-
Notifications
You must be signed in to change notification settings - Fork 87
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
designate: allow worker on cluster. #2104
Conversation
either way admins could have them deployed on multiple nodes without necessarily calling it HA |
Feel free to review but please do not merge, yet. We'll give this a spin in a QA environment first. |
If it can already be deployed individually to multiple nodes why does it need to be part of a cluster? |
@cmurphy count is set to 1, so only one node can be assigned atm |
It makes sense to have it running on all cluster nodes if an HA cluster is available (it is just a RabbitMQ driven backend server after all). This pull request is just a shortcut for convenience. |
This reverts commit 64e7ad9. We've are very close to the release and have multiple pull requests still open for Designate: * crowbar#2095 * crowbar#2104 * crowbar#2105 * SUSE-Cloud/doc-cloud#931 There are way too many question marks on this so I am switching the barclamp back to invisible mode. Lets get these pull requests squared away carefully and without the rush of imminent release.
Am I reading correctly that the service will use If yes, this needs to be fixed and systemd needs to be used instead. You can compare with how it's done for other services (like glance), which has a similar code structure for the cookbook. |
It should be on all cluster nodes right, unless Designate can tolerate single-point-of-failure for the workers node? fwiw, looking at the Ardana input models, it appears Designate workers are running on all the clustered nodes. |
[1] suggests that workers are there for designate-central to get work done. So having more of them is a good idea. A worker does not really have an idea of api so no vip or haproxy config is required. 1 - https://docs.openstack.org/designate/rocky/contributor/architecture.html
Author: Vincent Untz <[email protected]>
5223875
to
6ee734c
Compare
@guangyee thats what the code does now. |
This reverts commit 64e7ad9. We've are very close to the release and have multiple pull requests still open for Designate: * crowbar#2095 * crowbar#2104 * crowbar#2105 * SUSE-Cloud/doc-cloud#931 There are way too many question marks on this so I am switching the barclamp back to invisible mode. Lets get these pull requests squared away carefully and without the rush of imminent release.
Is it expected that the ha_want_designate job is failing? |
SUSE-Cloud/automation#3443 and https://jira.suse.com/browse/SOC-9342 is needed for tempest (smoke) test to pass. |
[1] suggests that workers are there for designate-central to get work
done. So having more of them is a good idea.
A worker does not really have an idea of api so no vip or haproxy
config is required.
1 - https://docs.openstack.org/designate/rocky/contributor/architecture.html