Description of problem:

2 scenarios:

A) When KFNBC is opened for the very first time, dashboard logs report the following error:

{"level":50,"time":1665141865242,"pid":68,"hostname":"rhods-dashboard-566db7747f-rdm9l","msg":"rolebinding
 rhods-notebooks-image-pullers could not be read, 
rolebindings.rbac.authorization.k8s.io \"rhods-notebooks-image-pullers\"
 not found"}

The rolebinding seems to be automatically created after the first access to KFNBC. 

Similar issue to https://issues.redhat.com/browse/RHODS-5079, If it is the expected behaviour to have the "rhods-notebooks-image-pullers" RB created after the first launch of KFNBC, then the dashboard should not report an error, since it is not.

B) on some RHODS installation I find more errors about rolebinding in the dashboard pods:

{"level":50,"time":1665089350751,"pid":108,"hostname":"rhods-dashboard-5657f845dc-4p9vl","msg":"Failed
 to list rolebindings for user, Failed to list groups filtered by 
username."}{"level":50,"time":1665089350757,"pid":108,"hostname":"rhods-dashboard-5657f845dc-4p9vl","msg":"Failed
 to list rolebindings for user, Failed to list groups filtered by 
username."}

It's not clear when this scenario is triggered.

And it's not clear if the 2 scenarios have the same root-cause

Prerequisites (if any, like setup, operators/versions):

RHODS Installed

Steps to Reproduce

1. Install RHODS
2. Access KFNBC spawner page
3. check the dashboard logs (you need to open all the dashboard pods, since the error is reported only by one of the 5 replicas)

Actual results:

error in fetching rolebinding is reported in the dashboard logs

Expected results:

no error about fetching rolebinding its reported in the dashboard logs 

Reproducibility (Always/Intermittent/Only Once):

scenario A  on a cluster with RHODS v1.18.0-3

scenario B on a cluster with RHODS v1.17.0-9

Build Details:

scenario A  on a cluster with RHODS v1.18.0-3 

scenario B on a cluster with RHODS v1.17.0-9

Workaround:

N/A

Additional info: