35 lines
1.3 KiB
Markdown
35 lines
1.3 KiB
Markdown
|
---
|
||
|
title: Roll back or fix forward?
|
||
|
date: 2025-02-19
|
||
|
permalink: daily/2025/02/19/back-or-forward
|
||
|
tags:
|
||
|
- software-development
|
||
|
cta: ~
|
||
|
snippet: |
|
||
|
How do you resolve issues after a deployment? Do you roll back to the previous version or fix the issue and deploy it and move forward?
|
||
|
---
|
||
|
|
||
|
You deploy a code change but it creates an issue.
|
||
|
|
||
|
Maybe it breaks a feature, adds a bug or takes down an environment completely.
|
||
|
|
||
|
How do you resolve it?
|
||
|
|
||
|
Larger releases will sometimes have a roll back plan that details how to revert to the previous release.
|
||
|
|
||
|
This can be difficult, especially for large releases and ones that change the database schema or values.
|
||
|
|
||
|
Because of the amount of change, diagnosing an issue in a large release can take time.
|
||
|
|
||
|
If it's been a while since the prior release, it can be some time since the code that introduced the issue was worked on - making it harder to fix.
|
||
|
|
||
|
I prefer to do small releases and do them often.
|
||
|
|
||
|
Some releases contain a single commit which was made only minutes since the previous release.
|
||
|
|
||
|
This makes it easier to identify the issue, fix it and deploy a new version.
|
||
|
|
||
|
No rolling back database changes or reverting to previous releases.
|
||
|
|
||
|
Small, iterative deployments are less risky than large infrequent ones, easier to fix and make changes available to end users sooner.
|