Periodical feature health check workflow tools(aka Canary)

This README describes how to set up Cadence canary, different types of canary test cases, and how to start the canary.

Setup

Prerequisite: Cadence server

Canary test suite is running against a Cadence server/cluster. See documentation for Cadence server cluster setup.

Note that some tests require features like Advanced Visibility and History Archival.

For local server env you can run it through:

Docker: Instructions for running Cadence server through docker can be found in docker/README.md. Either docker-compose-es-v7.yml or docker-compose-es.yml can be used to start the server.
Build from source: Please check CONTRIBUTING for how to build and run Cadence server from source. Please also make sure Kafka and ElasticSearch are running before starting the server with ./cadence-server --zone es start. If ElasticSearch v7 is used, change the value for --zone flag to es_v7.

Run canary

Different ways of start the canary:

1. Use docker image `ubercadence/cadence-canary:master`

You can pre-built docker-compose file to run against local server In the docker/ directory, run:

docker compose -f docker-compose-canary.yml up

This will start the canary worker and also the cron canary.

You can modify the canary worker config to run against a prod server cluster:

Use a different mode to start canary worker only for testing
Update the config to use Thrift/gRPC for communication
Use a different image than master tag. See docker hub for all the images. Similar to server/CLI images, the master image will be built and published automatically by Github on every commit onto the master branch.

2. Build & Run

In the project root, build cadence canary binary:

make cadence-canary

Then start canary worker & cron:

./cadence-canary start

This is essentially the same as

 ./cadence-canary start -mode all

By default, it will load the configuration in config/canary/development.yaml. Run ./cadence-canary -h for details to understand the start options of how to change the loading directory if needed.

To start the worker only for manual testing certain cases:

 ./cadence-canary start -mode worker

3. Monitoring

In production, it's recommended to monitor the result of this canary. You can use the workflow success metric emitted by cadence history service workflow_success. To monitor all the canary test cases, use workflowType of workflow.sanity.

Configurations

Canary workers configuration contains two parts:

Canary: this part controls which domains canary workers are responsible for what tests the sanity workflow will exclude.

yaml

canary:
  domains: ["cadence-canary"] # it will start workers on all those domains(also try to register if not exists)
  excludes: ["workflow.searchAttributes", "workflow.batch", "workflow.archival.visibility", "workflow.archival.history"] # it will exclude the three test cases. If archival is not enabled, you should exclude "workflow.archival.visibility" and"workflow.archival.history". If advanced visibility is not enabled, you should exclude "workflow.searchAttributes" and "workflow.batch". Otherwise canary will fail on those test cases.
  cron:
    cronSchedule: "@every 30s" #the schedule of cron canary, default to "@every 30s"
    cronExecutionTimeout: 18m #the timeout of each run of the cron execution, default to 18 minutes
    startJobTimeout: 9m #the timeout of each run of the sanity test suite, default to 9 minutes

An exception here is HistoryArchival and VisibilityArchival test cases will always use canary-archival-domain domain.

Cadence: this control how canary worker should talk to Cadence server, which includes the server's service name and address.

yaml

cadence:
  service: "cadence-frontend" # frontend service name
  address: "127.0.0.1:7833" # frontend address
  #host: "127.0.0.1:7933" # replace address with host if using Thrift for compatibility
  #tlsCaFile: "path/to/file" # give file path to TLS CA file if TLS is enabled on the Cadence server
  #metrics: ... # optional detailed client side metrics like workflow latency. But for monitoring, simply use server side metrics `workflow_success` is enough.

Metrics: metrics configuration. Similar to server metric emitter, only M3/Statsd/Prometheus is supported.
Log: logging configuration. Similar to server logging configuration.

Canary Test Cases & Starter

Cron Canary (periodically running the Sanity/starter suite)

The Cron workflow is not a test case. It's a top-level workflow to kick off the Sanity suite(described below) periodically. To start the cron canary:

 ./cadence-canary start -mode cronCanary

For local development, you can also start the cron canary workflows along with the worker:

 ./cadence-canary start -m all

The Cron Schedule is from the Configuration. However, changing the schedule requires you manually terminate the existing cron workflow to take into effect. It can be improved in the future.

The workflowID is fixed: "cadence.canary.cron"

Sanity suite (Starter for all test cases)

The sanity workflow is test suite workflow. It will kick off a bunch of childWorkflows for all the test to verify that Cadence server is operating correctly.

An error result of the sanity workflow indicates at least one of the test case fails.

You can start the sanity workflow as one-off run:

cadence --do <the domain you configured> workflow start --tl canary-task-queue --et 1200 --wt workflow.sanity -i 0

Or using the Cron Canary mentioned above to manage it.

Then observe the progress:

cadence --do cadence-canary workflow ob -w <...workflowID form the start command output>

NOTE 1:

tasklist(tl) is fixed to canary-task-queue
execution timeout(et) is recommended to 20 minutes(1200 seconds) but you can adjust it
the only required input is the scheduled unix timestamp, and 0 will uses the workflow starting time

NOTE 2: This is the workflow that you should monitor for alerting. You can use the workflow success metric emitted by cadence history service workflow_success. To monitor all the canary test cases use workflowType of workflow.sanity.

NOTE 3: This is the list of the test cases that it will start all supported test cases by default if no excludes are configured. You can find the workflow names of the tests cases in this file if you want to manually start certain test cases.

Echo

Echo workflow tests the very basic workflow functionality. It executes an activity to return some output and verifies it as the workflow result.

To manually start an Echo test case:

cadence --do <> workflow start --tl canary-task-queue --et 10 --wt workflow.echo -i 0

Then observe the progress:

cadence --do cadence-canary workflow ob -w <...workflowID form the start command output>

You can use these command for all other test cases listed below.