lis-hubs-infra-core

Part of the Livestock Information Service (LIS), which lets livestock keepers, owners, agents and delegates record and manage their animals.

@defra/lis-hubs-infra-core is a shared npm package with the runtime plumbing that every LIS hub and spoke needs: request context and correlation IDs, logging, calls to other LIS services with service-to-service tokens, and the hub's proxy to its spokes.

It deliberately does not cover authentication, sessions, roles or permissions (see lis-hubs-infra-access) or which modules a hub serves (see lis-hubs-infra-registry).

Used by:

• hubs: lis-hubs-front-office, lis-hubs-back-office
• spokes: lis-apps-cattle-home, lis-apps-cattle-register, lis-apps-cattle-move, lis-apps-cattle-death
• packages: lis-hubs-infra-access, lis-infra-ui-services

Installation

npm install @defra/lis-hubs-infra-core @defra/lis-hubs-infra-registry

@defra/lis-hubs-infra-registry is a peer dependency, so the consuming app must install it too.

Usage

Logging

logger is a single, shared pino logger that writes ECS format for CDP's OpenSearch. Log through it everywhere, including in request handlers. Don't use hapi's request.logger.

import { logger } from '@defra/lis-hubs-infra-core'

logger.level = config.get('log.level')
logger.enabled = config.get('log.enabled')
logger.format = 'ecs' // or 'pretty-print' for local development
logger.serviceName = 'lis-apps-cattle-home' // defaults to SERVICE_NAME
logger.serviceVersion = config.get('serviceVersion') // defaults to SERVICE_VERSION
logger.context.hashSecret = config.get('log.hashSecret')

await server.register(logger.hapiPlugin) // request and response logging

logger.info('Holding loaded')

logger.context holds request-scoped values that are added to every log line in the request. Pass true as the third argument to hash a value with hashSecret, so you can correlate identifiers without logging them:

logger.context.set('cph', cph, true)

Request context

requestContext.plugin takes the correlation ID from the x-cdp-request-id header, or generates one, and keeps it for the rest of the request. Register it before any other plugin.

import { requestContext } from '@defra/lis-hubs-infra-core'

await server.register(requestContext.plugin)

requestContext.getHeaders() // { 'x-cdp-request-id': '...' }, for your own outbound calls

Calling other LIS services

Extend BaseClient to call another LIS service (for example a spoke's be4fe). It:

• works out the service's base URL for the current environment (getServiceBaseUrl)
• adds the correlation ID header
• adds an Authorization: Bearer service token from AWS STS (GetWebIdentityToken), with the target service's name as the audience
• sends x-api-key if you give it an apiKey
• times out after 3 seconds by default (timeout, in milliseconds, is the total budget for the response headers and body; a timeout is thrown as a 504)
• holds a circuit breaker for the service (circuitBreaker sets its failureThreshold and resetTimeout, or false disables it)

import { BaseClient } from '@defra/lis-hubs-infra-core'

class CattleHomeBe4FeClient extends BaseClient {
  constructor({ environment, apiKey }) {
    super({
      environment,
      serviceName: 'lis-be4fe-cattle-home',
      port: 8087, // only used for 'local' and 'docker_compose'
      apiKey
    })
  }

  async getHolding(cph) {
    const { payload } = await this._get(`api/cphs/${encodeURIComponent(cph)}`)
    return payload
  }
}

_get, _post, _put, _patch and _delete take a path and Wreck options. If a service token can't be obtained, they throw ServiceTokenError; while the service's circuit is open, CircuitOpenError. Both carry a statusCode of 503. Other thrown errors carry statusCode, serviceName, method, path and cause for code that catches the error. The shared logger emits only the error's type, message and stack, so a caller that wants them in a log line must put them in the message or logger.context.

Locally, lis-dev's lis-fake-sts service stands in for AWS STS. See Service-to-service auth in lis-dev.

Hub proxy

createProxyPlugin is used by hubs. It registers a proxy route for every module that @defra/lis-hubs-infra-registry lists for the hub, forwarding requests to that module's spoke and forwarding only the hub JWT cookie and any cookies named in forwardedCookieNames (for example the cookie consent choice, which the spoke's layout needs to decide whether to show the banner), never the hub session cookie or the Authorization header.

import { createProxyPlugin } from '@defra/lis-hubs-infra-core'
import { COOKIES_POLICY_COOKIE_NAME } from '@defra/lis-infra-ui-services/cookies/cookies-policy'

const proxy = createProxyPlugin({
  hubId: 'front-office',
  environment: config.get('environment'),
  hubJwtCookieName: config.get('auth.hubJwt.cookieName'),
  forwardedCookieNames: [COOKIES_POLICY_COOKIE_NAME], // optional, none by default
  proxyTimeout: 30000, // optional, milliseconds to wait for a spoke's response headers; 30s by default
  circuitBreaker: { failureThreshold: 5, resetTimeout: 30000 } // optional, see below for the defaults
})

await server.register(proxy.plugin)

Each module is a Spoke, which owns the module's service name, base URL and circuit breaker and builds the h2o2 route. A spoke that does not answer within proxyTimeout gets a 504; one that cannot be reached, a 502.

