k3s: K3s Server fails to run on fresh Fedora 33 install

Environmental Info: Fedora 33 (workstation) K3s version 1.20.0+k3s2 (2ae6b163)

Describe the bug: k3s service endlessly flaps

Steps To Reproduce:

Startup Fedora 33
Install selinux policies etc. (per rancher docs)
Install k3s (sudo curl -sfL https://get.k3s.io | K3S_KUBECONFIG_MODE=“644” sh -)

Expected behavior: The service would start and the master node would be available

Actual behavior: The k3s server starts/stops endlessly until the user intervenes

Logs: These are seen initially once the service dies for the first time:

msg="Cluster-Http-Server 2021/01/09 23:22:32 http: TLS handshake error from 127.0.0.1:50154: remote error: tls: bad certificate"
msg="Cluster-Http-Server 2021/01/09 23:22:32 http: TLS handshake error from 127.0.0.1:50160: remote error: tls: bad certificate"
msg="runtime core not ready"
msg="Failed to retrieve agent config: https://127.0.0.1:6443/v1-k3s/serving-kubelet.crt: 500 Internal Server Error"

After these messages, many general Go errors are visible.

About this issue

Original URL
State: closed
Created 3 years ago
Reactions: 2
Comments: 18 (6 by maintainers)

Most upvoted comments

@T0MASD That is not a valid release version; it’s a dev build. You can install builds from CI using the INSTALL_K3S_COMMIT variable, for example to install the current master build you can do: curl -sfL https://get.k3s.io | INSTALL_K3S_EXEC="server --token=token" INSTALL_K3S_COMMIT=6e768c301e77738ffd69934363aa0479c4a516d6 bash -

brandond on Feb 10, 2021