Skip to content

[SN#140] Make AI understand NetworkManager logs

This is a proposed project for Google Summer of Code

  • Skill level: entry/junior

  • Skills required: ML, AI, git, python, shell, networking, markdown (for docs)

  • Mentor(s): Fernando F. Mancera, Iñigo Huguet, Wen Liang

  • Contacts (IRC & email): ffmancera&riseup.net ihuguet&redhat.com wenliang&redhat.com

  • Idea description:

AI summarizer or abstract generation is widely deployed and utilized in various applications,
which saves people’s time from reading lengthy or even garbled text.

Analyzing the NetworkManager log is hard for new developers or users, sometimes it is time-consuming to summarize the networking behavior from the NetworkManager log. With the assistance of an AI summarizer, we should expect the model to parse and understand the NM log, and give a summary of the networking behavior with a confidence score and the verbose level of the summary can also be easily controlled. With the assistance of this TUI tool, the developers and users can analyze the NetworkManager log more easily.

  • What are we looking for:

    • Get in touch with the Upstream (Fedora and NM)
    • Basic ML/AL and LLM knowledge is preffered.
    • Building a custom LLM which will be trained with dataset of NM and NMstate logs
    • Build TUI for score system and output
    • Write documentation
    • Python tests and CI automation

  • Notes & references:

https://networkmanager.dev/docs https://nmstate.io/ https://github.com/linux-system-roles/network https://fedoramagazine.org/writing-useful-terminal-tui-on-linux-with-dialog-and-jq/

  • Estimation Week1 Upstream and community bonding Week 2 Knowing mentors and Project structure Week 3 Basic training LLM Week 4 NM and all things around + Blog per 2 week Week 5 Sync up twice a week + public meeting + post updates in ML + Fedora Blog Eval : Week 6 Project stucture and maybe some system req Week 7 Generate outputs for a log Week 8 Parsing of the prompts + sanitize the output + sanitize the logs Week 9 Create the score system Week 10 Training the dataset + finding the dataset Week 11 Training the dataset Week 12 Training the dataset Week 13 Training the dataset Week 14 Evalate training and sum the findings Week 15 Final Eval

This issue ticket was originally created here on a Pagure repository, mentored-projects by FAS Fernando F. Mancera on Tue Feb 6 14:04:31 2024 UTC.

This issue ticket was automatically created by the Pagure Exporter.