Enhancement Automatic Roll of Deployments

[rifen]

As a Nautobot Helm Deployment maintainer I am having inconsistent deployments depending on what values I change using helm upgrade. It isn't breaking but it can be annoying.

Found a recommendation from the Helm Docs:
https://helm.sh/docs/howto/charts_tips_and_tricks/#automatically-roll-deployments

Seems like it would be straight forward to implement in this chart.

I couldn't find anywhere where it was utilizing this in any of the deployment charts eg.

helm-charts/charts/nautobot/templates/celery-deployment.yaml

Line 8 in 4241333

name: {{ include "common.names.fullname" $ }}-celery-{{ $celeryName }}

But maybe it isn't needed 🤷🏻.

Submitting this for some extra 👀

[ubajze]

Thank you for opening an issue.

I want to make sure I understand the problem correctly. If you change something that only updates a ConfigMap, your Deployments are not restarted, and a new environmental variable is not injected into pods. Is that correct?

If this is the case, it is a reasonable ask.

In the meantime, you can probably deploy a tool like Reloader for example: https://github.com/stakater/Reloader

[rifen]

Thank you for opening an issue.

I want to make sure I understand the problem correctly. If you change something that only updates a ConfigMap, your Deployments are not restarted, and a new environmental variable is not injected into pods. Is that correct?

If this is the case, it is a reasonable ask.

In the meantime, you can probably deploy a tool like Reloader for example: https://github.com/stakater/Reloader

Correct. My deployments never get the env var because my deployments don't get triggered.

I will take a look at that as well!