Skip to content

Latest commit

 

History

History
125 lines (92 loc) · 4.04 KB

CONTRIBUTING.md

File metadata and controls

125 lines (92 loc) · 4.04 KB
Raw

How to Contribute

Anyone is welcome to contribute through the official repository on github:

git clone -b dev https://github.com/ntop/ntopng.git

If you want to contribute with a patch, the first step to get it in the main tree is to run the regression tests included in the ntopng source code.

Run the Tests

An automated test suite is available under ntopng/tests, in order to run it:

  1. Compile ntopng. Please read the 'Memory Leak Detection' section in order to also enable the runtime memory leak detector.

  2. Make sure you have all the prerequisites installed:

  • redis-cli
  • curl
  • jq
  • shyaml (pip install shyaml)
  1. Run the run.sh script:
cd ntopng/tests
./run.sh

Please check that all the tests complete successfully before moving to the next step, sending the Pull Request.

Note: in case of failures, the output of the tests is stored in the ntopng/tests/rest/conflicts/ folder, in order to be able to compare it with the expected output in the ntopng/tests/rest/result folder. In case of test failures due to errors or warnings in the ntopng trace, the full ntopng log is stored in the ntopng/tests/rest/logs/ folder.

Create a PR

Please check the official GitHub documentation for instructions for sending Pull Requests:

https://docs.github.com/en/github/collaborating-with-issues-and-pull-requests/creating-a-pull-request

After the submission the ntop core team will review the patches to make sure they are well written and do the right things. If everything goes well, patches are applied to the dev branch and will be included in the nightly builds.

Add a Test

When implementing a new feature, it is recommended to write a new regression test to test the feature. This is based on the Rest API: the first time the test is executed, the output (this should be in json format) is stored in the 'result' folder, subsequent executions will compare the output of the test with the old one to make sure it is still the same.

Creating a new test is as simple as creating a small .yaml file, with the test name as name of the file, under ntopng/tests/rest/tests/ (where API version should be at least the latest API, e.g. v2) containing the below sections:

  • input: the name of a pcap in the 'ntopng/tests/rest/pcap' folder containing some traffic to be provided to ntopng as input
  • localnet: the local network(s) as usually specified with the -m option in ntopng
  • pre: a bash script with commands to be executed before processing the pcap in ntopng (initialization)
  • post: a bash script with commands to be executed after the pcap has been processed by ntopng and generating some json output (using the Rest API)
  • ignore: fields from the output to be ignored when comparing the output with the old JSON (this is usually used to ignore time, date or other fields that can change over time)
  • options: a list of extra options for the ntopng configuration file

Example:

input: traffic.pcap

localnet: 192.168.1.0/24

pre: |
  curl -s -u admin:admin -H "Content-Type: application/json" -d '{"ifid": 0, "action": "enable"}' http://localhost:3333/lua/toggle_all_user_scripts.lua

post: |
  sleep 10
  curl -s -u admin:admin  -H "Content-Type: application/json" -d '{"ifid": 0, "status": "historical-flows"}' http://localhost:3333/lua/rest/v2/get/alert/data.lua

ignore:
  - date

options:
  - -F=nindex

In order to run a specific test and avoid running all the suite, it is possible to specify -y=/ when running the run.sh script under ntopng/tests/rest:

cd ntopng/tests/rest
./run.sh -y=v2/get_alert_data_01

Memory Leak Detection

It is recommended to also enable the runtime memory leak detector based on the clang sanitizer when running the tests. This would check both memory errors and leaks (with no performance overhead).

In order to enable the sanitizer, please compile ntopng as below.

cd ntopng
./autogen.sh
./configure --with-sanitizer
make

This will report at the end of each test any memory leak or bad memory access, if any.