Pool update removes PTR records
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Designate |
Fix Released
|
Critical
|
Erik Olof Gunnar Andersson |
Bug Description
I am having some issue with OpenStack Designate that whenever I use the 'designate-manage update pool' command my PTR records are unset but the actual records stay active in the reverse zone. I have tried finding the problem but sadly I cannot pinpoint where and why to problem occurs.
The screenshots below are from our development environment deployed using Kolla-Ansible. But the same happens within a new devstack environment. The problems occurs in the versions "stable/2023.1" and "stable/yoga" I have reproduced this using Devstack and using a Kolla-Ansible deployment in "stable/2023.1".
Screenshots:
(ptr record list) Before designate-manage update pool:
https:/
(ptr record list) After designate-manage update pool:
https:/
(record list) Records in zone after designate-manage update pool:
https:/
How to reproduce in DevStack?
1. Install DevStack with OpenStack designate installed the Local.conf I used below:
[[local|localrc]]
ADMIN_
DATABASE_
RABBIT_
SERVICE_
HOST_IP=
SERVICE_
# Enable designate
enable_plugin designate https:/
enable_service designate,
2. After devstack has been installed create a Floating IP:
openstack floating ip create public
3. Now set a PTR record for the Floating IP
openstack ptr record set <Region (RegionOne)
4. Wait until the record has been set as "ACTIVE" in the reverse zone (don't know if this is required of not)
openstack record list --all <reverse_zone_id>. (Use --all because the zone is place under the 00000000-
5. Update the designate pool using the command: designate-manage update pool With or without the pools.yaml file has the same effect.
designate-manage pool update [--file /etc/designate/
6. Wait for the designate-manage pool update to finish after which the PTR record will be shown as empty (None) using the ptr record list command even tough they are still listed in the zone when u use the record list command again
openstack ptr record list
openstack record list --all <reverse_zone_id>
It almost seems like the records are removed from the database and are then are put back with a new ID and without the ptr:floatingip and managed_
If any more information is required or if I can help in any way please let me know.
Changed in designate: | |
importance: | Undecided → Critical |
assignee: | nobody → Erik Olof Gunnar Andersson (eandersson) |
Fix proposed to branch: master /review. opendev. org/c/openstack /designate/ +/894898
Review: https:/