Deploy Your First Stack
Table of contents
Table of contents
Summary
Generally, a happy environment would need to be set up before you can start deploying stacks–but that is boring. It is much more fun to to start using happy and deploy your first application! Luckily, we have a sandbox environment to do just that.
Ground Rules
Some ground rules:
- This is a very handy way to play with happy and experiment with new stack ideas–please don’t abuse it
- Stacks are cleaned up automatically at the end of the week
- This is not a production environment
- The sandbox is for CZI employees only
Deploy
Here’s how to get started:
- Install the latest version of happy CLI
- Clone happy
- Navigate to ./happy/tree/main/examples/typical_app (or any of the other example projects in /examples)
- Execute
happy create <stackname>where<stackname>is the name of the stack - You’re done!
This stack will deploy two services. To see the endpoints to access them, use happy list and it should look like the following:
$ happy list
[INFO]: Listing stacks from the happy api
│──────────│──────────│─────────│─────────────────────────────────────────│────────────────────────────│────────────────────────────────│─────────│───────────────────────────────────────────────────────────────│─────────────│
│ NAME │ OWNER │ APP │ REPO │ BRANCH │ HASH │ STATUS │ URLS │ LASTUPDATED │
│──────────│──────────│─────────│─────────────────────────────────────────│────────────────────────────│────────────────────────────────│─────────│───────────────────────────────────────────────────────────────│─────────────│
│ typical1 │ alokshin │ typical │ https://github.com/chanzuckerberg/happy │ alokshin/global-name-check │ dirty git tree (PLEASE COMMIT │ applied │ https://typical1.happy-playground-rdev.rdev.si.czi.technology │ 176h34m39s │
│ │ │ │ │ │ YOUR CHANGES) │ │ │ │
│──────────│──────────│─────────│─────────────────────────────────────────│────────────────────────────│────────────────────────────────│─────────│───────────────────────────────────────────────────────────────│─────────────│
Change the Dockerfile, application code, happy configuration however you like and redeploy using happy update <stackname>.
Inspect
Once your stack has successfully been deployed, you can obviously use your web browser to navigate to the application. But there are some other cool commands you can experiment with a running stack.
Logging & Monitoring
happy logs <stack_name> <service_name>- print the logs of the running application. Many of my stacks have way too many logs and I often only want to see the logs from a recent error. To do that, I often will dohappy logs <stack_name> <service_name> --since 10m --output /tmp/logs.json. Then I open the logfile in my editor for further inspection. This command does not stream logs, but happy does capture all logs in CloudWatch. At the end of the command it will ask if you’d like to view the full log stream. This will send you to the AWS console where you can see all the application logs.happy events <stack_name>- prints the latest events related to the stack’s deployment. This is handy for when a deployment doesn’t go so well. For instance, you might have specified the wrong port number, a health check is failing, or you accidently are deploying on the wrong architecture. All these types of events will either show up inhappy eventsorhappy logs.happy resources <stack_name>- prints the AWS resources created for your stack
$ happy resources typical1
[INFO]: Retrieving stack 'typical1' from environment 'rdev'
│─────────────────────────────────────────────────────────────────────────│────────────────────────────────────────│─────────────────────────────────────────│───────────│─────────────────────────────────────────────────────────────────────────────────────────│
│ MODULE (50) │ NAME │ TYPE │ MANAGEDBY │ INSTANCES │
│─────────────────────────────────────────────────────────────────────────│────────────────────────────────────────│─────────────────────────────────────────│───────────│─────────────────────────────────────────────────────────────────────────────────────────│
│ module.stack │ stack_dashboard │ datadog_dashboard_json │ terraform │ rde-hbv-vbw │
│ module.stack │ oidc_config │ kubernetes_secret │ terraform │ si-rdev-happy-eks-rdev-happy-env/typical1-oidc-config │
│ module.stack │ suffix │ random_pet │ terraform │ helping-macaque │
│ module.stack.module.services["frontend"] │ deployment │ kubernetes_deployment_v1 │ terraform │ si-rdev-happy-eks-rdev-happy-env/typical1-frontend │
│ module.stack.module.services["frontend"] │ hpa │ kubernetes_horizontal_pod_autoscaler_v1 │ terraform │ si-rdev-happy-eks-rdev-happy-env/typical1-frontend │
│ module.stack.module.services["frontend"] │ service │ kubernetes_service_v1 │ terraform │ si-rdev-happy-eks-rdev-happy-env/typical1-frontend │
│ module.stack.module.services["frontend"] │ this │ random_pet │ terraform │ clever-possum │
│ module.stack.module.services["frontend"].module.ecr │ lifecycle │ aws_ecr_lifecycle_policy │ terraform │ typical1/rdev/frontend │
│ module.stack.module.services["frontend"].module.ecr │ repo │ aws_ecr_repository │ terraform │ arn:aws:ecr:us-west-2:401986845158:repository/typical1/rdev/frontend │
│ module.stack.module.services["frontend"].module.iam_service_account │ role │ aws_iam_role │ terraform │ [] │
│ module.stack.module.services["frontend"].module.iam_service_account │ service_account │ kubernetes_service_account │ terraform │ si-rdev-happy-eks-rdev-happy-env/typical1-frontend-rdev-typical1 │
│ module.stack.module.services["frontend"].module.ingress[0] │ ingress │ kubernetes_ingress_v1 │ terraform │ si-rdev-happy-eks-rdev-happy-env/typical1-frontend │
│ module.stack.module.services["frontend"].module.ingress[0] │ ingress_bypasses │ kubernetes_ingress_v1 │ terraform │ si-rdev-happy-eks-rdev-happy-env/typical1-frontend-options-bypass │
│ module.stack.module.services["internal-api"] │ deployment │ kubernetes_deployment_v1 │ terraform │ si-rdev-happy-eks-rdev-happy-env/typical1-internal-api │
│ module.stack.module.services["internal-api"] │ hpa │ kubernetes_horizontal_pod_autoscaler_v1 │ terraform │ si-rdev-happy-eks-rdev-happy-env/typical1-internal-api │
│ module.stack.module.services["internal-api"] │ service │ kubernetes_service_v1 │ terraform │ si-rdev-happy-eks-rdev-happy-env/typical1-internal-api │
│ module.stack.module.services["internal-api"] │ this │ random_pet │ terraform │ smashing-catfish │
│ module.stack.module.services["internal-api"].module.ecr │ lifecycle │ aws_ecr_lifecycle_policy │ terraform │ typical1/rdev/internal-api │
│ module.stack.module.services["internal-api"].module.ecr │ repo │ aws_ecr_repository │ terraform │ arn:aws:ecr:us-west-2:401986845158:repository/typical1/rdev/internal-api │
│ module.stack.module.services["internal-api"].module.iam_service_account │ role │ aws_iam_role │ terraform │ arn:aws:iam::401986845158:role/si-playground-eks-v2/typical1-internal-api-rdev-typical1 │
│ module.stack.module.services["internal-api"].module.iam_service_account │ service_account │ kubernetes_service_account │ terraform │ si-rdev-happy-eks-rdev-happy-env/typical1-internal-api-rdev-typical1 │
│ module.stack.module.services["internal-api"].module.ingress[0] │ ingress │ kubernetes_ingress_v1 │ terraform │ si-rdev-happy-eks-rdev-happy-env/typical1-internal-api │
│ module.stack.module.services["internal-api"].module.ingress[0] │ ingress_bypasses │ kubernetes_ingress_v1 │ terraform │ si-rdev-happy-eks-rdev-happy-env/typical1-internal-api-options-bypass │
│ │ *typical1-frontend-rdev-typical1 │ ServiceAccount │ k8s │ │
│ │ *typical1-internal-api-rdev-typical1 │ ServiceAccount │ k8s │ │
│ │ typical1-frontend │ Service │ k8s │ │
│ │ typical1-internal-api │ Service │ k8s │ │
│ │ *typical1-oidc-config │ Secret │ k8s │ │
│ │ typical1-frontend │ Deployment │ k8s │ │
│ │ typical1-internal-api │ Deployment │ k8s │ │
│ │ typical1-frontend │ HorizontalPodAutoscaler │ k8s │ │
│ │ typical1-internal-api │ HorizontalPodAutoscaler │ k8s │ │
│ │ typical1-frontend │ Ingress │ k8s │ │
│ k8s:Ingress/typical1-frontend │ │ Application Load Balancer │ k8s │ k8s-stacktypical1help-2fe1caf63c-78548467.us-west-2.elb.amazonaws.com │
│ k8s:Ingress/typical1-frontend │ │ Route 53 Entry │ k8s │ typical1.happy-playground-rdev.rdev.si.czi.technology │
│ │ typical1-frontend-options-bypass │ Ingress │ k8s │ │
│ k8s:Ingress/typical1-frontend-options-bypass │ │ Application Load Balancer │ k8s │ k8s-stacktypical1help-2fe1caf63c-78548467.us-west-2.elb.amazonaws.com │
│ k8s:Ingress/typical1-frontend-options-bypass │ │ Route 53 Entry │ k8s │ typical1.happy-playground-rdev.rdev.si.czi.technology │
│ │ typical1-internal-api │ Ingress │ k8s │ │
│ k8s:Ingress/typical1-internal-api │ │ Application Load Balancer │ k8s │ k8s-stacktypical1help-2fe1caf63c-78548467.us-west-2.elb.amazonaws.com │
│ k8s:Ingress/typical1-internal-api │ │ Route 53 Entry │ k8s │ typical1.happy-playground-rdev.rdev.si.czi.technology │
│ │ typical1-internal-api-mybypass-bypass │ Ingress │ k8s │ │
│ k8s:Ingress/typical1-internal-api-mybypass-bypass │ │ Application Load Balancer │ k8s │ k8s-stacktypical1help-2fe1caf63c-78548467.us-west-2.elb.amazonaws.com │
│ k8s:Ingress/typical1-internal-api-mybypass-bypass │ │ Route 53 Entry │ k8s │ typical1.happy-playground-rdev.rdev.si.czi.technology │
│ │ typical1-internal-api-mybypass2-bypass │ Ingress │ k8s │ │
│ k8s:Ingress/typical1-internal-api-mybypass2-bypass │ │ Application Load Balancer │ k8s │ k8s-stacktypical1help-2fe1caf63c-78548467.us-west-2.elb.amazonaws.com │
│ k8s:Ingress/typical1-internal-api-mybypass2-bypass │ │ Route 53 Entry │ k8s │ typical1.happy-playground-rdev.rdev.si.czi.technology │
│ │ typical1-internal-api-options-bypass │ Ingress │ k8s │ │
│ k8s:Ingress/typical1-internal-api-options-bypass │ │ Application Load Balancer │ k8s │ k8s-stacktypical1help-2fe1caf63c-78548467.us-west-2.elb.amazonaws.com │
│ k8s:Ingress/typical1-internal-api-options-bypass │ │ Route 53 Entry │ k8s │ typical1.happy-playground-rdev.rdev.si.czi.technology │
│─────────────────────────────────────────────────────────────────────────│────────────────────────────────────────│─────────────────────────────────────────│───────────│─────────────────────────────────────────────────────────────────────────────────────────│
happy get <stack_name>- print metadata information about the deployed stack
$ happy get typical1
[INFO]: Retrieving stack 'typical1' from environment 'rdev'
│──────────│──────────│─────────│─────────────────────────────────────────│────────────────────────────│────────────────────────────────│─────────│───────────────────────────────────────────────────────────────│─────────────│
│ NAME │ OWNER │ APP │ REPO │ BRANCH │ HASH │ STATUS │ URLS │ LASTUPDATED │
│──────────│──────────│─────────│─────────────────────────────────────────│────────────────────────────│────────────────────────────────│─────────│───────────────────────────────────────────────────────────────│─────────────│
│ typical1 │ alokshin │ typical │ https://github.com/chanzuckerberg/happy │ alokshin/global-name-check │ dirty git tree (PLEASE COMMIT │ applied │ https://typical1.happy-playground-rdev.rdev.si.czi.technology │ 177h2m8s │
│ │ │ │ │ │ YOUR CHANGES) │ │ │ │
│──────────│──────────│─────────│─────────────────────────────────────────│────────────────────────────│────────────────────────────────│─────────│───────────────────────────────────────────────────────────────│─────────────│
│─────────────────────────│──────────────────────────────────────────────────────────────────────────────│
│ RESOURCE (23) │ VALUE │
│─────────────────────────│──────────────────────────────────────────────────────────────────────────────│
│ Environment │ │
│ TFE │ │
│ Environment Workspace │ https://si.prod.tfe.czi.technology/app/happy-playground/workspaces/env- │
│ Stack Workspace │ https://si.prod.tfe.czi.technology/app/happy-playground/workspaces/-typical1 │
│ Backlog size │ 2 outstanding runs │
│ │ edu-platform-infra │
│ │ (planning)->1 │
│ │ lp-infra (applying)->1 │
│ AWS │ │
│ Account ID │ [401986845158] │
│ Region │ us-west-2 │
│ Profile │ czi-playground │
│ Service │ frontend │
│ Compute │ K8S │
│ namespace │ si-rdev-happy-eks-rdev-happy-env │
│ deployment_name │ typical1-frontend │
│ auth_method │ eks │
│ kube_api │ https://FF38057B0DB0747C32795D4B8265796A.gr7.us-west-2.eks.amazonaws.com │
│ Service │ internal-api │
│ Compute │ K8S │
│ namespace │ si-rdev-happy-eks-rdev-happy-env │
│ deployment_name │ typical1-internal-api │
│ auth_method │ eks │
│ kube_api │ https://FF38057B0DB0747C32795D4B8265796A.gr7.us-west-2.eks.amazonaws.com │
│─────────────────────────│──────────────────────────────────────────────────────────────────────────────│
Iterate and Develop
happy shell <stack_name> <service_name>- shell into your application. This is very handy for debugging running applications. The shell interactions are authenticated using IAM permissions.happy config <cp,delete,diff,exec,get,list,set>- add a key, value pair to your deployment. A key, value pair will be stored and added to your running container as an environment variable. Key, value pairs can be set on the stack-level or on the environment-level (all stacks in the environment get the same key value pairs).happy restart <stack_name>- performs a quick redeployment of the running containers. This is handy if you want to update yourhappy configof your application and have your application pick up the changes. The docker images will be the same and no new code will be uploaded. Everything will be the same except your containerized applications will be redeployed with updated configuration values.happy delete <stack_name>- deletes the stack