Hey everyone! I'm curious if anyone else is struggling with managing a bunch of different versions across various deployments to multiple customers. My company has over five helm charts, each with its own separate versioning and release process, plus we have distinct app versions that are also versioned individually. To make things even trickier, our customers might be using different combinations of these versions at any given time.
I'm trying to figure out how to keep tabs on all of this chaos. Full disclosure: I'm thinking about creating a web app to visualize and manage these release schedules, as the current project management tools aren't cutting it for me. Ideally, I'd like a timeline view that shows what version each component is on and which specific version a customer is using. Do you know of any tools that could help with this sort of tracking and display?
3 Answers
I had my own struggles with the devs just tagging everything as 'new.' It caused so many issues that we ended up enforcing a system where we tag by time and Gitlab pipeline number instead. With around 100 apps, it’d be impossible to diagnose anything if we didn’t track them properly per app, time, and pipeline!
I totally get where you're coming from! We have a CICD setup with a script that sends data to an API. It stores stuff like build times and release info in a database, which helps me keep track of all service versions across our customers. We've got a custom app where devs can choose the specific versions or commits they want for each release plan. It really makes things less chaotic, and we can filter everything to find what we need easily!
Thanks! When you mention devs picking versions, that means you know which version goes to each customer, right? Are your release plans specific to customers? Also, do you integrate this setup with Azure DevOps or GitHub Actions for automated testing and release?
Have you considered using existing tools like ArgoCD? It could save you from building a web app from scratch. Also, Helm has built-in release tracking that includes the version of your charts and the parameters used, which can reference your Docker image versions too. Might be worth looking into!
Must be nice!