Update dependency projectcalico/calico to v3.26.3

[renovate]

This PR contains the following updates:

Package	Update	Change
projectcalico/calico	patch	v3.26.1 -> v3.26.3

---

Release Notes

projectcalico/calico (projectcalico/calico)

v3.26.3

Compare Source

Release notes can be found on GitHub

Attached to this release are the following artifacts:

• release-v3.26.3.tgz: container images, binaries, and kubernetes manifests.
• calico-windows-v3.26.3.zip: Calico for Windows.
• tigera-operator-v3.26.3.tgz: Calico Helm v3 chart.
• ocp.tgz: Manifest bundle for OpenShift.

Additional links:

• VPP data plane release information

v3.26.2

Compare Source

Release notes can be found on GitHub

Attached to this release are the following artifacts:

• release-v3.26.2.tgz: container images, binaries, and kubernetes manifests.
• calico-windows-v3.26.2.zip: Calico for Windows.
• tigera-operator-v3.26.2.tgz: Calico Helm v3 chart.
• ocp.tgz: Manifest bundle for OpenShift.

Additional links:

• VPP data plane release information

---

Configuration

📅 Schedule: Branch creation - At any time (no schedule defined), Automerge - At any time (no schedule defined).

🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.

♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.

🔕 Ignore: Close this PR and you won't be reminded about this update again.

---

• If you want to rebase/retry this PR, check this box

---

This PR has been generated by Mend Renovate. View repository job log here.

use_cilium = "false"

[DEiselt]

Because cilium is used by default and this is related to calico, thus i will test this manually instead of using zuul.

[DEiselt]

This also waits for #610 as discussed in today's container meeting.

[matofeder]

@DEiselt SCS Zuul has been temporarily unavailable, hence I re-labeled this PR with the e2e-quick-test label. Now the e2e test works.

EDIT: e2e test executed again. The test executed above failed, because Management cluster with prefix pr623-67d7ab already exists. This is not supported due to unexpected side-effects.. This was probably caused by the Zuul hiccups I mentioned.

[scszuulapp]

Build failed (e2e-quick-test pipeline).
https://zuul.scs.community/t/SCS/buildset/26d4980fd99c4fee9296ceafacb07841

❌ k8s-cluster-api-provider-e2e-quick FAILURE in 12m 01s

Warning:

Custom ENV config

use_cilium = "false"

[scszuulapp]

Build failed (e2e-quick-test pipeline).
https://zuul.scs.community/t/SCS/buildset/8857f34878ff45d59ea5c2fef1ef0a2e

❌ k8s-cluster-api-provider-e2e-quick FAILURE in 5m 12s

Warning:

Custom ENV config

use_cilium = "false"

[DEiselt]

@matofeder can you help me with this? The run shows

Management cluster with prefix pr623-67d7ab already exists. This is not supported due to unexpected side-effects.
make: *** [Makefile:87: create] Error 1

I suspect this means that some cleanup is missing / has failed but i don't know how to fix it in zuul.

[scszuulapp]

Build failed (e2e-quick-test pipeline).
https://zuul.scs.community/t/SCS/buildset/95c1de2741364517b51a23883d81c93a

❌ k8s-cluster-api-provider-e2e-quick FAILURE in 9m 20s

Warning:

Custom ENV config

use_cilium = "false"

[matofeder]

@matofeder can you help me with this? The run shows

Management cluster with prefix pr623-67d7ab already exists. This is not supported due to unexpected side-effects.
make: *** [Makefile:87: create] Error 1

I suspect this means that some cleanup is missing / has failed but i don't know how to fix it in zuul.

@DEiselt sorry for the delay. @chess-knight is on it. He let you know once the CI project is without any leftovers.
It seems that Zuul's hiccups caused the issues we have with e2e.

[scszuulapp]

Build failed (e2e-quick-test pipeline).
https://zuul.scs.community/t/SCS/buildset/c169765a8f5049d9a3143f5bbe3c215e

❌ k8s-cluster-api-provider-e2e-quick FAILURE in 6m 08s

Warning:

Custom ENV config

use_cilium = "false"

[scszuulapp]

Build failed (e2e-quick-test pipeline).
https://zuul.scs.community/t/SCS/buildset/fe7ad81fb93145abb3bbc416748ae05d

❌ k8s-cluster-api-provider-e2e-quick FAILURE in 5m 17s

Warning:

Custom ENV config

use_cilium = "false"

[scszuulapp]

Build failed (e2e-quick-test pipeline).
https://zuul.scs.community/t/SCS/buildset/2272f2d5a1854a43b0d8af1b50123159

❌ k8s-cluster-api-provider-e2e-quick FAILURE in 4m 07s

Warning:

Custom ENV config

use_cilium = "false"

[chess-knight]

Build failed (e2e-quick-test pipeline). https://zuul.scs.community/t/SCS/buildset/2272f2d5a1854a43b0d8af1b50123159

❌ k8s-cluster-api-provider-e2e-quick FAILURE in 4m 07s

Warning:

Custom ENV config

use_cilium = "false"

Something very strange happened. According to the task summary, we can clearly see that make create failed:

make create GITREFERENCE=renovate/projectcalico-calico-3.x GITREPO=https://github.com/SovereignCloudStack/k8s-cluster-api-provider.git
non-zero return code
Management cluster with prefix pr623-e18fc0 already exists. This is not supported due to unexpected side-effects.
make: *** [Makefile:87: create] Error 1

But according to the logs we can see that terraform apply run:

...
openstack_compute_instance_v2.mgmtcluster_server: Creating...
...
openstack_compute_instance_v2.mgmtcluster_server: Still creating...
openstack_compute_instance_v2.mgmtcluster_server: Creation complete after 13s [id=d982b3c3-6676-46b2-be0b-70d2fc7b80f5]

But then:

[Zuul] Log Stream did not terminate
ubuntu-jammy | ERROR
ubuntu-jammy | {
ubuntu-jammy |   "delta": "0:00:02.965970",
ubuntu-jammy |   "end": "2023-10-30 13:23:10.345649",
ubuntu-jammy |   "msg": "non-zero return code",
ubuntu-jammy |   "rc": 2,
ubuntu-jammy |   "start": "2023-10-30 13:23:07.379679"
ubuntu-jammy | }

So I assume there is some second run which failed but the logs that we see are from the first run.

[matofeder]

@DEiselt After further investigation and discussions in Matrix chat SCS|Zuul we suspect that this issue is connected to the WaveStack infrastructure issues. Unfortunately, it seems that infra issues are not resolved yet, therefore it seems that we are not able to run the e2e test using SCS Zuul.

As a result of the temporarily unhealthy SCS Zuul instance, I removed the failed-e2e-quick-test label and I would suggest doing the review of this PR in an "old" way.

[chess-knight]

LGTM
TESTED