This message was deleted.
# rke2
a
This message was deleted.
g
Note that I'm currently getting 401 when the control plane nodes are trying to pull images:
failed with status code [manifests v0.1.8]: 401 Unauthorized
I have also updated the node's CA store with the self-signed cert & restarted the RKE2 service.
c
why are you trying to add a TLS secret when you have already configured authentication?
you only need a TLS secret if you want to use client certificates to auth to the registry, instead of basic auth
👍 1
g
OK, so mTLS? What about CA Cert Bundle?
OK, the nodes seem to be pulling images now... looks like I needed
rancher/system-agent-installer-rke2:v1.23.17-rke2r1
to successfully update the RKE servers and agents (which isn't in the RKE2 release bundle on github btw)
Hmm, this seems to occur with certain images. The calico node pod has managed to pull down calico-node from the registry, however now it's throwing a 401 trying to get the
rancher/pause:3.6
image (replaced my registry hostname with {registryaddress}):
Copy code
Failed to create pod sandbox: rpc error: code = Unknown desc = failed to get sandbox image "{registryaddress}/rancher/pause:3. │
│ 6": failed to pull image "{registryaddress}/rancher/pause:3.6": failed to pull and unpack image "{registryaddress}/rancher/pause:3.6": failed to resolve reference "{registryaddress}/rancher/pause: │
│ 3.6": pulling from host {registryaddress} failed with status code [manifests 3.6]: 401 Unauthorized
c
system-agent-installer-rke2 image is part of Rancher’s provisioning framwork, not RKE2
but yeah, you’ll need that
👍 1
if registry auth is working for some images it should work for all of them
g
but yeah, you’ll need that
Does it use that in a different way? It's in the registry but the worker nodes can't seem to fetch it
rancher-system-agent.service logs are suspect:
Copy code
Oct 05 15:18:06 rancher-system-agent[915]: time="2023-10-05T15:18:06+13:00" level=info msg="[Applyinator] Extracting image rancher/system-agent-installer-rke2:v1.23.17-rke2r1 to directory /var/lib/rancher/agent/work/20231005-151806/352756b4d64bd84d9b8064b13d710a637636d0fbd8094df75c28f56a5ddb838e_0"
Oct 05 15:18:06 rancher-system-agent[915]: time="2023-10-05T15:18:06+13:00" level=info msg="Using private registry config file at /etc/rancher/agent/registries.yaml"
Oct 05 15:18:06 rancher-system-agent[915]: time="2023-10-05T15:18:06+13:00" level=info msg="Pulling image index.docker.io/rancher/system-agent-installer-rke2:v1.23.17-rke2r1"
Oct 05 15:18:06 rancher-system-agent[915]: time="2023-10-05T15:18:06+13:00" level=error msg="error while staging: Get \"<https://index.docker.io/v2/>\": read tcp 172.20.101.204:57676->34.194.164.123:443: read: connection reset by peer: failed to get image index.docker.io/rancher/system-agent-installer-rke2:v1.23.17-rke2r1"
Oct 05 15:18:06 rancher-system-agent[915]: time="2023-10-05T15:18:06+13:00" level=error msg="error executing instruction 0: Get \"<https://index.docker.io/v2/>\": read tcp 172.20.101.204:57676->34.194.164.123:443: read: connection reset by peer: failed to get image index.docker.io/rancher/system-agent-installer-rke2:v1.23.17-rke2r1"
I added
<http://index.docker.io|index.docker.io>
to the list of mirrors in Rancher, but didn't help
Will I need to set the Rancher default registry to point to my private one? Or is it enough to set it at a cluster level? Would I need to rebuild my clusters if I change this value?