/infra This comes from the tenant-stack pipeline repository
/infra/silo There is a helm chart deployed by terraform with values populated by terraform execution.
The one you see in gitlab pipeline.
Variables are passed by terraform.
/infra/silo/service-definitions.yaml This file contains the configuration used by terraform to create a base infrastrucutre
for each microservices.
In this case this is also passed as values to helm chart in infra/silo see the simulate.sh and the result in /out/tenant-infra-app.yaml
/argo This comes from the argocd gitops repository
/argo/stacks/silo/bootstrap Contains helm chart referenced by the application deployed by terraform .
It contains the base namespace definition and another application set that point to apps directory
/argo/stacks/silo/apps The list of apps deployed for the specific stack
/argo/envs Contains the values files for a specific environments for each microservice
/tenants This file is optionals. Ideally a pipeline create this dir only the first time and can be used to overwrite some values for a tenant. This file is included (without fail if does not exists) in all chart deployed with the tenant. The values into this file overwrite everything else
- A service definition file is passed (filtered by stack or other criterias) as helm values.
- Services values is used in a list generator to deploy only the services needs for that stack
- To onboard a new microservices (and deploy the infrastructure) the steps are:
- Create a gitops folder in gitops repo (that is not deployed becouse git generator does not exists anymore)
- Update the service-definition-file
sh simulate.sh
Will generate files in out directory
/out
tenant-infra-app #the application generated by the gitlab pipeline from terraform
helm template -f infra/silo/values.yaml -f infra/silo/service-definitions.yaml --set common.tenant=tenant1 --set common.stack=silo infra/silo | k apply -f -
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent