Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
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.
isn't the right fix to make
kind get nodes
work with podman?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.
podman works with kind get nodes, it logs the experimental disclaimer ... podman is not stable enough and still lacks some important capabilities #1778
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.
There's some small discussion on the linked issue that highlights why the current
kind get nodes
output is a challenge with podman.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.
@wbrefvem I didn't realize, but the script does not rely on kubectl commands, since you can not guarantee you are targeting the right cluster
Since all kind nodes have a
-control-plane
or-worker
suffix we may use that inside the loop to skip those, WDYT @BenTheElderThere 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.
oh, my bad again, it does rely on
kubectl
at the end of the script ... nevermind, let me reassing this to @BenTheElder , he knows better this script/assign @BenTheElder
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.
It should be stderr, which should not be captured:
kind/pkg/internal/runtime/runtime.go
Line 16 in 0483056
kind/pkg/cmd/logger.go
Lines 31 to 36 in 0483056
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.
kind/pkg/cluster/internal/providers/podman/provider.go
Line 45 in 0483056
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.
Are we sure the script is broken?
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.
For podman running on macOS, yes, but I may have misunderstood the reason. Here's what I'm seeing when the script starts to
docker exec
the nodes:@BenTheElder you're right that the two lines about podman are not going to stdout. It must be a race condition, and since my solution involves an API call, it's slow enough to lose the race.
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.
Interesting, we really should be safe to exec once
kind create cluster
succeeds ... uhhhDid the container crash or something?
Let's follow up more on the issue.