Skip to content
This repository has been archived by the owner on Nov 29, 2024. It is now read-only.

Latest commit

 

History

History

tools

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
README.md
README.md
 
 
canary_latencies.py
canary_latencies.py
 
 

README.md

Canary Latencies Script

Table of Contents

  1. Description
  2. Prerequisites
  3. Usage
  4. Resources

Description

This is a script to calculate the average producer/consumer/connection latencies from canary logs. This can be useful to determine if the histogram buckets for metrics are set optimally.

It also sorts the information into quantiles and returns them based on the number of cuts specified.

Prerequisites

You must have access to the logs from a canary that's connected to a Kafka cluster (and configured to have the "VERBOSITY_LOG_LEVEL" turned up to "1"), either stored in files or available to be streamed in.

Usage

The basic usage of the script:

python canary_latencies.py -f <your-file(s)>

If you would like to use it with input from STDIN, remove the -f flag:

cat <your-file> | python canary_latencies.py

Changing Defaults

  • Number of Cuts

By default, the number of cuts in the quantile method is set to 4. To change this, run the script with the following flag:

python canary_latencies.py -c <your-number-of-cuts>

  • Quantile Method

The quantile method being used is exclusive by default. To change this, run the script with the following flag:

python canary_latencies.py -m inclusive

Resources