Logan is a log analysis tool based on Django Meerkat. For now it is a full Django project, but the goal is to refactor it in a Meerkat plugin once plugins are supported by Meerkat.
This tool is not installable through pip yet.
Clone the repository with git clone git@gitlab.com:pawamoy/logan
.
This project is executed through Docker.
Run the whole building process by typing make
or make all
,
or follow the step-by-step build below:
make build
to build the Docker images;make up-no-start
to initialize the containers. It will actually create the database in the volume, which is necessary before trying to connect to the database in the next step;make build-database
to initialize the database (create the tables);make build-superuser
to create a super-user within the application;- See Loading fixtures and Running sections.
Copy your NginX logs file in the repository.
The file must be named nginx-access.log
.
Now run make load-fixtures
to load the logs into the database.
To run the project, use the make up
command.
Then go to http://localhost:8000/admin
and login with your super-user name and password.
Once logged in, go to the requests logs page in the admin interface at http://localhost:8000/admin/meerkat/requestlogs. Use the available buttons and links to filter your selection.
Select one or more or all items in the current filtered queryset, then click on the action called "Generate analysis for the selected items."
Once the analysis is done, a message will appear with the link to see it, something like /analysis/1.
Click on the link to go to the analysis page.
Use the mouse wheel to zoom/unzoom the graph.
You can see all the analysis you generated on the home page.