HyperMeter: HTTP metric tool

Usage

npm install -g hypermeter
hypermeter my-config-file.json

Config file

The config file is a JSON file containing the following fields:

urls (required)

An array of URLs for hypermeter to measure

reporters (optional)

An object containing reporter/config pairs. The currently available reporters are:

• 'console' (which takes no further configuration options)
• 'json', which creates/updates a JSON file that can be used to draw graphs (see the wiki)
• 'graph', which makes use of plotly (see below)

If no reporters are specified, a single console reporter is used by default.

checks (optional)

An object containing check/config pairs. The currently available checks are 'status' (which takes no further configuration options), and 'time', which takes an integer specifying the maximum allowable response time (in milliseconds).

client (optional)

Settings for the HTTP client. If any settings are specified then hypermeter uses curlrequest and passes your settings through to cURL. Otherwise, the native node HTTP module is used.

Note that cURL is generally slower than a native HTTP call, so timings won't be consistent between the two.

Examples

A minimal config file might look like the following:

{
    "urls": [
        "https://twitter.com",
        "https://google.com",
        "https://facebook.com"
    ]
}

A more advanced config file using the graph reporter might look like the following:

{
    "urls": [
        ...
    ],
    "reporters": {
        "console": null,
        "graph": {
            "username": "my_plotly_username",
            "apiKey": "my_plotly_api_key",
            "build": "$BUILD_NUMBER",
            "filename": "plotly_filename",
            "fileId": "56"
        }
    },
    "checks": {
        "status": null,
        "time": 5000
    }
    "client": {
        "cert": "/etc/pki/my_cert.pem"
    }
}

Note that config values can be environment variables (prefixed by $).

If the graph reporter is used with no fileId specified, the filename will be created (or overwritten!) and the URL of the file will be output to the console. You can take the ID from this and add it to your config file for subsequent runs.