03/21/2023, 3:32 PM
I'm making good progress toward adopting Harvester. Now I'm assessing the basic capabilities by creating a single-node K3s cluster through Rancher Server UI. I am stuck trying to diagnose a pending state. Can you direct me to find the cause? Cluster status:
waiting for viable init node
Conditions log:
[Disconnected] Cluster agent is not connected
Here's how I arrived. 1. On node, boot and install ISO on metal 2. On PC, Install Rancher Desktop w/ K8s v1.24 3. On PC, Install Rancher with Helm and set
of reverse proxy reachable by both Rancher Desktop cluster and Harvester node 4. Use Rancher UI via reverse proxy from this point forward 5. In Rancher, verify Harvester node health and practice creating a simple VM from Ubuntu cloud minimal image. 6. In Rancher, create a cluster using same image. The switch did not work with a VLAN tag so I have created an untagged VM network with working DHCP.
The "Machine Pools" sub-tab in Cluster Management UI is continually attempting to create the pool.
• Harvester 1.1.1 • Rancher 2.7.1
I never found a way to proceed, so I rebuilt Rancher Server with a new hostname so I could register Harvester. This time, I bypassed the proxy. I didn't have the same issue, but it's not clear whether the changes I made to storage or the proxy was the cause of the issue.


03/27/2023, 11:16 PM
The Rancher cluster agent tries to actively open a connection to Rancher. A proxy or NAT in between prevents this most of the times...


03/27/2023, 11:19 PM
Though Rancher Manager UI was usable through the proxy, it seems the agent could not connect to the API. I will update this thread if I revert to the proxy, but for now I am using a direct IP connection.