docker-log-gen

A quick and dirty tool to spew logs for benchmarking purposes. Can operate in several modes:

• As a standalone container, using command-line flags to determine logging quantity and options. Acts in a similar manner to ab
• As a web server, using query-string options to determine logging quantity and options
• As part of a container orchestration system, like DC/OS or k8s, using the system's healthcheck to periodically force log generation

Basic CLI Usage

• docker run bryanlatten/docker-log-gen
• Spits out random lorem ipsum-generated log lines on a timer
• Customize output based on the following command-line options:

  Options:

    -V, --version             output the version number
    -n, --number <size>       Number of outputs to perform (default 10 )
    -l, --length <size>       Bytes to log per output (default 64 bytes)
    -d, --dated               Include date per output (default 0)
    -c, --counter             Include a consistently-spaced counter to each output (default 1)
    -i, --identifier <text>   Add an identifier per outputs
    -r, --repeat <delay>      Repeat output every X seconds (default 2 sec), 0 to disable, incompatible with -w
    -s, --separator <symbol>  Use a symbol to separate log components (default  |  )
    -m, --multiline           Add newline and tab control characters into each output
    -w, --web                 Launch an HTTP server to log only during requests (default port = 3000)
    -h, --help                output usage information

Web Server Usage

• Run container with its exposed web port, docker run -p 3000:3000 bryanlatten/docker-log-gen -w
• Spits out random lorem ipsum-generated log lines per request to / on the web interface
• Customize the output based on the following querystring parameters:

Variable	Example	Default	Description
n	n=100	1	Number of log lines to produce per request
l	l=1024	64	Number of bytes to log per output
d	d=0	1	Adds ISO 8601 date to each output
i	i=TRANSACTION_ID	(none)	Identifier to add per line
s	s=/	|	Separates line components from each other
c	c=0	1	Adds a consistently-spaced counter to each output

Clustered Usage, via DC/OS (Marathon) or Kubernetes

Run the container, using the healthcheck's frequency + query-string parameters to predictably control the scale of throughput. Or, hit / with standard benchmarking tools (ab, jmeter, gatling, locust.io) for maximum output.

• examples/marathon.json: sample DC/OS application definition
• examples/kubernetes.yaml: sample Kubernetes Pod definition

Local Development

1. Clone repo, cd docker-log-gen
2. Install dependencies, npm i
3. Run the app ./bin/start or node ./bin/start