Public | Automated Build

Last pushed: 3 years ago
Short Description
Short description is empty for this repo.
Full Description


A simple container that you can use to run
HTTP load tests through the awesome vegeta.

~/projects/docker-vegeta (master ✘)✭ ᐅ docker run -ti vegeta
root@cb6740a3f8b7:/go# echo "GET" | vegeta attack -duration=3s -rate=5 -timeout=1m | tee results.bin | vegeta report
Requests    [total]                15
Duration    [total, attack, wait]        2.949280725s, 2.798913162s, 150.367563ms
Latencies    [mean, 50, 95, 99, max]        288.16816ms, 239.913779ms, 579.811387ms, 579.811387ms, 617.736089ms
Bytes In    [total, mean]            2441521, 162768.07
Bytes Out    [total, mean]            0, 0.00
Success        [ratio]                100.00%
Status Codes    [code:count]            200:15  
Error Set:


First go inside the contaner:

docker run -ti odino/vegeta

then have fun with vegeta:

echo "GET" | vegeta attack -duration=3s -rate=5 -timeout=1m | tee results.bin | vegeta report

Why not simply using the compiled vegeta binary from your machine?

Because this lets you do real world load tests from
a common hardware, without the need to spin up
good machines on a provider like AWS.

Vegeta is a very good tool but at the same time, on
a single HW it will hit system bottlenecks like
open files limit and so on -- not superfun at all!

Luckily you can simply virtualize some "machines"
through docker and trigger less stressing requests
from the containers themselves.

ie. on my machine I used something like:

echo "GET" | vegeta attack -duration=10m -rate=100 -timeout=1m | tee results.bin | vegeta report

and started seeing some funky error with file descriptors,
ulimit and so on.

You can instead launch 10 containers and simply do:

echo "GET" | vegeta attack -duration=10m -rate=10 -timeout=1m | tee results.bin | vegeta report

and everything will work like a charm :)


The container comes with vim preinstalled: this
is to facilitate your life if you need to play
around with your /etc/hosts (ie. you wanna stress
test different machines / setups).

Docker Pull Command
Source Repository