-
Notifications
You must be signed in to change notification settings - Fork 526
Rework ha-single-zone
kustomizations (move to /dev-setup
)
#12345
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
Rework ha-single-zone
kustomizations (move to /dev-setup
)
#12345
Conversation
cb89ad7
to
d4852bf
Compare
/assign |
086ffd4
to
2d321cc
Compare
/assign |
2d321cc
to
11d5236
Compare
afca986
to
7b03eed
Compare
fb2bd6b
to
f01d743
Compare
That was easy 😅 /lgtm |
This way, `make operator-up` also works out-of-the-box. Next step: Make `make operator-seed-up` work, i.e., registering a `Garden` and a `Gardenlet` for this new multi-node (3 nodes in the same zone) setup.
TODOs: - remove ha-single-zone from skaffold.yaml (and probably many other places) - validate that zones must be unique in Garden and Seed
Otherwise, we might run only 2 replicas, but we have 3 nodes. If none of the `istio-ingressgateway` pods runs on the control plane node (which has the port mapping `172.18.255.3:443->31443/tcp`), the virtual garden cannot be contacted. ``` $ docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES ab387f3ecc27 kindest/node:v1.32.5 "/usr/local/bin/entr…" 8 minutes ago Up 8 minutes 0.0.0.0:5001->5001/tcp, 127.0.0.1:62613->6443/tcp, 172.18.255.1:5353->30053/tcp, 172.18.255.1:443->30443/tcp, 172.18.255.3:443->31443/tcp, 172.18.255.1:8132->32132/tcp gardener-operator-local-control-plane 27257fee5eab kindest/node:v1.32.5 "/usr/local/bin/entr…" 8 minutes ago Up 8 minutes gardener-operator-local-worker2 c7da6eec61f6 kindest/node:v1.32.5 "/usr/local/bin/entr…" 8 minutes ago Up 8 minutes gardener-operator-local-worker ``` For whatever reason, traffic is not forwarded to the `istio-ingressgateway` pods running on the worker nodes. This commit is a temporary workaround and should be reverted later on when the actual networking problem is fixed. 🎁 ScheererJ 🙈
…one-{up,down}` In the next commit, we will rename `kind-ha-single-zone-{up,down}`
This way, `gardener-operator` running in first kind cluster deploys the gardenlet to the second kind cluster.
This way, `gardener-operator` running in the first kind cluster will be able to talk to the second kind cluster.
Also, the `gardener-operator` folder is no longer used within `dev/local-backupbuckets` now that the provider extension manages the `BackupBucket` resource. It creates new folders with `garden-<uid>`, so let's rather cleanup those.
more consistency
- GOP deletes the kubeconfig secret after successful installation of `gardenlet` - However, the e2e tests need a kubeconfig secret for the seed in order to be able to communicate with it - Hence, let's simply create a second secret just for `gardener-operator`'s installation process
The e2e tests still require the original server, see https://prow.gardener.cloud/view/gs/gardener-prow/pr-logs/pull/gardener_gardener/12345/pull-gardener-e2e-kind-migration-ha-single-zone/1937417741346541568
Ref: https://prow.gardener.cloud/view/gs/gardener-prow/pr-logs/pull/gardener_gardener/12345/pull-gardener-e2e-kind-migration-ha-single-zone/1937573427510513664 Also, the `Gardenlet` resource itself is already deleted by the `skaffold delete`, so no need to delete it again.
55940f6
to
f65b22f
Compare
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.
/lgtm
/approve
LGTM label has been added. Git tree hash: 96de496f8ed2d2cedf59eec30107ee53744da12b
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: timuthy The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
How to categorize this PR?
/area dev-productivity
/kind cleanup
What this PR does / why we need it:
This PR reworks the dev/e2e setup for
ha-single-zone
in the/dev-setup
folder.gardener-operator
setup.gardenlet
in the second kind cluster is now also deployed via theseedmanagement.gardener.cloud/v1alpha1.Gardenlet
resource viagardener-operator
.istio-ingressgateway
running on the control plane node. Our workaround increases the replicas to3
such that at least one replica gets scheduled to the control plane node (due to the spread constraints). Eventually, we should revert it and rather fix the actual networking problem. cc @ScheererJWhich issue(s) this PR fixes:
Part of #11958
Special notes for your reviewer:
/cc @oliver-goetz
Release note: