025e43 95bfd2d257
Committed: https:/
Submitter: "Zuul (22348)"
Branch: master
commit b17f59ebcb913e1 625390637025e43 95bfd2d257
Author: Erickson Silva de Oliveira <email address hidden>
Date: Wed Jan 10 16:35:02 2024 -0300
Add 'public_addr' in ceph.conf by default
After upgrading the platform on AIO-DX, the ceph configuration ceph.conf) is missing the "public_addr" field,
file (/etc/ceph/
which causes the playbook to fail when performing backup & restore.
The problem is caused because this field is defined in the ceph
puppet, only in the monitor configuration, and during the upgrade
it does not enter this part, precisely because the host being
upgraded is not the active controller.
To resolve this, this field is defined together with the
"host" field in the initial ceph configuration.
PASS: (AIO-DX) Upgrade stx.7.0 to stx.8.0
PASS: Check that public_addr is present in ceph.conf on
both controllers.
PASS: Run backup & restore
Closes-Bug: 2049047
Change-Id: Ibac139fdf6d86d 44f8efca1ca96f5 123d639f77c
Signed-off-by: Erickson Silva de Oliveira <email address hidden>