Configure Probes for your Application

SAAP provides robust mechanisms for managing the health and availability of container applications. Liveness, readiness, and startup probes are essential features that help ensure application reliability and stability.

In this tutorial, we will explore how to leverage the capabilities of SAAP to define different probes using the values.yaml file.

Objective

• Configure your application's liveness, readiness, and startup probes with their parameters.

Key Results

• Configure probe settings to effectively monitor application health and readiness.

Tutorial

Probes Error

When Deploying Stakater Nordmart Review API application, you might see this error in your application pods.

The pod status shows CrashLoopBackOff. Let's see pod's events for more detailed logs:

Because of the liveness and readiness probes got failed, the pods are not in running state.

So, to deploy your application smoothly you need to configure probes.

1. To add a liveness probe in your application, you need to define it in your deploy/values.yaml. Liveness probes continuously monitor the health of containers and automatically restart any instances that fail the health check.

   # Liveness Probe Configuration
   livenessProbe:
       # Enable liveness probe for the container
     enabled: true
       # Number of consecutive failures to consider the container as unhealthy
     failureThreshold: 50
       # Interval at which the liveness probe will be performed
     periodSeconds: 50
       # Number of consecutive successes required to consider the container as healthy again
     successThreshold: 1
       # Maximum time the probe can take before it is considered failed
     timeoutSeconds: 50
       # Initial delay before the first liveness probe is performed
     initialDelaySeconds: 50
       # HTTP GET request used for the liveness probe
     httpGet:
       # Path for the health check endpoint
       path: /actuator/health
       # Port number where the health check endpoint is exposed
       port: 8080
   

   Note

   Indentation should be followed as: application.deployment.livenessProbe.

   It should look like this:

   

   Note

   You can configure the parameters of the probe per your application's requirement. To know more about liveness probes, visit here.

2. To add a readiness probe in your application, you need to define it in your deploy/values.yaml. Readiness probes determine if a container is ready to serve traffic and delay routing until the container is ready.

   # Readiness Probe Configuration
   readinessProbe:
       # Enable readiness probe for the container
     enabled: true
       # Number of consecutive failures to consider the container as not ready
     failureThreshold: 50
       # Interval at which the readiness probe will be performed
     periodSeconds: 50
       # Number of consecutive successes required to consider the container as ready
     successThreshold: 1
       # Maximum time the probe can take before it is considered failed
     timeoutSeconds: 50
       # Initial delay before the first readiness probe is performed
     initialDelaySeconds: 50
       # HTTP GET request used for the readiness probe
     httpGet:
       # Path for the health check endpoint
       path: /actuator/health
       # Port number where the health check endpoint is exposed
       port: 8080
   

   Note

   Indentation should be followed as: application.deployment.readinessProbe.

   It should look like this:

   

   Note

   You can configure the parameters of the probe as per your application's requirement. To know more about readiness probes, visit here.

3. To add startup probe in your application, you need to define it in your deploy/values.yaml. Startup probe is specifically designed to check if an application has completed its initialization and is fully ready to serve traffic.

   # Startup Probe Configuration
   startupProbe:
       # Enable the startup probe for the container
     enabled: true
       # Number of consecutive failures to consider the startup as unsuccessful
     failureThreshold: 5
       # Interval at which the startup probe will be performed
     periodSeconds: 5
       # Number of consecutive successes required to consider the startup as successful
     successThreshold: 1
       # Timeout for the startup probe
     timeoutSeconds: 5
       # Delay before the first startup probe is performed
     initialDelaySeconds: 5
       # HTTP GET request used for the startup probe
     httpGet:
       # Path for the startup check endpoint
       path: /actuator/health
       # Port number where the startup check endpoint is exposed
       port: 8080
   

   Note

   Indentation should be followed as: application.deployment.startupProbe.

   It should look like this:

   

   Note

   You can configure the parameters of the probe as per your application's requirement. To know more about startup probes, visit here.

4. Save the values.yaml and run tilt up at the root of your directory. Press the space key to view the progress in Tilt web UI. The application should be running in the namespace used in tilt_options.json file.

   

5. In your Tiltfile, there is a port forwarding command written on line #60, this command will forward the port of the application to your local machine. To see the output of the deployment:

   curl localhost:8080/api/review/329199
   

   

   Review the JSON output on browser:

   

6. let's make one change; we will update the first review text to "Tilt Demo"

   

   Switch back to the tilt browser, you will see it has started picking up changes

   

   Within a few seconds, the change will be deployed, and you can refresh the route to see the change

   

   Awesome! you made it

7. Run tilt down to delete the application and related configuration from the namespace.

Good Job! Let's move on to next tutorial, where we'll see how we can persist data of our application.