Why is the transport endpoint not connected?

Virtual Private Cloud Classic infrastructure

When you create a PVC, you see an error message similar to the following:

Transport endpoint is not connected.

To find the cause of the issue, gather system logs by deploying an inspectnode pod.

Follow the steps to gather logging information.

1. Create a debug-pvc.yaml file and specify the ibm.io/debug-level: "info" and ibm.io/curl-debug: "true" annotations.

   apiVersion: v1
   kind: PersistentVolumeClaim
   metadata:
      annotations:
        ibm.io/auto-create-bucket: "false"
        ibm.io/auto-delete-bucket: "false"
        ibm.io/bucket: <bucket-name> # Enter the name of your bucket.
        ibm.io/secret-name: <cos-secret-name> # Enter the name of your Kubernetes secret that contains your COS credentails
        ibm.io/debug-level: "info"
        ibm.io/curl-debug: "true"
      name: debug-pvc
      namespace: default
   spec:
     storageClassName: ibmc-s3fs-standard-perf-regional
     accessModes:
     - ReadWriteOnce
     resources:
       requests:
         storage: 10Gi
   

2. Create the debug-pvc PVC in your cluster.

   oc create -f debug-pvc.yaml>
   

3. Redeploy your app and reference the debug-pvc. Alternatively, you can use the following sample pod.

   apiVersion: v1
   kind: Pod
   metadata:
     name: debug-pod
   spec:
     volumes:
     - name: nginx
       persistentVolumeClaim:
           claimName: cos-debug
     containers:
     - name: nginx
       image: nginx
       volumeMounts:
       - mountPath: /debug
         name: nginx
   

4. Save the following DaemonSet configuration file as inspectnode-ds.yaml. After you deploy an app that references the PVC that you created earlier, deploy the DaemonSet when the Transport endpoint is not connected. error occurs to gather the logs.

   apiVersion: apps/v1
   kind: DaemonSet
   metadata:
       namespace: default
       name: ibm-inspectnode
       labels:
         app: ibm-inspectnode
   spec:
       selector:
         matchLabels:
             app: ibm-inspectnode
       updateStrategy:
         type: RollingUpdate
       template:
         metadata:
             labels:
               app: ibm-inspectnode
         spec:
             tolerations:
             - operator: "Exists"
             hostNetwork: true
             containers:
               - name: inspectnode
               image: alpine:latest
               command: ["/bin/sh"]
               args: ["-c", "while true; do msg=$(date); echo $msg; sleep 30; done"]
               securityContext:
                   runAsUser: 0
               volumeMounts:
                 - mountPath: /host/var/log
                   name: host-log
             volumes:
               - name: host-log
                 hostPath:
                   path: /var/log
   

5. Create the DaemonSet in your cluster when the Transport endpoint is not connected. error occurs.

   oc create -f ./inspectnode-ds.yaml
   

6. Get the name of the inspectnode pod.

   oc get pods -n default -l app=ibm-inspectnode -o wide
   

7. Copy the logs from inspectnode pod to your local machine. In the copy command, give your directory a name so that you can identify which node the logs are from. For example, node-one.log. You might also use the node IP in the directory name, for example: 10.XXX.XX.XXX.log.

   oc cp <inspectnode_pod_name>:/host/var/log  ./<node_name>.log
   

8. Review the syslog and the s3fslog for information about the Transport endpoint error. [Open a support ticket](/docs/openshift?topic=openshift-get-help and share the log files that you gathered.

9. Delete the ibm-inspectnode DaemonSet that you deployed.

   oc delete daemonset ibm-inspectnode