-
### Is there an existing issue for this?
- [X] I have searched the existing issues
### Version
higher than v1.16.0-rc.1 and lower than v1.17.0
### What happened?
While adding a new node to the ex…
-
## Describe the bug
I found this while running some other experiments with fast failover and auto-delete of workload pods. In one run, the dead share manager pod wasn't deleted until more than 20 …
-
Affected repositories:
- https://github.com/giantswarm/aws-iptables-legacy-cleaner/pull/10
- https://github.com/giantswarm/awscnfm/pull/497
- https://github.com/giantswarm/aws-operator/pull/3553
-…
-
I have deployed the crate operator using Helm following the documentation https://cratedb.com/docs/guide/install/container/kubernetes/kubernetes-operator.html
and I have run `dev-cluster.yaml`.
but I …
-
### Enhancement Description
- One-line enhancement description (can be used as a release note): Projected service account tokens for authenticated image pulls via kubelet image credential providers…
-
## Describe the bug
When running negative test case `Power Off Node One By Once For More Than Pod Eviction Timeout While Workload Heavy Writing` on longhorn `v1.6.2-rc2`, accidentally encountered a…
-
## Describe the bug
warning from longhorn-manager trying to read /sys/class/block/XXX:
time="2024-09-17T18:19:51Z" level=warning msg="Failed to get filesystem device type of /var/lib/longhorn/" …
-
Currently, the swizzle playbook is installing the kubernetes packages (and prereqs) to the etcd nodes. This is fine when deploying stacked masters, but seems unnecessary when deploying etcd as an exte…
-
## Is this a BUG REPORT or FEATURE REQUEST?
FEATURE REQUEST
## Anything else we need to know?
This issue wants to remove container runtime interface (CRI) socket annotations from Node objects…
-
## Bug Report
After upgrading to 1.7.6. If a pod listen to hostPort, any request to services with the same port number on the same node will be redirected to this hostPort.
### Description
I have…