CallHttpAsync HttpRetryOptions not retrying on .NET isolated worker

[m-spurgeon]

I'm using .NET isolated Azure Durable Functions and calling CallHttpAsync, with HttpRetryOptions configured, from my orchestrator. The HTTP call fails with a retryable status (e.g. 503), but the orchestration does not retry, it fails after a single attempt.

Example:

HttpRetryOptions retryPolicy = new HttpRetryOptions(new List<HttpStatusCode>()
{
    HttpStatusCode.RequestTimeout,
    HttpStatusCode.TooManyRequests,
    HttpStatusCode.InternalServerError,
    HttpStatusCode.BadGateway,
    HttpStatusCode.ServiceUnavailable,
    HttpStatusCode.GatewayTimeout
})
{
    FirstRetryInterval = TimeSpan.FromSeconds(20),
    MaxNumberOfAttempts = 10
};

DurableHttpRequest durableHttpRequest = new DurableHttpRequest(HttpMethod.Post, new Uri("https://httpbin.org/status/503"))
{
    Content = System.Text.Json.JsonSerializer.Serialize(new { Name = "Test" }),
    Timeout = TimeSpan.FromMinutes(15),
    HttpRetryOptions = retryPolicy
};

DurableHttpResponse response = await context.CallHttpAsync(durableHttpRequest);

The error we are getting is below, but a retry isn't triggered from this.

{
   "$type": "System.Net.Http.HttpRequestException, System.Net.Http",
   "HttpRequestError": 0,
   "StatusCode": null,
   "Message": "Status code ServiceUnavailable received, retry requested.",
   "Data": {
      "$type": "System.Collections.ListDictionaryInternal, System.Private.CoreLib"
   },
   "InnerException": null,
   "HelpLink": null,
   "Source": "Microsoft.Azure.WebJobs.Extensions.DurableTask",
   "HResult": -2146233088,
   "StackTrace": "   at Microsoft.Azure.WebJobs.Extensions.DurableTask.TaskHttpActivityShim.RunAsync(TaskContext context, String rawInput) in /_/src/WebJobs.Extensions.DurableTask/Listener/TaskHttpActivityShim.cs:line 83"
}

Is there any reason this wouldn't be working with Isolated? The equivalent code had been working as expected using in-process.