prep release: v1.57.0

[abernix]

Note

When approved, this PR will merge into the 1.57.0 branch which will — upon being approved itself — merge into main.

Things to review in this PR:

• Changelog correctness (There is a preview below, but it is not necessarily the most up to date. See the Files Changed for the true reality.)
• Version bumps
• That it targets the right release branch (1.57.0 in this case!).

---

Important

If you have enabled Distributed query plan caching, updates to the query planner in this release will result in query plan caches being re-generated rather than re-used. On account of this, you should anticipate additional cache regeneration cost when updating between these versions while the new query plans come into service.

🚀 Features

Remove legacy schema introspection (PR #6139)

Schema introspection in the router now runs natively without JavaScript. We have high confidence that the new native implementation returns responses that match the previous Javascript implementation, based on differential testing: fuzzing arbitrary queries against a large schema, and testing a corpus of customer schemas against a comprehensive query.

Changes to the router's YAML configuration:

• The experimental_introspection_mode key has been removed, with the new mode as the only behavior in this release.
• The supergraph.query_planning.legacy_introspection_caching key is removed, with the behavior in this release now similar to what was false: introspection responses are not part of the query plan cache but instead in a separate, small in-memory—only cache.

When using the above deprecated configuration options, the router's automatic configuration migration will ensure that existing configuration continue to work until the next major version of the router. To simplify major upgrades, we recommend reviewing incremental updates to your YAML configuration by comparing the output of ./router config upgrade --config path/to/config.yaml with your existing configuration.

By @SimonSapin in #6139

Support new request_context selector for telemetry (PR #6160)

The router supports a new request_context selector for telemetry that enables access to the supergraph schema ID.

You can configure the context to access the supergraph schema ID at the router service level:

telemetry:
  instrumentation:
    events:
      router:
        my.request_event:
          message: "my request event message"
          level: info
          on: request
          attributes:
            schema.id:
              request_context: "apollo::supergraph_schema_id" # The key containing the supergraph schema id

You can use the selector in any service at any stage. While this example applies to events attributes, the selector can also be used on spans and instruments.

By @bnjjj in #6160

Support reading and setting port on request URIs using Rhai (Issue #5437)

Custom Rhai scripts in the router now support the request.uri.port and request.subgraph.uri.port functions for reading and setting URI ports. These functions enable you to update the full URI for subgraph fetches. For example:

fn subgraph_service(service, subgraph){
    service.map_request(|request|{
        log_info(``);
        if request.subgraph.uri.port == {} {
            log_info("Port is not explicitly set");
        }
        request.subgraph.uri.host = "api.apollographql.com";
        request.subgraph.uri.path = "/api/graphql";
        request.subgraph.uri.port = 1234;
        log_info(``);
    });
}

By @lleadbet in #5439

🐛 Fixes

Fix various edge cases for __typename field (PR #6009)

The router now correctly handles the __typename field used on operation root types, even when the subgraph's root type has a name that differs from the supergraph's root type.

For example, given a query like this:

{
  ...RootFragment
}

fragment RootFragment on Query {
  __typename
  me {
    name
  }
}

Even if the subgraph's root type returns a __typename that differs from Query, the router will still use Query as the value of the __typename field.

This change also includes fixes for other edge cases related to the handling of __typename fields. For a detailed technical description of the edge cases that were fixed, please see this description.

By @IvanGoncharov in #6009

Support uri and method properties on router "request" objects in Rhai (PR #6147)

The router now supports accessing request.uri and request.method properties from custom Rhai scripts. Previously, when trying to access request.uri and request.method on a router request in Rhai, the router would return error messages stating the properties were undefined.

An example Rhai script using these properties:

fn router_service(service) {
  let router_request_callback = Fn("router_request_callback");
  service.map_request(router_request_callback);
}

fn router_request_callback (request) {
  log_info(`Router Request... Host: , Path: `);
}

By @andrewmcgivery in #6114

Cost calculation for subgraph requests with named fragments (PR #6162)

In some cases where subgraph GraphQL operations contain named fragments and abstract types, demand control used the wrong type for cost calculation, and could reject valid operations. Now, the correct type is used.

This fixes errors of the form:

Attempted to look up a field on type MyInterface, but the field does not exist

By @goto-bus-stop in #6162

Federation v2.9.3 (PR #6161)

This release updates to Federation v2.9.3, with query planner fixes:

• Fixes a query planning bug where operation variables for a subgraph query wouldn't match what's used in that query.
• Fixes a query planning bug where directives applied to __typename may be omitted in the subgraph query.
• Fixes a query planning inefficiency where some redundant subgraph queries were not removed.
• Fixes a query planning inefficiency where some redundant inline fragments in @key/@requires selection sets were not optimized away.
• Fixes a query planning inefficiency where unnecessary subgraph jumps were being added when using @context/@fromContext.

By @sachindshinde in #6161

[svc-apollo-docs]

✅ Docs Preview Ready

No new or changed pages found.

[router-perf]

CI performance tests

• connectors-const - Connectors stress test that runs with a constant number of users
• const - Basic stress test that runs with a constant number of users
• demand-control-instrumented - A copy of the step test, but with demand control monitoring and metrics enabled
• demand-control-uninstrumented - A copy of the step test, but with demand control monitoring enabled
• enhanced-signature - Enhanced signature enabled
• events - Stress test for events with a lot of users and deduplication ENABLED
• events_big_cap_high_rate - Stress test for events with a lot of users, deduplication enabled and high rate event with a big queue capacity
• events_big_cap_high_rate_callback - Stress test for events with a lot of users, deduplication enabled and high rate event with a big queue capacity using callback mode
• events_callback - Stress test for events with a lot of users and deduplication ENABLED in callback mode
• events_without_dedup - Stress test for events with a lot of users and deduplication DISABLED
• events_without_dedup_callback - Stress test for events with a lot of users and deduplication DISABLED using callback mode
• extended-reference-mode - Extended reference mode enabled
• large-request - Stress test with a 1 MB request payload
• no-tracing - Basic stress test, no tracing
• reload - Reload test over a long period of time at a constant rate of users
• step-jemalloc-tuning - Clone of the basic stress test for jemalloc tuning
• step-local-metrics - Field stats that are generated from the router rather than FTV1
• step-with-prometheus - A copy of the step test with the Prometheus metrics exporter enabled
• step - Basic stress test that steps up the number of users over time
• xlarge-request - Stress test with 10 MB request payload
• xxlarge-request - Stress test with 100 MB request payload