Hi all.
I would like to ask how are you measuring devops success in your applications which are on aws. i have read about metrics such as deployment frequency, mean time to recover etc exists but I am trying to figure out how to collect this kind of metrics automatically.
thanks!
Believe they have links to detailed guides on each of these here https://docs.aws.amazon.com/solutions/latest/devops-monitoring-dashboard-on-aws/devops-metrics-list.html
It’s mostly about configuring CloudWatch metrics correctly and tying those into dashboards.
Thanks I will give it a look! What about metrics that could theocratically be extracted with GitHub API. such as a build time for a workflow or the push frequency to master branch or the number of hotfixes that are created during a time period. Is there a tool that people here are using to gather those metrics?
The links you provided works great if you are on code dommit and codebuild
Good question. I’m not sure how you’d bring the GH data into AWS, but there probably is some way.
The mean time to recovery and similar metrics don’t have anything to do with GitHub because they measure what’s happening on AWS itself, so you’re good there.
For the others, maybe you could look into publishing custom CloudWatch metrics: https://docs.aws.amazon.com/AmazonCloudWatch/latest/monitoring/publishingMetrics.html
Maybe there is a better way and I’m not sure what the state of the art is. But I would think it shouldn’t be too difficult to trigger calls to AWS to publish a metric on pushes, merged, etc.
Are you also using GitHub Actions for building?
Yes I am using github actions for building and deploying my containers. they are deployed to ecs. i believe i can also gather some metrics from images published to ecr.
Nice. I’m using ECS as well…
ECS with EC2 or Fargate?
Btw i found this solution which combines scripting magic with lambda functions and an api gateway
