Bug 1766851: Add migration for affinity and deployment strategy #321
Conversation
openshift-ci-robot
added
the
bugzilla/valid-bug
|
@Miciah: This pull request references Bugzilla bug 1766851, which is valid. The bug has been moved to the POST state. The bug has been updated to refer to the pull request using the external bug tracker. DetailsIn response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
openshift-ci-robot
added
the
size/M
openshift-ci-robot
added
the
approved
|
Provisioning errors. |
|
Works in manual testing. |
|
Tricky one, thanks! /lgtm |
|
openshift/origin#24085 should have fixed the test flake. /test e2e-aws |
When updating a router deployment, check whether the deployment strategy is changing; if it is, update the affinity policy first, and only update the deployment strategy once all pods have the new affinity policy. This two-step migration is necessary because an old affinity policy may have an anti-affinity rule that prohibits colocation of pods belonging to the same ingresscontroller, the new deployment strategy may require surge, and the presence of pods with the old affinity policy would block a rolling update using the new deployment strategy. Follow-up to commit 0f6fd1c. This commit fixes bug 1766851. https://bugzilla.redhat.com/show_bug.cgi?id=1766851 * pkg/operator/controller/ingress/deployment.go (updateRouterDeployment): If the deployment strategy is changing, make sure to update the affinity policy first.
Miciah
force-pushed
the
BZ1766851-add-migration-for-affinity-and-deployment-strategy
branch
from
50fcaf9 to
c31f498
Compare
|
/lgtm |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: ironcladlou, Miciah 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 |
|
@Miciah: All pull requests linked via external trackers have merged. Bugzilla bug 1766851 has been moved to the MODIFIED state. DetailsIn response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
5 participants
When updating a router deployment, check whether the deployment strategy is changing; if it is, update the affinity policy first, and only update the deployment strategy once all pods have the new affinity policy.
This two-step migration is necessary because an old affinity policy may have an anti-affinity rule that prohibits colocation of pods belonging to the same ingresscontroller, the new deployment strategy may require surge, and the presence of pods with the old affinity policy would block a rolling update using the new deployment strategy.
Follow-up to #280.
pkg/operator/controller/ingress/deployment.go(updateRouterDeployment): If the deployment strategy is changing, make sure to update the affinity policy first.