This article helps to fix issue with Telemetrymanager pod stuck in init state.

If telemetrymanager pod is stuck in init state -> it will appear like below-

kubectl get pods -n fault | grep telemetry

fault telemetrymanager-5d4c64db69-6tdth 0/1 Init:0/1 0 10m

fault telemetrymanager-7585b95578-6szt7 0/1 Init:0/1 0 9m39s

fault telemetrymanager-sync-devices-0.0.2-g10e37e0897-7fsxl 0/1 Init:0/1 0 9m3s

[root@####### ~]# kubectl logs -n fault telemetrymanager-5d4c64db69-6tdth

Error from server (BadRequest): container "telemetrymanager" in pod "telemetrymanager-5d4c64db69-6tdth" is waiting to start: PodInitializing

To fix the issue, 1st try with below workaround-

• Delete the pod and check if it restarts properly.
• Try with rollout restart of the deployment and check if it helps
• Try with setting replica=0 and then replica=1 *(Scale down and scale up of replica) -> this might help to fix the issue.

If the above steps doesn't work then follow below steps-

- Uninstalled TM using ansible:  

./run -c /root/config-dir destroy -t telemetrymanager

- Destroyed any remaining TM jobs, e.g.:  

kubectl -n fault get jobs  kubectl -n fault delete jobs telemetrymanager-resync-devices-cron-job-28474025 telemetrymanager-resync-faults-cron-job-28474080

- After that re-installed TM:  

./run -c /root/config-dir deploy -t telemetrymanager

Please Note:

1. We need access to the Ansible node which was used for initial deployment and configuration during paragon installation.
2. The location of the /root/config-dir might be different in every customer setup
   1. Need to ensure same config directory is used which was used during installation.

For any issues, please contact JTAC support.