generating state of devOps dashboard in aws cloudwatch based on the metrics data generated by pipeline-dashboard and Ops-metrics serverless application repositories
Prerequisite: State of devOps solution consists of three different Serverless Application Repositories. In order to have this solution working, You are required to deploy below three serverless applications:
-
ApplicationId: arn:aws:serverlessrepo:us-east-1:611781478414:applications/mechanicalrock-ops-metrics
-
ApplicationId: arn:aws:serverlessrepo:us-east-1:611781478414:applications/mechanicalrock-pipeline-dashboard
-
ApplicationId: arn:aws:serverlessrepo:ap-southeast-2:611781478414:applications/mechanicalrock-state-of-devops-dashboard
Once the above applications are deployed in your account, follow below 3 steps to get a full advantage of your State Of DevOps metrics and monitoring
Step 1:
- Find /state-of-devops/app-names in ssm parametor store
- Replace the value with your apps/systems' names as a comma separated string eg "app1,app2"
Step 2:
- Make sure each code pipeline related to your app will have the app name included in the pipeline name. for instance imagine you have deployed an app called app1 into your aws account. You are also using two code piplines one for frontend and one for backend deployments of your app. Your pipline names should have app1 in their naming convention. e.g. app1-frontend-codepipeline and app1-backend-codepipeline or codepipeline-app1-frontend and codepipeline-app1-backend
Step 3:
- Create cloudwatch alarms for your app failures which their names are starting with your app name. For example app1-backend-lambda-failures or app1-cloudfront-400-errors or app1-downstream-system-failure Make sure all of your alarm names start with your app names