k8s/watchers: Add missing v1 EndpointSlice group on init

[christarazi]

The K8sWatcher's InitK8sSubsystem() calls resourceGroups() which returns
a list of resources that should be waited on via WaitForCacheSync(). The
v1 version of EndpointSlice was missing from this list. This commit adds
it.

Signed-off-by: Chris Tarazi chris@isovalent.com

[joestringer]

nice catch, looks like this was introduced by one of my refactoring PRs during v1.11 development cycle. v1.10 looks fine in this regard.

Curious also how you noticed this, since CI doesn't appear to catch this condition? (at least not in a reliable way; for all I know this was causing a flake but I don't know).

[nathanjsweet]

good on 🍏

[christarazi]

@joestringer My guess for why we don't see it in CI is because we probably don't disable the original Endpoints resource, so even if the v1.EndpointSlice group is missing from the wait, then we still have to wait on the former. I imagine that if Cilium ran in a cluster with the former completely disabled, then we would have seen some flakes at the startup.

I found it through regular code inspection while working on #17714.

[christarazi]

/test

Job 'Cilium-PR-K8s-1.20-kernel-5.4' failed and has not been observed before, so may be related to your PR:

Click to show.

Test Name

K8sDatapathConfig Iptables Skip conntrack for pod traffic

Failure Output

FAIL: Missing '-j CT --notrack' iptables rule

If it is a flake, comment /mlh new-flake Cilium-PR-K8s-1.20-kernel-5.4 so I can create a new GitHub issue to track it.

Job 'Cilium-PR-K8s-1.21-kernel-4.19' failed and has not been observed before, so may be related to your PR:

Click to show.

Test Name

K8sDatapathConfig Iptables Skip conntrack for pod traffic

Failure Output

FAIL: Missing '-j CT --notrack' iptables rule

If it is a flake, comment /mlh new-flake Cilium-PR-K8s-1.21-kernel-4.19 so I can create a new GitHub issue to track it.

Job 'Cilium-PR-K8s-1.16-net-next' failed and has not been observed before, so may be related to your PR:

Click to show.

Test Name

K8sDatapathConfig Iptables Skip conntrack for pod traffic

Failure Output

FAIL: Missing '-j CT --notrack' iptables rule

If it is a flake, comment /mlh new-flake Cilium-PR-K8s-1.16-net-next so I can create a new GitHub issue to track it.

[christarazi]

/test

[pchaigno]

Reviews are in. Multicluster has the usual failure and k8s-1.21-kernel-4.19 hit known flake #13552. Marking ready to merge and merging.