loading...
Cover image for The Developer's Deployment Checklist

The Developer's Deployment Checklist

flippedcoding profile image Milecia McG ・1 min read

I was struggling setting up a new deployment pipeline for this project I'm working on using Conducto, so I decided to make a checklist to help. Hopefully it'll help somebody else!

  1. Make a list of all of the things that should be included in this release
  2. Make sure that any open PRs have been adequately reviewed
  3. Make sure that everything is merged or rebased to the same branch
  4. Run all of your unit tests locally
  5. Do some manual user testing
  6. Build/update a pipeline that will handle your code and deploy to the staging environment
    • Set environment variables
    • Do any code formatting
    • Run any unit tests
    • Run any integration tests
    • Set the rules around deploying from new branches, tags, or repos
    • Set the path to the cloud service you use
  7. Let the team know that the new features are up on staging and get some testing in to catch any weird behavior
  8. Build/update a pipeline that will handle your code and deploy to production
    • Clear the server cache
    • Set environment variables
    • Do any code formatting
    • Run any unit tests
    • Run any integration tests
    • Set the rules around deploying from new branches, tags, or repos
    • Set the path to the cloud service you use
  9. Track server utilization
  10. Track exception rates
  11. Have a post deploy meeting to see if there's anything that could be done better next time

This is just a quick, simple deployment strategy. Let me know if there's something you would add or move around!

Posted on by:

flippedcoding profile

Milecia McG

@flippedcoding

Starting classes soon! | Software/Hardware Engineer | International tech speaker | Random inventor and slightly mad scientist with extra sauce

Discussion

markdown guide
 

I've yet to find the right balance between merging and rebasing to get work brought together.

It's easy as a single developer, but when there's a team working in similar areas of the system things become interesting.

I'm probably doing something wrong