/healthcheck doesn't work in ECS

[louislatreille-cb]

Describe the bug
As per the documentation and the examples in this repository, we can set the ECS healthCheck to the "/healthcheck" command.

After deploying the ADOT collector with this, the container remains unhealthy in ECS

Steps to reproduce
Deploy the ADOT collector with a healthcheck command as described in the documentation

What did you expect to see?
The container is becomes healthy after a few seconds

What did you see instead?
The container stays unhealthy forever

Environment
I'm using AWS CDK. Here's my container definition:

const adotContainer = taskDefinition.addContainer(`${id}-adot-container`, {
    image: ContainerImage.fromRegistry("amazon/aws-otel-collector:v0.37.0"),
    command: ["--config=/etc/ecs/ecs-cloudwatch-xray.yaml"],
    logging: LogDriver.awsLogs({
      logGroup: new LogGroup(scope, `${id}-adot-logs`, {
        logGroupName: `${id}-adot-logs`,
      }),
      streamPrefix: `${id}-adot-container`,
    }),
    healthCheck: {
      command: ["/healthcheck"],
    },
  });

Additional context
Add any other context about the problem here.

[zehsor]

you have to set the healthcheck the following way @louislatreille-cb:

healthCheck: {
             command: ["CMD", "/healthcheck"],
             timeout: Duration.seconds(5),
             interval: Duration.seconds(10),
             retries: 3,
             startPeriod: Duration.seconds(5),
       },

If you only set it to be "/healthcheck" the underlying cloudformation script will set it to use CMD-Shell. The ADOT collector does not come with a default shell.

Hope this was useful!

[Louis Latreille]

Thanks for the hint @zehsor, it worked!

[github-actions]

This issue is stale because it has been open 60 days with no activity. Remove stale label or comment or this will be closed in 30 days.

[github-actions]

This issue was closed because it has been marked as stale for 30 days with no activity.