Skip to content

Conversation

@mtrmac
Copy link
Contributor

@mtrmac mtrmac commented Mar 11, 2019

- What I did

Replaced the symlink added in #535 with using a CRI-O config option introduced by
cri-o/cri-o#2115 .

Note that this requires CRI-O ≥ 1.12.9 / ≥ 1.13.2 .

- How to verify it

I hope the CI will do that? I honestly have no idea what I am doing.

All I verified was that the unit tests work (with updates to the expected results).

- Description for the changelog

@openshift-ci-robot openshift-ci-robot added the size/M Denotes a PR that changes 30-99 lines, ignoring generated files. label Mar 11, 2019
@kikisdeliveryservice
Copy link
Contributor

cc: @cgwalters - Thoughts?

@runcom
Copy link
Member

runcom commented Mar 11, 2019

Thanks Miloslav for putting this up so quickly. Context here is that the symlink option is a stopgap but CRI-O added an option to directly speficy the Auth file to look for when pulling the pause image.

@cgwalters
Copy link
Member

What I did to test the other PR was to create a separate PR with a test commit. Cherry picking this commit onto another PR should be enough to get a good test.

@mtrmac mtrmac force-pushed the pause_image_auth_file branch from d2eb549 to 142b068 Compare March 12, 2019 14:27
@mtrmac mtrmac mentioned this pull request Mar 12, 2019
@openshift-ci-robot openshift-ci-robot added the needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. label Mar 16, 2019
... to include ImageConfig.PauseImageAuthFile, so that
updateCRIOConfig does not remove the value.

Signed-off-by: Miloslav Trmač <[email protected]>
@mtrmac mtrmac force-pushed the pause_image_auth_file branch from 142b068 to 4e82446 Compare March 19, 2019 18:47
@openshift-ci-robot openshift-ci-robot removed the needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. label Mar 19, 2019
mtrmac added 2 commits March 19, 2019 19:48
The path is hard-coded here as well as in the templates that
create the file, in templates/*/00-*/_base/files/pull-secret.yaml .

Signed-off-by: Miloslav Trmač <[email protected]>
This reverts commit 8b08dfb.

We now configure CRI-O to use the Kubelet's path explicitly, so this
symlink should no longer be necessary.

Signed-off-by: Miloslav Trmač <[email protected]>
@mtrmac
Copy link
Contributor Author

mtrmac commented Mar 19, 2019

/retest

1 similar comment
@kikisdeliveryservice
Copy link
Contributor

/retest

@cgwalters
Copy link
Member

/approve

OK based on the results in #542 this
/lgtm

@openshift-ci-robot openshift-ci-robot added the lgtm Indicates that a PR is ready to be merged. label Mar 22, 2019
@openshift-ci-robot
Copy link
Contributor

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: cgwalters, mtrmac

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Details Needs approval from an approver in each of these files:

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

@openshift-ci-robot openshift-ci-robot added the approved Indicates a PR has been approved by an approver from all required OWNERS files. label Mar 22, 2019
@openshift-merge-robot openshift-merge-robot merged commit ab6b97f into openshift:master Mar 22, 2019
@mtrmac mtrmac deleted the pause_image_auth_file branch March 22, 2019 22:27
runcom added a commit to runcom/machine-config-operator that referenced this pull request Mar 26, 2019
This PR is further fixing https://bugzilla.redhat.com/show_bug.cgi?id=1677198 by
allowing an old cluster to contain links (coming from openshift#535) which are
then removed when upgrading to post openshift#540.

What's happening to QE is that:

1) they're starting a cluster with an MCO version which contains openshift#535
2) they're upgrading to a payload which doesn't have openshift#535 cause it has openshift#540 which reverts openshift#535

The above means that point 1) generates MachineConfigs with an unsupported symlink and when
upgrading to 2), the symlink is removed causing drift and an unreconcilable error.

QE can start testing with a newer MCO version to avoid this, but for
ease of testing, let's add this snippet to make sure getting rid of
links, which we don't support anyway, just works.

Signed-off-by: Antonio Murdaca <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

approved Indicates a PR has been approved by an approver from all required OWNERS files. lgtm Indicates that a PR is ready to be merged. size/M Denotes a PR that changes 30-99 lines, ignoring generated files.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

6 participants