add better migration for deprecated metric apollo_router_http_request…

…_retry_total (#6384)

Signed-off-by: Benjamin <[email protected]>

apollo-router/src/plugins/telemetry/config_new/attributes.rs

@@ -264,6 +264,9 @@ impl DefaultForLevel for SubgraphAttributes {
                 if self.graphql_operation_type.is_none() {
                     self.graphql_operation_type = Some(StandardAttribute::Bool(true));
                 }
+                if self.http_request_resend_count.is_none() {
+                    self.http_request_resend_count = Some(StandardAttribute::Bool(true));
+                }
             }
             DefaultAttributeRequirementLevel::None => {}
         }

docs/source/reference/router/telemetry/instrumentation/standard-instruments.mdx

@@ -127,43 +127,4 @@ The following metrics have been deprecated and should not be used.
 - `apollo_authentication_success_count` - **Deprecated**: use the `apollo.router.operations.authentication.jwt` metric instead. If the `authentication.jwt.failed` attribute is *absent* or `false`, the authentication succeeded.
 - `apollo_require_authentication_failure_count` - **Deprecated**: use the `http.server.request.duration` metric's `http.response.status_code` attribute. Requests with authentication failures have HTTP status code 401.
 - `apollo_router_timeout` - **Deprecated**: this metric conflates timed-out requests from client to the router, and requests from the router to subgraphs. Timed-out requests have HTTP status code 504. Use the `http.response.status_code` attribute on the `http.server.request.duration` metric to identify timed-out router requests, and the same attribute on the `http.client.request.duration` metric to identify timed-out subgraph requests.
-- `apollo_router_http_request_retry_total` **Deprecated**: this can be achieved with custom telemetry instead. See [below for an example](#migrate-from-apollo_router_http_request_retry_total).
-
-#### Migrate from `apollo_router_http_request_retry_total`
-
-The below configuration filters the deprecated metric, and defines a custom histogram recording the number of retried HTTP requests to subgraphs.
-
-```yaml title="config.router.yaml"
-telemetry:
-  exporters:
-    metrics:
-      common:
-        views:
-        - name: apollo_router_http_request_retry_total # Drop the deprecated metric
-          aggregation: drop
-        - name: http.client.request.retry # Set the right buckets for the new metric
-          aggregation:
-            histogram:
-              buckets:
-                - 1
-                - 10
-                - 100
-                - 1000
-  instrumentation:
-    instruments:
-      subgraph:
-        http.client.request.retry: # Create custom histogram measuring the number of retries for http request to a subgraph
-          type: histogram
-          value:
-            subgraph_resend_count: true
-          unit: hit
-          description: histogram of subgraph request resend count
-          condition:
-            gt:
-            - subgraph_resend_count: true
-            - 0
-          attributes:
-            subgraph.name: true
-            supergraph.operation.name:
-              supergraph_operation_name: string
-```
+- `apollo_router_http_request_retry_total` **Deprecated**: use the `http.client.request.duration` metric's `http.request.resend_count` attribute. Requests with retries will contains `http.request.resend_count` set with the number of retries.