-
Notifications
You must be signed in to change notification settings - Fork 2
Add cilium support in Kubespray #28
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
8c569c2 to
8a61324
Compare
8a61324 to
d331994
Compare
d331994 to
e264e4a
Compare
|
Question: Did you use |
e264e4a to
a294ea2
Compare
I used |
You should probably switch to the release branch in that case since that's where this will be used |
a294ea2 to
2e2a6fc
Compare
Switched back to the release-2.27.0-ck8s branch as base. |
2e2a6fc to
1d70e40
Compare
|
Opened kubernetes-sigs#12408 for upstreaming the generic hooks required for this task. |
Refactor Cilium CNI installation
Give users two options: besides skip Cilium, add `cilium_remove_old_resources`, default is `false`, when set to `true`, it will remove the content of the old version, but it will cause the downtime, need to be careful to use. Signed-off-by: ChengHao Yang <[email protected]>
`cilium install` is equivalent to `helm install`, it will failed if cilium relase exist. `cilium version` can know the release exist without helm binary Signed-off-by: ChengHao Yang <[email protected]>
1d70e40 to
f6503cd
Compare
Apparently there is another PR in the upstream pipes that will allow for generic helm values for cilium: kubernetes-sigs#12375 so we can take advantage of that instead of using hooks. |
|
Closing in favor of #30 which consists entirely of cherry-picks and backports. |
PoC that Cilium is a viable CNI plugin for Kubespray, but boy there were a lot of hoops to jump through:
cilium-clifor installation, which internally useshelmwhich means we finally get access to avalues.yamltemplate which we can modify. This is included in thev2.28.0upstream release.- Added extra values sections to enable the "policy audit mode" (because our built-in network policies were blocking traffic to the Kube API server and all hell broke loose - need further investigation on why this happens on apps apply).- Added extra values section to enableServiceMonitorsfor the cilium agent pods and operator pods so the pretty Grafana dashboards added by Haorui are now showing pretty graphs.What type of PR is this?
What this PR does / why we need it:
Which issue(s) this PR fixes:
Fixes https://github.com/elastisys/ck8s-issue-tracker/issues/535
Special notes for your reviewer:
Worth knowing that the first few commits in the PR are cherry-picks, so not really in scope for change requests.
The only deviation from upstream (with high chances of getting upstreamed) is in this commit: 1d70e40
Also, might be worth reviewing in tandem with: elastisys/compliantkubernetes-kubespray#445
Does this PR introduce a user-facing change?: