Run health checks on background services
In a CI pipeline, health checks confirm that services are running before the build runs other steps that need to interact with those services. This topic explains how to run a health check on services running in Background steps before running the rest of the steps in the stage. This example uses step groups to run health checks on multiple background services.
This example assumes you have a CI pipeline with a Build stage and at least one Background step.
Create step groups
Add one step group for each background service that you want to run a health check on. If you have multiple health check step groups, organize the step groups to run in parallel.
For example, if you want to run health checks on two services, create two step groups and run them in parallel, as shown in the following YAML example.
stages:
- stage:
identifier: build
type: CI
name: build
spec:
cloneCodebase: false
execution:
steps:
- parallel: ## Parallel flag.
- stepGroup: ## First step group.
name: group 1
identifier: sg1
steps:
...
- stepGroup: ## Second step group.
name: group 2
identifier: sg2
steps:
...
Add Background steps
Add a Background step to each step group. A Background step runs a service in the background. Required Background step settings depend on the service you're running and your build infrastructure. The following YAML examples use Background steps to run multiple PostgreSQL instances.
For a Background step to run a service, the build environment must have the necessary binaries. Depending on the stage's build infrastructure, Background steps can use binaries that exist in the build environment or pull an image, such as a public or private Docker image, that contains the required binaries. For more information about when and how to specify images, go to the Background step Container Registry and Image settings.
- Harness Cloud
- Self-managed Kubernetes cluster
- stage:
identifier: build
type: CI
name: build
spec:
cloneCodebase: false
execution:
steps:
- parallel:
- stepGroup:
name: group 1
identifier: sg1
steps:
- step: ## This step is in the first step group.
identifier: Background_1
type: Background
name: background service 1
spec: ## Configure specs for the service you want to run.
shell: Sh
envVariables:
POSTGRES_USER: postgres
POSTGRES_DB: test1
POSTGRES_PASSWORD: password
entrypoint:
- docker-entrypoint.sh
- "-p 5433"
- stepGroup:
name: group 2
identifier: sg2
steps:
- step: ## This step is in the second step group.
identifier: Background_2
type: Background
name: Background service 2
spec: ## Configure specs for the service you want to run.
shell: Sh
envVariables:
POSTGRES_USER: postgres
POSTGRES_DB: test2
POSTGRES_PASSWORD: password
entrypoint:
- docker-entrypoint.sh
- "-p 5434"
- stage:
identifier: build
type: CI
name: build
spec:
cloneCodebase: false
infrastructure:
...
execution:
steps:
- parallel:
- stepGroup:
name: group 1
identifier: sg1
steps:
- step: ## This step is in the first step group.
identifier: Background_1
type: Background
name: background service 1
spec: ## Configure specs for the service you want to run.
connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
image: postgres
shell: Sh
envVariables:
POSTGRES_USER: postgres
POSTGRES_DB: test1
POSTGRES_PASSWORD: password
PGDATA: /tmp/pgdata1
entrypoint:
- docker-entrypoint.sh
- "-p 5433"
- stepGroup:
name: group 2
identifier: sg2
steps:
- step: ## This step is in the second step group.
identifier: Background_2
type: Background
name: Background service 2
spec: ## Configure specs for the service you want to run.
connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
image: postgres
shell: Sh
envVariables:
POSTGRES_USER: postgres
POSTGRES_DB: test2
POSTGRES_PASSWORD: password
PGDATA: /tmp/pgdata2
entrypoint:
- docker-entrypoint.sh
- "-p 5434"
Add health checks
In each step group, after the Background step, add a Run step that runs a health check on that service. The commands necessary for a health check depend on the service you're running. The following YAML examples use psql commands to run health checks on PostgreSQL services. For more information about this use case, go to Run multiple PostgreSQL instances in Background steps.
For the Run step to run the health check commands, the build environment must have the necessary binaries. Depending on the stage's build infrastructure, Run steps can use binaries that exist in the build environment or pull an image, such as a public or private Docker image, that contains the required binaries. For more information about when and how to specify images, go to the Run step Container Registry and Image settings.
- Harness Cloud
- Self-managed Kubernetes cluster
- stage:
identifier: build
type: CI
name: build
spec:
cloneCodebase: false
execution:
steps:
- parallel:
- stepGroup: ## First step group.
name: group 1
identifier: sg1
steps:
- step: ## First background service.
identifier: Background_1
type: Background
name: background service 1
spec:
shell: Sh
envVariables:
POSTGRES_USER: postgres
POSTGRES_DB: test1
POSTGRES_PASSWORD: password
entrypoint:
- docker-entrypoint.sh
- "-p 5433"
- step: ## First health check.
identifier: Run_1
type: Run
name: health check 1
spec: ## Configure specs according to the service you're checking.
shell: Sh
command: |
sleep 15
psql -U postgres -d test1 -h Background_1 -p 5433
- stepGroup: ## Second step group.
name: group 2
identifier: sg2
steps:
- step: ## Second background service.
identifier: Background_2
type: Background
name: Background service 2
spec:
shell: Sh
envVariables:
POSTGRES_USER: postgres
POSTGRES_DB: test2
POSTGRES_PASSWORD: password
entrypoint:
- docker-entrypoint.sh
- "-p 5434"
- step: ## Second health check.
type: Run
name: health check 2
identifier: Run_2
spec: ## Configure specs according to the service you're checking.
shell: Sh
command: |-
sleep 15
psql -U postgres -d test2 -h Background_2 -p 5434
- stage:
identifier: build
type: CI
name: build
spec:
cloneCodebase: false
infrastructure:
...
execution:
steps:
- parallel:
- stepGroup: ## First step group.
name: group 1
identifier: sg1
steps:
- step: ## First background service.
identifier: Background_1
type: Background
name: background service 1
spec:
connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
image: postgres
shell: Sh
envVariables:
POSTGRES_USER: postgres
POSTGRES_DB: test1
POSTGRES_PASSWORD: password
PGDATA: /tmp/pgdata1
entrypoint:
- docker-entrypoint.sh
- "-p 5433"
- step: ## First health check.
identifier: Run_1
type: Run
name: health check 1
spec: ## Configure specs according to the service you're checking.
connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
image: postgres:9-alpine
shell: Sh
command: |
sleep 15
psql -U postgres -d test1 -h localhost -p 5433
- stepGroup: ## Second step group.
name: group 2
identifier: sg2
steps:
- step: ## Second background service.
identifier: Background_2
type: Background
name: Background service 2
spec:
connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
image: postgres
shell: Sh
envVariables:
POSTGRES_USER: postgres
POSTGRES_DB: test2
POSTGRES_PASSWORD: password
PGDATA: /tmp/pgdata2
entrypoint:
- docker-entrypoint.sh
- "-p 5434"
- step: ## Second health check.
type: Run
name: health check 2
identifier: Run_2
spec: ## Configure specs according to the service you're checking.
connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
image: postgres:9-alpine
shell: Sh
command: |-
sleep 15
psql -U postgres -d test2 -h localhost -p 5434
Test and finalize the pipeline
Run your pipeline to test your background services and health checks. You can monitor and review build logs on the Build details page.
Once you've confirmed that the background services and health checks are functioning as expected, configure any remaining steps in this stage, as necessary. Make sure these steps are not in your background service step groups or in parallel with the step groups.
Background steps don't persist across stages.
Pipeline YAML examples
These pipeline YAML examples use two step groups to run health checks on two PostgreSQL services.
- Harness Cloud
- Self-managed Kubernetes cluster
pipeline:
name: default
identifier: default
projectIdentifier: default
orgIdentifier: default
tags: {}
stages:
- stage:
identifier: build
type: CI
name: build
spec:
cloneCodebase: false
execution:
steps:
- parallel: ## This flag causes the two step groups to run in parallel. The steps within each group run sequentially, but the two groups start at the same time.
- stepGroup: ## First step group.
name: group 1
identifier: sg1
steps:
- step: ## This is the background service in the first group.
identifier: Background_1
type: Background
name: background service 1
spec: ## This example runs PostgreSQL. Configure your step's specs for the service you want to run.
shell: Sh
envVariables:
POSTGRES_USER: postgres
POSTGRES_DB: test1
POSTGRES_PASSWORD: password
entrypoint:
- docker-entrypoint.sh
- "-p 5433"
- step: ## This is the health check in the first group.
identifier: Run_1
type: Run
name: health check 1
spec: ## This example uses PostgreSQL. Configure your step's specs according to the service you're checking.
shell: Sh
command: |
sleep 15
psql -U postgres -d test1 -h Background_1 -p 5433
- stepGroup: ## Second step group.
name: group 2
identifier: sg2
steps:
- step: ## This is the background service in the second group.
identifier: Background_2
type: Background
name: Background service 2
spec: ## This example runs PostgreSQL. Configure your step's specs for the service you want to run.
shell: Sh
envVariables:
POSTGRES_USER: postgres
POSTGRES_DB: test2
POSTGRES_PASSWORD: password
entrypoint:
- docker-entrypoint.sh
- "-p 5434"
- step: ## This is the health check in the second group.
type: Run
name: health check 2
identifier: Run_2
spec: ## This example uses PostgreSQL. Configure your step's specs according to the service you're checking.
shell: Sh
command: |-
sleep 15
psql -U postgres -d test2 -h Background_2 -p 5434
- step: ## Add other steps to your stage as needed. Make sure they are not in the step groups or under the step groups' parallel flag.
identifier: Run_3
type: Run
name: Run_3
spec:
...
platform:
os: Linux
arch: Amd64
runtime:
type: Cloud
spec: {}
pipeline:
name: default
identifier: default
projectIdentifier: default
orgIdentifier: default
tags: {}
stages:
- stage:
identifier: build
type: CI
name: build
spec:
cloneCodebase: false
infrastructure:
type: KubernetesDirect
spec:
connectorRef: YOUR_KUBERNETES_CLUSTER_CONNECTOR_ID
namespace: YOUR_KUBERNETES_NAMESPACE
volumes: ## This example uses mounted volumes for PostgreSQL data.
- mountPath: /tmp/pgdata1
type: EmptyDir
spec:
medium: ""
- mountPath: /tmp/pgdata2
type: EmptyDir
spec:
medium: ""
automountServiceAccountToken: true
nodeSelector: {}
os: Linux
execution:
steps:
- parallel: ## This flag causes the two step groups to run in parallel. The steps within each group run sequentially, but the two groups start at the same time.
- stepGroup: ## First step group.
name: group 1
identifier: sg1
steps:
- step: ## This is the background service in the first group.
identifier: Background_1
type: Background
name: background service 1
spec: ## This example runs PostgreSQL. Configure your step's specs for the service you want to run.
connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
image: postgres
shell: Sh
envVariables:
POSTGRES_USER: postgres
POSTGRES_DB: test1
POSTGRES_PASSWORD: password
PGDATA: /tmp/pgdata1
entrypoint:
- docker-entrypoint.sh
- "-p 5433"
- step: ## This is the health check in the first group.
identifier: Run_1
type: Run
name: health check 1
spec: ## This example uses PostgreSQL. Configure your step's specs according to the service you're checking.
connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
image: postgres:9-alpine
shell: Sh
command: |
sleep 15
psql -U postgres -d test1 -h localhost -p 5433
- stepGroup: ## Second step group.
name: group 2
identifier: sg2
steps:
- step: ## This is the background service in the second group.
identifier: Background_2
type: Background
name: Background service 2
spec: ## This example runs PostgreSQL. Configure your step's specs for the service you want to run.
connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
image: postgres
shell: Sh
envVariables:
POSTGRES_USER: postgres
POSTGRES_DB: test2
POSTGRES_PASSWORD: password
PGDATA: /tmp/pgdata2
entrypoint:
- docker-entrypoint.sh
- "-p 5434"
- step: ## This is the health check in the second group.
type: Run
name: health check 2
identifier: Run_2
spec: ## This example uses PostgreSQL. Configure your step's specs according to the service you're checking.
connectorRef: YOUR_IMAGE_REGISTRY_CONNECTOR
image: postgres:9-alpine
shell: Sh
command: |-
sleep 15
psql -U postgres -d test2 -h localhost -p 5434
- step: ## Add other steps to your stage as needed. Make sure they are not in the step groups or under the step groups' parallel flag.
identifier: Run_3
type: Run
name: Run_3
spec:
...