Circuit breaker

CircuitBreaker protects calls to one downstream service, and is what BaseClient and the hub proxy use. Only a transport failure (connection refused, timeout) counts against it; any response, whatever its status, proves the service is up. Outcomes of calls that started before the circuit opened are ignored.

• Closed: calls go through. failureThreshold consecutive failures (5 by default) open it.
• Open: calls fail fast with CircuitOpenError, a 503, for resetTimeout milliseconds (30 seconds by default). The UI catch-all renders the "service unavailable, try again later" page.
• Half-open: one probe call goes through and the rest fail fast. Its success closes the circuit; its failure reopens it for another resetTimeout.

The state is per process, so each hub or spoke instance decides for itself. Every transition is logged through the shared logger: a warning when a circuit opens, info when it goes half-open and when the probe closes it.

import { CircuitBreaker } from '@defra/lis-hubs-infra-core'

const breaker = new CircuitBreaker({ name: 'lis-be4fe-cattle-home' })

breaker.assertCallAllowed() // throws CircuitOpenError while open; call immediately before the transport call
try {
  res = await Wreck.request('GET', url, options)
} catch (error) {
  breaker.recordFailure() // no response at all
  throw error
}
breaker.recordSuccess() // any response, whatever its status

Environment variables

The package reads these, falling back to them when the consumer doesn't set the values directly:

| Variable | Used for | | ---------------------- | ---------------------------------------------- | | SERVICE_NAME | logger.serviceName | | SERVICE_VERSION | logger.serviceVersion | | AWS_REGION | AWS STS region, for service tokens | | AWS_ENDPOINT_URL_STS | AWS STS endpoint override (lis-dev's fake STS) |

On CDP the platform sets these. In lis-dev, the compose files set them.

Prerequisites

• Node.js 24 (see .nvmrc, use nvm use)
• To check changes against a consuming app: a local clone of lis-dev with its one-time setup done

Setup

Development

nvm use
npm install

Test

There are no tools to install beyond npm install. Tests use Vitest.

Running in development

This is a library, so it has no server of its own. To try a change in a consuming app without publishing, link the local copy into the lis-dev stack:

cd lis-dev
.bin/sideload link hubs/infra-core # every installed copy now points at this repo
.bin/sideload up front-office       # optional: run a consumer locally with hot reload
.bin/sideload unlink hubs/infra-core # when done

Running tests

npm test                 # unit tests
npm run test:coverage    # with coverage
npm run lint             # ESLint
npm run format:check     # Prettier

Before releasing a change, also run the tests of each consuming repo against the linked package.

Pipelines

Build, versioning and publishing run in GitHub Actions (.github/workflows) using Cake scripts (*.cake):

• Pull request to main: validate (format, lint, test), Sonar analysis, version check
• Push to a feature/ or hotfix/ branch: prerelease version and tag
• Merge to main: release tag, then publish to npm

Versions come from git tags (SemVer). After a merge, wait for the publish job to finish before bumping the package in consuming repos.

Branching policy

The repo uses mainline development. Branch from main and name the branch:

• feature/LREG-<ticket>-<short-kebab-description>
• hotfix/LREG-<ticket>-<short-kebab-description>

LREG must be uppercase, because the GitHub ruleset rejects other branch names. Rebase onto main rather than merging main in.

Contributing to this project

• Follow the Defra software development standards.
• npm install sets up a pre-commit hook (Husky) that runs the security audit, format check and lint.
• Open a pull request against main using the PR template. CI must pass and the Sonar quality gate must be green before merge.

Licence

THIS INFORMATION IS LICENSED UNDER THE CONDITIONS OF THE OPEN GOVERNMENT LICENCE found at:

http://www.nationalarchives.gov.uk/doc/open-government-licence/version/3

The following attribution statement MUST be cited in your products and applications when using this information.

Contains public sector information licensed under the Open Government licence v3

About the licence

The Open Government Licence (OGL) was developed by the Controller of His Majesty's Stationery Office (HMSO) to enable information providers in the public sector to license the use and re-use of their information under a common open licence.

It is designed to encourage use and re-use of information freely and flexibly, with only a few conditions.