How to recover Kerberos authentication after a VM restart

Restore Kerberos SQL authentication after cluster VM shutdown in Automation Suite.

The following procedure applies to Automation Suite deployments configured with Kerberos authentication to SQL Server.

In scenarios where all cluster VMs were shut down for more than 6 hours, Kerberos tickets may expire and services may fail to connect to SQL Server after restart, resulting in errors such as Ticket expired or Cannot generate SSPI context.

To recover Kerberos authentication, take the following steps:

1. Verify that all nodes are in the Ready state:

   kubectl get nodes
   

2. Restart the kerberos-tgt-update CronJob to renew the Kerberos tickets:

   kubectl create job --from=cronjob/kerberos-tgt-update kerberos-tgt-manual-refresh -n uipath
   

3. Wait for the job to complete:

   kubectl wait --for=condition=complete job/kerberos-tgt-manual-refresh -n uipath --timeout=300s
   

4. Restart all deployments to pick up the renewed tickets:

   kubectl rollout restart deployment -n uipath
   

5. Wait for all services to sync and stabilize in ArgoCD. If any applications remain in a degraded state, manually sync them.