Status Page Controller

Automates actions on a Statuspage.io page based on New Relic Alerts.

It will change the status of status page components if the component names matches the exact name of a New Relic Alert Policy.

Mapping components

Mapping is done by exact name match with optional namespacing for component groups.

Example:

# New Relic alert policy name 
    Downloads
     
# Statuspage.io component name
    Downloads

If a New Relic incident is created for the "Downloads" policy, status page controller will update the matching "Downloads" component in Statuspage.io.

Component Groups

Component groups are mapped by namespacing the New Relic alert policy with the following convention:

"group-component"

Example mapping of component groups on status.redhat.com:

# New Relic alert policy names
access.redhat.com-Downloads
access.redhat.com-Knowledgebase
developers.redhat.com-Downloads
developers.redhat.com-Blog

# Statuspage.io component groups
access.redhat.com
    L Downloads
    L Knowledgebase
developers.redhat.com
    L Downloads
    L Blog

Tresholds

1. yellow = Incident is at least 10 minutes old
2. orange = Incident is at least 20 minutes old
3. red = incident is 30+ minutes old

NOTE: the above can be configured, see below.

Install

npm install statuspage-controller

Usage

With the default config

var StatuspageController = require('statuspage-controller')
var spc = new StatuspageController();
spc.start();

This usage will use all config defaults and expect the following environment variables to be set in order to work:

• NR_API_KEYS - Your New Relic API key(s), this can be a single key or comma separated list of keys
• SPIO_PAGE_ID - Your Statuspage.io Page ID
• SPIO_API_KEY - Your Statuspage.io API key

Specify a config object:

var StatuspageController = require('statuspage-controller')
var config = {
    POLL_INTERVAL: 10000,
    PORT: 8080,
    NR_API_KEYS: process.env.NR_API_KEYS,
    SPIO_PAGE_ID: process.env.SPIO_PAGE_ID,
    SPIO_API_KEY: process.env.SPIO_API_KEY,
};
var spc = new StatuspageController(config);
spc.start();

Configuring with custom thresholds:

var config = {
    POLL_INTERVAL: 10000,
    PORT: 8080,
    NR_API_KEYS: process.env.NR_API_KEYS,
    SPIO_PAGE_ID: process.env.SPIO_PAGE_ID,
    SPIO_API_KEY: process.env.SPIO_API_KEY,
    THRESHOLDS: [
        {
            "duration": 600,
            "status": "degraded_performance"
        },
        {
            "duration": 1200,
            "status": "partial_outage"
        },
        {
            "duration": 1800,
            "status": "major_outage"
        }
    ]
};
var spc = new StatuspageController(config);
spc.start();

Thresholds are in seconds. The above will change the component to degraded_performance after 600 seconds, and so on. You can have just one threshold if you want to change the component status to a single status after a given time. For example, if you wanted to go strait to major outage after 10 minutes then you would do:

THRESHOLDS: [
    {
        "duration": 600,
        "status": "major_outage"
    }
]

Debug

Enable extra debug logging to the console. Don't enable this in production.

var StatuspageController = require('statuspage-controller')
var config = {
    DEBUG: true,
};
var spc = new StatuspageController(config);
spc.start();

API

Methods

There are a few API methods that check the health of the controller and allow you to set component overrides:

• GET /api/healthcheck.json - Check the health of the controller
• POST /api/overrides.json - Create a temporary override of a given statuspage.io component
• GET /api/overrides.json - List current active overrides

Healthcheck

Request

curl http://localhost:8080/api/healthcheck.json

Response

{"message":"New Relic and statuspage.io connections established.","ok":true}

Overrides

Request

curl -X POST -H 'content-type:application/json' http://localhost:8080/api/overrides.json -d '{"component_name": "Downloads", "seconds":60, "new_status":"major_outage"}'

Response

{"message":"Successfully added override","component_name":"Downloads","seconds":60}

Authentication

Optionally you can enable basic auth for the API. You'll need to configure an htpasswrd file with user(s) created with htpasswd command.

1. Either have htpasswd command installed with Apache, or npm-htpasswd
2. Create a user and new htpasswd file: htpasswd -c /path/to/users.htpasswd myuser
3. Point the config at the password file: HTPASSWD_FILE: '/path/to/users.htpasswd

Example:

var StatuspageController = require('statuspage-controller')
var config = {
    HTPASSWD_FILE: '/path/to/users.htpasswd',
};
var spc = new StatuspageController(config);
spc.start();

SSL

Optionally you can configure the API server to use SSL. See node.js https

1. Either use an existing key and cert of create self-signed cert using the following method: openssl req -x509 -nodes -days 365 -newkey rsa:2048 -keyout /path/to/selfsigned.key -out /path/to/selfsigned.crt
2. Point the config at your key and cert files using the tls object:

Example:

var StatuspageController = require('statuspage-controller')
var config = {
    TLS: {
        key:  "/path/to/selfsigned.key",
        cert: "/path/to/selfsigned.crt",
    }
};
var spc = new StatuspageController(config);
spc.start();

Full Example

Here is an example with all optional config options set:

var StatuspageController = require('statuspage-controller')
var config = {
    POLL_INTERVAL: 10000,
    PORT: 8080,
    DEBUG: false,
    NR_API_KEYS: process.env.NR_API_KEYS,
    SPIO_PAGE_ID: process.env.SPIO_PAGE_ID,
    SPIO_API_KEY: process.env.SPIO_API_KEY,
    HTPASSWD_FILE: '/path/to/users.htpasswd',
    TLS: {
        key:  "/path/to/selfsigned.key",
        cert: "/path/to/selfsigned.crt",
    }
};
var spc = new StatuspageController(config);
spc.start();

Sync vs Push

Both New Relic and StatusPage.io have ways to automate via push. New Relic alerts can post to a webhook, or send an email, and StatusPage.io components can be updated with a unique email. The problem with this method is that if a message is ever lost, then the state between New Relic and StatusPage.io will get out of sync. Also you might not want to change the status of a component until a minimum threshold of failure has occurred.

By synchronizing both systems with their APIs, the states between the two will always be kept in sync, even if an alert message is never received. It also updates states faster than email. As soon as a violation is created it will be picked up on the next sync. If Status Controller ever goes down, the next time it is started it will sync the statuses to their current state.