-
Notifications
You must be signed in to change notification settings - Fork 462
Add extra filter for checking if registries have changed #461
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
Conversation
|
/test e2e-aws |
|
/test e2e-aws |
|
/test e2e-aws-op |
|
/test images |
|
FTR: Lots of CI flakes with 504 gateway timeouts hence the retests. Also these seem to be known flakes being worked on. See https://github.com/openshift/release/issues/2905 Taking a look/testing locally. |
|
Given the below comment, disregard |
|
/hold discussed offline, we'll need something lighter and more state oriented |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Drop this
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
woops, dropping.
|
@runcom updated code as discussed on slack. |
|
/hold cancel /assign kikisdeliveryservice |
|
hi @umohnani8 since the docs are pretty thin on using this feature and im less familiar with it, can you please add step by step instructions on how to test this? |
|
@kikisdeliveryservice this is what an example image CR looks like with insecure and blocked registries in it You can add as many or as little you want there and should see the changes on all the nodes once the controller does its magic :) |
|
Looks like the test flakes being fixed in #457 |
|
thanks @umohnani8 :) will try this out. |
|
/assign kikisdeliveryservice |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Let's call this applied
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
done
A resync happens about every 20 minutes, which sends an updated event to the image informer even if nothing in the image CR has changed. Adding and extra filter that checks if there has been any changes to the registries part of the image CR before syncing the image handler again. Signed-off-by: Urvashi Mohnani <[email protected]>
|
Testing results:
Checking MCD: new config was successfully applied and daemons rebooted. |
|
/lgtm |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: kikisdeliveryservice, runcom, umohnani8 The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:
Approvers can indicate their approval by writing |
|
Flakes: /retest |
|
/retest Please review the full test history for this PR and help us cut down flakes. |
1 similar comment
|
/retest Please review the full test history for this PR and help us cut down flakes. |
|
/retest |
|
/retest Please review the full test history for this PR and help us cut down flakes. |
|
CI errors tracked in slack /retest |
The CRC (container runtime config) controller recently added a check to avoid resyncing and recreating the very same registries config if nothing has changed on the image crd side [1]. While that's correct, during an upgrade, the controllers need to generate the MC fragments using the controller version they're at. Since we weren't checking the versions of the controller that generated the registries config, we wrongly assumed the configurations were equal and never generated a new one with the new controller. This patch fixes that by adding a version check before skipping a regeneration on equal content in the registries configs. Fixes: openshift#487 [1] openshift#461 Signed-off-by: Antonio Murdaca <[email protected]>
A resync happens about every 20 minutes, which sends an updated
event to the image informer even if nothing in the image CR has
changed. Adding and extra filter that checks if there has been
any changes to the registries part of the image CR before syncing
the image handler again.
Helps fix #453 where changes to the image config will not be applied unless and actual change happened in the CR.
Signed-off-by: Urvashi Mohnani [email protected]