# Feature Request: Dev-Ops pipeline improvements

**URL:** <https://discourse.nodered.org/t/feature-request-dev-ops-pipeline-improvements/93146>\
**Category:** FlowFuse\
**Created:** [14 November 2024 14:48 UTC](https://discourse.nodered.org/t/feature-request-dev-ops-pipeline-improvements/93146 "2024-11-14T14:48:38Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![butteronadonut](https://sea2.discourse-cdn.com/flex026/user_avatar/discourse.nodered.org/butteronadonut/32/41374_2.png) [@butteronadonut](https://discourse.nodered.org/u/butteronadonut)\
**Post date:** [14 November 2024 14:48 UTC](https://discourse.nodered.org/t/feature-request-dev-ops-pipeline-improvements/93146/1 "2024-11-14T14:48:38Z")

</div>

My team often finds themselves in a scenario where we need to stop production on our manufacturing floor to deploy changes to a set of machines, each running device-agent.

It would be a big advantage of we could deploy a devops pipeline in a variety of ways, perhaps having the option to deploy in certain ways

0 - as-is

1 - scheduled for a certain time/date. If this existed, we'd like to be able to cancel a scheduled deployment. In this case, we could just schedule it for break time or end of day as to not disrupt production.

2- deploy on next restart. It just gives a little more control over a deployment, making sure a deployment doesn't interfere with a currently running flow. If restarts are scheduled, it is kinda like #1

3- deploy with end-user acknowledgment. Perhaps more tricky, but this is a strategy that we use with a different product by blue clover devices ([Operation Guide — PLT Documentation](https://docs.pltcloud.com/plt-300a/operation/)). This forces the device-user to acknowledge and apply an update. There are big-advantages to this when deploying off-site and to suppliers where they may not be so happy to have an update pushed to hardware without awareness.