Establish a method to block requests when kafka consumption itself is broken

[harshad16]

Is your feature request related to a problem? Please describe.
We have come across a scenario in which Kafka consumers couldn't consume due to a change in offset value, which can happen if the data is lost out of Kafka, if such a scenario is encountered in future, maybe it would be a good idea to let users know the system is under maintenance, instead of losing request of users, while users are waiting for results.

[sesheta]

@harshad16: This issue is currently awaiting triage.
One of the @thoth-station/devsops will take care of the issue, and will accept the issue by applying the
triage/accepted label and provide further guidance.

The triage/accepted label can be added by org members by writing /triage accepted in a comment.

Details

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[harshad16]

/priority backlog

[sesheta]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

/lifecycle stale

[sesheta]

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

/lifecycle rotten