CONFIG.md

How to configure the monitor

1. Copy config/config.js to config/config.local.js Do not modify config.js, this could prevent you from pulling updates from GitHub.
2. Edit config.local.js as needed.

A note on peerId

The monitor uses libp2p to connect with other monitor peers. At 1st startup, each monitor will generate a peerId (saved as keys/peer-id.json) The local monitor id (peerId) is printed to console at startup:

Our monitorId  12D3KooWH1XvGgPjRoMLi4tykATZ8UUcKng8sRU8WcmftoW1ZvJh

Example config.local.js file

WARNING: config/config.local.js potentially contains password and other sensitive info. This file is excluded in .gitignore. DO NOT CHECK THIS FILE IN!

// TODO: check if this is useful? process.env.GOSSIP_PORT could be set in the Dockerfile?
const GOSSIP_PORT = 30000`

// TODO: check if this is useful? process.env.HTTP_PORT could be set in the Dockerfile?
const HTTP_PORT = process.env.HTTP_PORT || 30001`

// This is the dateTime format used on the user interface
dateTimeFormat: 'DD/MM/YYYY HH:mm',`

// Connection to the datastore
sequelize: {
  database: 'ibp_monitor',
  username: 'ibp_monitor',
  password: 'ibp_monitor',
  options: {
    dialect: 'mariadb',
    // hostname = docker service name
    host: 'ibp-datastore',
    port: 3306,
    // optionally turn on Sequelize logs 
    logging: false,
  }
}

// Connection to Redis
redis: {
  // hostname = docker service name
  host: 'ibp-redis',
  port: 6379
},

// @deprecated: the peerId is calculated at 1st run
peerId: {
  // each member should register a known peerId
},

// List of other monitors (peerId is printed on startup)
knownPeers: [
  // TODO: list all peers/members public keys...?
  // Not currently used.
],
// TODO: amend lib/MessageHandler to check `cfg.knownPeersOnly` when gossip messages.
knownPeersOnly: false,

// http port for frontend
httpPort: HTTP_PORT,

// tcp port for gossip
listenPort: GOSSIP_PORT,

// allowed topics (channels) for the p2p protocol
allowedTopics: [
  // TBC
  '/ibp',
  // publish a list of services
  '/ibp/services',
  // publish a list of results
  '/ibp/healthCheck'
],

// how often to advertise our services and perform healthchecks
updateInterval: 30 * 1000, // 30 seconds

// where to find bootstrap peers (monitors)
bootstrapPeers: [
  '/ip4/31.22.13.147/tcp/30000/p2p/12D3KooWK88CwRP1eHSoHheuQbXFcQrQMni2cgVDmB8bu9NtaqVu',
  '/dnsaddr/boot.metaspan.io/tcp/30000/p2p/12D3KooWK88CwRP1eHSoHheuQbXFcQrQMni2cgVDmB8bu9NtaqVu',
],

// should we healthCheck our own services?
checkOwnServices: false,

// should we healthCheck services of other monitors?
checkOtherServices: true,

// send out results to p2p peers
gossipResults: true,

// libp2p: allow our node to relay messages to other nodes
relay: null,

// IMPORTANT
// these services will be advertised to other nodes
services: [
// put these in your config.local.js
  {
    serviceUrl: "wss://ibp-rpc.metaspan.io/westend",
    name: "Metaspan Westend RPC",
    chain: "westend",
  }
],

// prune the datastore
pruning: {
  // age of message
  age: 90 * 24 * 60 * 60, // 90 days as seconds
  // how often to prune
  interval: 1 * 60 * 60  // 1 hour as seconds
},

// not used: TODO, implement full SLA monitoring
performance: {
  sla: 500 // ms - used for performance graph, and later for alerts
}