You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
error detecting the lowest Kubernetes version Get "https://10.5.0.2:6443/api/v1/namespaces/kube-system/pods": dial tcp 10.5.0.2:6443: i/o timeout
#10111
Open
BenHiramTaylor opened this issue
Jan 9, 2025
· 1 comment
I may have discovered a bug with the Talosctl cli, please close this if it is expected local functionality.
TLDR
When using local cluster made from talosctl cluster create --workers 3 --controlplanes 2 --kubernetes-version 1.31.4 running an upgrade-k8s command fails with i/o timeout.
I was playing around with Talos specifically for learning purposes, I am familiar with K8s but wanted to see how Talos would handle K8s upgrades on HA clusters.
I installed the latest talosctl version and create a local cluster in Docker Desktop with the above cluster create command.
All elements of the cluster are showing and running fine as shown below
When I attempt to try an upgrade to 1.32.0 with talosctl --nodes 10.5.0.2 upgrade-k8s --to 1.32.0 (same error with --dry-run)
it fails with the following error
I know that the optimal way to upgrade a local test cluster is to delete and recreate it but I wanted to see the upgrade in action, please close if this is expected functionality.
This is a limitation on Docker non-Linux which doesn't expose container IPs. It's actually a bug in talosctl as well.
As a workaround you can see the exposed endpoint on the localhost in the first Talos contianer port and force talosctl upgrade-k8s to use that with the endpoint like --endpoint=https://127.0.0.1:<exposed port>
Bug Report
Description
I may have discovered a bug with the Talosctl cli, please close this if it is expected local functionality.
TLDR
When using local cluster made from
talosctl cluster create --workers 3 --controlplanes 2 --kubernetes-version 1.31.4
running an upgrade-k8s command fails with i/o timeout.I was playing around with Talos specifically for learning purposes, I am familiar with K8s but wanted to see how Talos would handle K8s upgrades on HA clusters.
I installed the latest talosctl version and create a local cluster in Docker Desktop with the above cluster create command.
All elements of the cluster are showing and running fine as shown below
When I attempt to try an upgrade to 1.32.0 with
talosctl --nodes 10.5.0.2 upgrade-k8s --to 1.32.0
(same error with --dry-run)it fails with the following error
I know that the optimal way to upgrade a local test cluster is to delete and recreate it but I wanted to see the upgrade in action, please close if this is expected functionality.
Logs
talosctl cluster create --workers 3 --controlplanes 2 --kubernetes-version 1.31.4
talosctl --nodes 10.5.0.2 upgrade-k8s --to 1.32.0
error detecting the lowest Kubernetes version Get "https://10.5.0.2:6443/api/v1/namespaces/kube-system/pods": dial tcp 10.5.0.2:6443: i/o timeout
Environment
The text was updated successfully, but these errors were encountered: