This repository has been archived by the owner on Jul 3, 2021. It is now read-only.
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Outdated suggestions cannot be applied.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.
What this PR does / why we need it:
Configure docker daemon with standard networking settings (docker0 bridge), letting Kubelet/CNI manage cni0 bridge.
With this change the cni0 configuration is no longer kept in the docker store (when persistent disks are attached to the worker nodes) so we solve the problem of stale data sticking around when flannel leases expire, and it should also make it easier to keep cni0 in sync with flannel.1
Idea credit and co-author: @alekssaul
How can this PR be verified?
Applying the Ops-file including in the changes and
Is there any change in kubo-release?
Not at this time
Is there any change in kubo-ci?
Not required, but recommended when approved
Does this affect upgrade, or is there any migration required?
No impact.
Which issue(s) this PR fixes:
When flannel lease expire and flannel.1 and cni0 go out of sync, rebooting/recreating the worker nodes doesn't solve the problem but make it worse, because the old/stale subnet info is kept in the docker store (
/var/vcap/store/docker/docker/network/files/local-kv.d
) and that is still used to configure cni0 even after recreating the worker node. With this fix, docker no longer owns cni0 so we can rely on rebootings/bosh-recreate to recover worker nodes out of syncRelease note:
NONE