TCA partner system Harbor stuck on initiated
search cancel

TCA partner system Harbor stuck on initiated

book

Article ID: 377029

calendar_today

Updated On:

Products

VMware Telco Cloud Automation

Issue/Introduction

Within the TCA's Partner System, the Harbor instance remains stuck in the initiated state.

Environment

2.2

2.3

Resolution

1. Edit the TCA-CPs registration in Virtual Infrastructure, provide the credentials, and re-validate it.

2. Reboot the TCA Manager and TCA-CP; or restart web-engine, app-engine service

    systemctl stop app-engine
    systemctl stop web-engine
   
    systemctl start app-engine
    systemctl start web-service



3. After reboot, If more than one harbor instance is stuck in initiated, usually one harbor instance goes into a "running" state, while subsequent harbor instances take several hours.

4. If the harbor is still stuck in initiated 

Perform a sample Add/Remove cluster for the harbor partner system and see if an error is generated, example
"Login to Remote Cloud failed with errors..." 

This could be an expired cert or outdated kubeconfig

5. For each cluster selected in the harbor partner system may contain a outdated cert on the TCA Manager or TCACP

6. To isolate the cluster to the TCACP that has the issue, search the TCACP for error:

”Exception creating kubernetes session” 

ssh into TCACP

​​grep -irH "Exception creating kubernetes session" /common/logs/*

If the error is found, then one or multiple clusters on that TCACP have a certificate issued or a kubeconfig issue.

For each cluster suspected cluster

copy the kubeconfig from cluster control plane node at /etc/kubernetes/admin
update in the TCA Manager Clusters kubeconfig

 

 

Powered by Wolken Software