Open hansmbakker opened 3 years ago
Do you see this being used directly against the worker? Currently, the only communication the worker has is with the host and the host already has the monitoring features, it would make sense to expose that there and perhaps have the host also report on worker. provided metrics.
Thoughts?
For my usecase it would be sufficient to have the host report the health checks. However, I think the host currently does not provide all the information that the ASP.Net Core ones can do (database connectivity, connectivity to other systems, etc). Can you give more insight into that?
Also, people running Azure Functions theirselves in a container, do not use the Azure Functions host I believe so they also do not get the health info from the host. See https://github.com/Azure/azure-functions-host/issues/5259 from @jeffhollan. There I suppose it would make sense to have the worker report the health status?
Triaging stale issues.
Keeping this open, this this will be addressed more broadly when support for customization of the ASP.NET features/pipeline is exposed, as part of the ASP.NET Core Integration effort (https://github.com/Azure/azure-functions-dotnet-worker/issues/1387)
ASP.net core has health checks apis.
Does the out of process dotnet functions worker have support for this or an alternative?
Also see https://github.com/Azure/azure-functions-host/issues/5259