Hi, I have a question regarding Rancher's behavior...
# general
a
Hi, I have a question regarding Rancher's behavior when switching the TLS certificate used in an existing setup. I'm using Rancher versions 2.9.3 and 2.10.0, and initially I had a certificate generated by cert-manager, set using
tls.secret: rancher
in the Helm values. Later on, I switched to a self-signed certificate (or a certificate signed by a private CA). At that point, there was already an existing downstream cluster registered and functioning. After changing the certificate, the
rancher-system-agent
on the cluster nodes started throwing the following error:
Copy code
rancher-system-agent[3597227]: time="2025-02-20T16:20:12Z" level=info msg="Initial connection to Kubernetes cluster failed with error Get \"
<https://mi.dominio.com/version>\": tls: failed to verify certificate: x509: certificate signed by unknown authority, removing CA data and trying again"
rancher-system-agent[3597227]: time="2025-02-20T16:20:12Z" level=fatal msg="error while connecting to Kubernetes cluster with nullified CA data: Get \"<https://mi.dominio.com/version>\": tls: failed to verify certificate: x509: certificate signed by unknown authority"
As a workaround, I manually installed the private CA certificate on each host node, and after that, the error went away. My question is: Is this the expected behavior when switching to a self-signed or private CA certificate with an existing downstream cluster, or could this be considered a bug? Note: From the agent pods, I can reach the Rancher URLs without needing to specify the CA certificate explicitly. Thanks in advance for your help.
c
a
thanks