Hello, I am trying to deploy a "bare metal" cluste...
# generalb
better-telephone-21557
05/01/2025, 8:37 PMHello, I am trying to deploy a "bare metal" cluster from Rancher in VMware as a POC on Oracle Linux 8 and have gotten to where the
cattle-cluster-agent Copy code
│ INFO: Environment: CATTLE_ADDRESS=192.168.44.65 CATTLE_CA_CHECKSUM=f184280d010533f239f1f720f1badbcadc22d223a4a1db9cc80b8149d4adb259 CATTLE_CLUSTER=true CATTLE_CLUSTER_AGENT_PORT=<tcp://192.168.54.162:80> CATTLE_CLUSTER_AGENT_PORT_443_TCP=<tcp://192.168.54.162:443> CATTLE_ │
│ INFO: Using resolv.conf: search cattle-system.svc.cluster.local svc.cluster.local cluster.local nameserver 192.168.54.10 options ndots:5 │
│ ERROR: <https://rancherhost.test.io/ping> is not accessible (Could not resolve host: <http://rancherhost.test.io|rancherhost.test.io>)better-telephone-21557
05/01/2025, 8:40 PMI can login to the host and it will resolve the hostname of our rancher host.
better-telephone-21557
05/01/2025, 9:08 PMI also see that the
rke2-coredns-rke2-corednsc
creamy-pencil-82913
05/01/2025, 10:21 PMRKE2 will fall back to 8.8.8.8 if the dns server in the node’s resolv.conf points does not point at a valid upstream. For example, if you’re using systemd-resolved and resolv.conf points at 127.0.0.53 or something.
creamy-pencil-82913
05/01/2025, 10:22 PMYou should ensure that resolv.conf on your node points at a valid upstream, and then restart rke2.
creamy-pencil-82913
05/01/2025, 10:23 PMIf you search the RKE2 logs, there should be a warning about the fact that it is falling back to 8.8.8.8
5 Views