Out-of-memory if can’t read Kubernetes service account

[nmaguiar]

Describe the bug
If, for some reason, the Kubernetes service account that nAttrMon was using with nInput_Init to list the available pods can’t be read, it will start throwing the error but will eventually run into an out-of-memory error.

To Reproduce
Steps to reproduce the behavior:

1. Configure inputs to list pods using nInput_Init (with the cron setting) and gather metrics from pods
2. Set the configuration on a nAttrMon Kubernetes pod with an associated service account with the necessary permissions
3. Change or revoke permissions of the associated service account
4. See exceptions starting to occur
5. Eventually the pod will reach an out-of-memory situation

Expected behavior
There should be a log message indicating the listing of pods is no longer possible and it shouldn’t run out-of-memory.

Screenshots
n/a

Additional context
The error hasn’t been reproduce yet successfully so there might be missing steps.

cc: @abrizida