For now we prefer to wait and see who want that and have concretes use cases.

I am interested in this feature. My use case is the following:
I have a backend whose health reporting is done through graphql and the response status is 200 even when he graphql response contains an unhealthy state.

If there could be a way to specify a certain parsing for health check I could reuse that graphql response.

An alternative would be to configure health check to go through a different backend that would perform the parsing like

healthcheck:
 url: http://healthchecker

I have a use case where my service only responds to POST resquest with a specific body. In order to check it healthiness I need to be able to send POST request to the endpoint. I'm able to do that using Nomad check but not with traefik.

Hello,

Thanks for suggesting this, we also think that the support for a recheck interval could be a great addition.

Unfortunately, this issue is not on our roadmap, but anyone willing to contribute is more than welcome. In case anyone is interested but unsure on how to proceed check out our contributing guide.

Hey,
I would be interested in working on this feature. I'll contribute a PR in the following days for it :)

Is this issue still valid ?

Hello,

@sswastik02 yes it still valid we are looking for some from the community on this one.
@Muscaw do you have some time to propose your implementation?

Hi @mmatur,
I would like to work on this issue. If not already implemented by @Muscaw.

Hi @mmatur,

If it's okay with @Muscaw, I would like to work on this issue.

Hi @mmatur @sswastik02 ,

I unfortunately did not have enough free time to complete my PR. @sswastik02 feel free to work on it 👍

Hi @mmatur and @Muscaw,

In an effort to better understand this issue, I delved into the documentation provided here. As someone relatively new to this, I find that many of the functionalities outlined in the documentation appear to be relevant to our current issue. However, I'm struggling to identify the specific area I should focus on or where to begin my work.

Could you kindly offer some guidance or provide a starting point for me to look into? Any advice would be greatly appreciated.

Thank you in advance.

Hello, @mmatur
This here is where I think the patch should be implemented. When calls to function executeHealthCheck fail, the plan is to incorporate (perhaps another) ticker . Can you please tell whether I'm heading in the correct direction?

Hi @mmatur ,
I would like to propose this plan for this issue, steps which include

1. Implement GetStatus Function: Introduce a GetStatus function similar to the one found in this example. This function will enable the retrieval of the status of child services.

2. Introduce recheck Field: Add a new field named recheck in ServerHealthChecker to store the duration of the recheck interval. This will allow for configuring how frequently the health checks should be performed when the service returns an unhealthy status.

3. Utilize GetStatus Function: Modify the logic in healthcheck.go to utilize the GetStatus function. This will enable the healthcheck function to check if the current status is new and reset the ticker accordingly.

If the proposed plan looks good, I will proceed with implementing these changes and submit a pull request accordingly.

Your feedback and suggestions on this proposal are highly appreciated.

Hello @sswastik02,

Thanks for sharing and explaining the approach, sure, please let's open the PR!

Hi @rtribotte,
Thank you very much for your validation of my approach. I will go ahead and create a pull request shortly.

BR