V2

CQ V2

ARCHITECTURE

COIN SUPPORT

• BTC
• BTCTESTNET
• BCH
• DGB
• LTC

ADDING COINS

• Add coin configuration using default values to getCoins function in infra/config.ts
• Add coin configuration to config/<prod&&stage>.json
• Get in touch with SRE so that they can provision a database to host this new coin
• Add schemaPrefix to pkg/postgres/postgres.go
• Add coin template to ingester-sql-ddl
• Run sqitch against new database to ensure schema is valid (see build/ingester-sql-ddl/README.md)

INSTALL DEPENDENCIES

• Install ZMQ using your package manager of choice brew install zmq

DEVELOPMENT

SETUP

• cp sample.env .env and add your AWS creds for KMS

RUNNING THE API

• make dev-{ticker} (lowercase ticker) will spin up your dev environment with the db deployed with sqitch (ingester-sql-ddl) and api service with hot reloading.
• DB=stage make dev-{ticker} to connect the api to the staging database

RUNNING THE INDEXER

• go run cmd/indexer/indexer -coin [coin] -start [startBlock] -end [endBlock] the -sync flag can be used to sync from last block in db

RUNNING THE MONITOR

• go run cmd/monitor/main.go -config={absolute-path-to}/config.json -coin={coin}

View your GoDocs

• Run make godoc
• Open browser to localhost:3000
• Click packages button on top nav
• Click Third party link under Packages
• Navigate to shapeshift-legacy > coinquery > V2 and you will see our packages

PORTS

4000 => API
4001 => Monitor

DEPLOYMENT

• To change the coins deployed via circle-ci (and depending on what coins you want to deploy to which environment) update the STAGE_COINS and PROD_COINS env vars in circle-ci. Format is comma delimited - no spaces - all lowercase

  Prod coins

  btc,bch,dgb,ltc
  

  Stage coins

  btc,bch,dgb,ltc
  

• Make sure to create the following ECR repos for the new coin:

  • coinquery/{environment}-monitor-{coin}
  • coinquery/{environment}-indexer-{coin}
  • coinquery/{environment}-api-{coin}

Profiling

• local builds have pprof support enabled by default that you can use to extract cpu and memory profiles

• some very basic memory and go-routine information can be found by navigating to http://localhost:4000/debug/pprof/

• To create a cpu or memory profile, activate profiling with the following command

  go tool pprof -seconds 30 http://localhost:4000/debug/pprof/profile
  // Then proceed to put the server under load by making requests to the endpoints you care about
  

• To interactively explore your generated profile

  pprof -http localhost:8080 {profile_file_name}
  

Tests

• use make target test to run all unit tests and integration tests
• set TEST_PATH to only run tests in the specified directory. default all directories
• set TIMEOUT to specify a timeout limit. default 10m

Unit

• use the build tag unit for any unit test suits
• use package_unit_test.go for test file
• use make target test-unit to run all unit tests

Integration

• use the build tag integration for any unit test suits
• use package_integration_test.go for test file
• use make target test-integration to run all unit tests

Benchmarks

• use the build tag benchmark for any benchmark test suites
• use package_bench_test.go for test file
• use make target test-bench to run all benchmark tests
• set BENCH_TIME to a duration (1s for 1 second) or count (1x for 1 execution) for more granularity. default 1s
• update db uri in test as needed for db/api benchmarks note that using the default duration for benchmarks in the insight package will fail. use count in this case

HTTP API LOAD TESTING

• make attack see ./vegeta/README.md for usage