cluster-api-provider-openstack
cluster-api-provider-openstack copied to clipboard
kubernetes-sigs
HTTPS for Loadbalancer Health Monitor
/kind feature
Describe the solution you'd like
Be able to use HTTPS as type for the load balancer's health monitor (on the /readyz path).
Anything else you would like to add:
Hopefully this would solve problems such as:
- https://github.com/kubernetes-sigs/cluster-api-provider-openstack/issues/1749
Related issues:
- Most-recent improvement (to avoid incidents seen in #1221 and #1374):
- https://github.com/kubernetes-sigs/cluster-api-provider-openstack/pull/1375
- Making load balancer health checks configurable overall (though only TCP settings discussed yet):
- https://github.com/kubernetes-sigs/cluster-api-provider-openstack/issues/1221
- https://github.com/kubernetes-sigs/cluster-api-provider-openstack/pull/1360
Here's for example how OpenShift documents how to configure the load balancer health checks in this manner (and why not to use TCP port probe, but instead the /readyz endpoint):
- https://github.com/openshift/installer/blob/master/docs/dev/kube-apiserver-health-check.md
Though on the topic of setting/syncing the monitor's values with the api server's --shutdown-delay-duration setting, while searching for shutdown-delay-duration in CAPI repo, I found issues like these:
- https://github.com/kubernetes-sigs/cluster-api/issues/2753 (mentioned, but solved other issues)
- https://github.com/kubernetes-sigs/cluster-api/issues/2937
...closed in favor of:
- https://github.com/kubernetes-sigs/cluster-api/issues/1250
...closed in favor of:
- https://github.com/kubernetes-sigs/cluster-api/pull/8500
...merged 2 days ago, aiming to solve:
- https://github.com/kubernetes-sigs/cluster-api/issues/7494
- https://github.com/kubernetes-sigs/cluster-api/pull/8500
...merged 2 days ago, aiming to solve:
- https://github.com/kubernetes-sigs/cluster-api/issues/1250
...closed in favor of:
I think we should also take into account what @mnaser said over in #1374:
I would have liked to change the following too:
type: HTTPSurl_path:/healthzThis will give us actual health checks that involve the API endpoint telling us that it's happy, and it should be OK to accept the unknown cert since openstack/octavia@
8c793f2/octavia/common/jinja/haproxy/split_listeners/templates/macros.j2#L178-L182 is OK with it however OVN doesnt support this so we can keep TCP for now.
For example, here's one current monitor:
And here's a manually drafted one which I'd like to be able to use instead:
@dulek What do you think about using HTTPS health monitors here? There's a question about how to support the OVN provider. Does it support HTTPS?
@dulek What do you think about using HTTPS health monitors here? There's a question about how to support the OVN provider. Does it support HTTPS?
It makes perfect sense from my perspective as long as it's opt-in. The OVN provider does not support HTTP or HTTPS health monitors and most likely never will as it lives in layers below. That shouldn't block us from trying to improve things for Amphora.
The Kubernetes project currently lacks enough contributors to adequately respond to all issues.
This bot triages un-triaged issues according to the following rules:
- After 90d of inactivity,
lifecycle/staleis applied - After 30d of inactivity since
lifecycle/stalewas applied,lifecycle/rottenis applied - After 30d of inactivity since
lifecycle/rottenwas applied, the issue is closed
You can:
- Mark this issue as fresh with
/remove-lifecycle stale - Close this issue with
/close - Offer to help out with Issue Triage
Please send feedback to sig-contributor-experience at kubernetes/community.
/lifecycle stale
The Kubernetes project currently lacks enough contributors to adequately respond to all issues.
This bot triages un-triaged issues according to the following rules:
- After 90d of inactivity,
lifecycle/staleis applied - After 30d of inactivity since
lifecycle/stalewas applied,lifecycle/rottenis applied - After 30d of inactivity since
lifecycle/rottenwas applied, the issue is closed
You can:
- Mark this issue as fresh with
/remove-lifecycle stale - Close this issue with
/close - Offer to help out with Issue Triage
Please send feedback to sig-contributor-experience at kubernetes/community.
/lifecycle stale
The Kubernetes project currently lacks enough active contributors to adequately respond to all issues.
This bot triages un-triaged issues according to the following rules:
- After 90d of inactivity,
lifecycle/staleis applied - After 30d of inactivity since
lifecycle/stalewas applied,lifecycle/rottenis applied - After 30d of inactivity since
lifecycle/rottenwas applied, the issue is closed
You can:
- Mark this issue as fresh with
/remove-lifecycle rotten - Close this issue with
/close - Offer to help out with Issue Triage
Please send feedback to sig-contributor-experience at kubernetes/community.
/lifecycle rotten
The Kubernetes project currently lacks enough contributors to adequately respond to all issues.
This bot triages un-triaged issues according to the following rules:
- After 90d of inactivity,
lifecycle/staleis applied - After 30d of inactivity since
lifecycle/stalewas applied,lifecycle/rottenis applied - After 30d of inactivity since
lifecycle/rottenwas applied, the issue is closed
You can:
- Mark this issue as fresh with
/remove-lifecycle stale - Close this issue with
/close - Offer to help out with Issue Triage
Please send feedback to sig-contributor-experience at kubernetes/community.
/lifecycle stale
The Kubernetes project currently lacks enough active contributors to adequately respond to all issues.
This bot triages un-triaged issues according to the following rules:
- After 90d of inactivity,
lifecycle/staleis applied - After 30d of inactivity since
lifecycle/stalewas applied,lifecycle/rottenis applied - After 30d of inactivity since
lifecycle/rottenwas applied, the issue is closed
You can:
- Mark this issue as fresh with
/remove-lifecycle rotten - Close this issue with
/close - Offer to help out with Issue Triage
Please send feedback to sig-contributor-experience at kubernetes/community.
/lifecycle rotten
The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.
This bot triages issues according to the following rules:
- After 90d of inactivity,
lifecycle/staleis applied - After 30d of inactivity since
lifecycle/stalewas applied,lifecycle/rottenis applied - After 30d of inactivity since
lifecycle/rottenwas applied, the issue is closed
You can:
- Reopen this issue with
/reopen - Mark this issue as fresh with
/remove-lifecycle rotten - Offer to help out with Issue Triage
Please send feedback to sig-contributor-experience at kubernetes/community.
/close not-planned
@k8s-triage-robot: Closing this issue, marking it as "Not Planned".
In response to this:
The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.
This bot triages issues according to the following rules:
- After 90d of inactivity,
lifecycle/staleis applied- After 30d of inactivity since
lifecycle/stalewas applied,lifecycle/rottenis applied- After 30d of inactivity since
lifecycle/rottenwas applied, the issue is closedYou can:
- Reopen this issue with
/reopen- Mark this issue as fresh with
/remove-lifecycle rotten- Offer to help out with Issue Triage
Please send feedback to sig-contributor-experience at kubernetes/community.
/close not-planned
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.
