tree: d8ea85898a4a8d1ec244adcf678cfbaf0218ee47 [path history] [tgz]
  1. Dockerfile
  2. README.md
  3. analysis_options.yaml
  4. bin/
  5. lib/
  6. pubspec.lock
  7. pubspec.yaml
current_results/README.md

Current results of Dart CI

This server runs on Google Cloud Run, and provides the most recent test results from the Dart CI (continuous integration) testing visible at https://ci.chromium.org/p/dart/g/be/console

Build and deploy

The service takes about forty seconds to load all the existing latest results.json files from all the configurations, so its reponse to the first request after deploying will take at least that long. When an instance is started, a health check verifies that it is responding to web requests, and this health check has a 4 minute timeout. An instance is also started, and the health check run, when deploying a new version.

Generating protogen classes

To generate the Dart code reading the records from results.json, and the gRPC server and client code from query.proto, run protogen on the .proto files declaring them:

protoc --dart_out=lib/src/generated ../common/result.proto
protoc --dart_out=grpc:lib/src/generated -I/usr/local/include -Ilib/protos lib/protos/query.proto
dartfmt -w lib/src/generated

Our gRPC api protocol uses the google/protobuf/Empty message, and we use the gRPC api of Pub/Sub from google/pubsub/v1, so we need to check out the googleapis protocol buffer definitions and generate them. They are not checked into the repository, and must be generated.

protoc --dart_out=lib/src/generated -I/usr/local/include /usr/local/include/google/protobuf/*.proto
protoc --dart_out=grpc:lib/src/generated -I/src/googleapis /src/googleapis/google/pubsub/v1/pubsub.proto

Staging

To build the server and deploy to cloud run on staging, run

gcloud builds submit --project=dart-ci-staging --tag gcr.io/dart-ci-staging/current_results
gcloud run deploy --project=dart-ci-staging current-results --image gcr.io/dart-ci-staging/current_results --platform=managed --allow-unauthenticated

Unauthenticated access is allowed only while the server being developed does not modify any data, remove that flag before adding APIs that modify data.

Production XXX NOT READY YET

To build the server and deploy to cloud run on production, run

gcloud builds submit --project=dart-ci --tag gcr.io/dart-ci/current_results
gcloud run deploy --project=dart-ci current-results --image gcr.io/dart-ci/current_results --platform=managed

Services

Data cached by server

The server reads the current results.json and flaky.json from cloud storage at gs://dart-test-results/configuration/master/[configuration name]/[build number]/ and stores selected fields from those records in memory. (Flaky.json reads not implemented).

Updating results data

The server exposes a service to Pub/Sub on the same project, which is triggered by any object creation on the dart-test-results bucket. If the object is a /configuration/master/[configuration name]/latest file, the build number is read from that file and the cached results.json and flaky.json are updated from that build.

Serving results data

A gRPC service is defined, with a query message type and a response message type defined by protocol buffers. It will allow queries by test name or test name prefix and by configuration or configuration set.

Managing flakiness

APIs will be defined to query current flakiness data, and to revoke the flakiness mark on test-configuration pairs. When designing the API to revoke flakiness, security and auditing concerns must be addressed by requiring authentication.