What Do I Do If a Public Image Can't Be Pulled During Workload Creation?

Symptom

The following information is displayed when a workload is created in a CCE Autopilot cluster:

Failed to pull image "100.125.**.**:32334/**/nginx:1.0": rpcerror: code =DeadlineExceeded desc = failed to pull and unpack image "100.125.**.**:32334/**/nginx:1.0": failed to resolve reference "100.125.**.**:32334/**/nginx:1.0": failed to do request Head: Head "https://100.125.**.**:32334/v2/**/nginx/manifests/1.0": dial tcp 100.125.**.**:32334: i/o timeout

Fault Location

When the CCE Autopilot cluster pulls the image from the public network, the NAT gateway cannot access the public network because there is no route destined for the NAT gateway in the route table of the subnet.

Solution

Add the route destined for 0.0.0.0/0 over the NAT gateway in the default route table or custom route table of the subnet.

1. Log in to the CCE console and click the cluster name to access the cluster console.
2. In the navigation pane, choose Overview. In the Networking Configuration area, view the cluster subnet.
3. Switch to the Network Console. In the navigation pane, choose Virtual Private Cloud > Subnets. Locate the subnet by name and click the route table name.

   

4. On the Summary tab, check whether a route to the NAT gateway exists.

   If no, manually add a route. Click Add Route.

   • Destination: Set this parameter to 0.0.0.0/0, which means any IP address.
   • Next Hop Type: Select NAT gateway.
   • Next Hop: Select the NAT gateway configured for the subnet.

   Click OK.