Week 4

(June 24, 2025 - June 30, 2025)

Meeting 1:

(June 25, 2025)

In our weekly sync-up call with Avinal and Shaheem, we discussed several important updates and next steps:

• Decided to switching from Minikube to Kind for a smoother, faster and more stable local development experience.

• Explored the root cause of the scheduler pod’s CrashLoopBackOff and potential fixes.

• We finalized the agenda for upcoming weeks, which includes stabilizing the scheduler pod, refining the microservices setup, and polishing deployments for long-term maintainability.

• My mentors introduced me to rebasing in version control. They explained its significance in maintaining a clean Git history and walked me through how to rebase, resolve conflicts and prepare a clean commit for review.

• After this meeting, I successfully applied the guidance, created a clean commit and pushed the updated microservices configuration to the forked repository for mentor review.

Submitted a Functional Kubernetes-Based FOSSology Setup

I finalized and submitted a fully working version of the FOSSology microservices-based setup. After addressing deployment issues over the past weeks, I bundled all the required configuration files, updated manifests and Dockerfiles into a single, functional version and then sent to my Mentor for review.

All changes were pushed to Omar’s forked repository under the following branch:

• https://github.com/SalmanDeveloperz/fossology/tree/OmarAbdelSamea/GSoC/Microservices

The associated commit can be viewed here:

• https://github.com/SalmanDeveloperz/fossology/commit/1af934a48c449a2aacdb3bdec285a79534488db4

Highlights of the Commit:

• Polished Dockerfiles and .yaml files for successful image builds

• Web and scheduler pod configuration fixes

• Init containers added to manage PostgreSQL readiness

• Verified end-to-end setup with FOSSology UI loading successfully

Switching from Minikube to Kind:

Initially, Minikube was used for Kubernetes orchestration, but we decided to switch to Kind due to several limitations:

Minikube	Kind
Uses Virtual Machines (VMs)	Uses Docker containers
Slower startup and resource-heavy	Lightweight and faster startup
Port forwarding can be inconsistent	Easier and more reliable port handling
Persistent volumes sometimes reset on reboot	More stable volume management
Difficult to debug inside container network	Easier debugging with native Docker tooling

Deep Dive into Scheduler Pod CrashLoopBackOff

A significant part of the week was spent troubleshooting why the scheduler pod was crashing repeatedly. With Kind in place, I could gather more reliable logs and iterate quickly.

Areas investigated:-

• PostgreSQL schema migration errors

• Improper environment variables

• Readiness and liveness probes

• Docker image and startup sequences

While a final fix is still pending, we’ve narrowed down the potential issues and documented our findings to streamline the next round of debugging.

Meeting 2:

(June 26, 2025)

During the second weekly community call, I shared:

• Progress on the repository and successful UI

• The decision and rationale for moving to Kind

• Troubleshooting steps taken for the scheduler

What’s next for Week 5:

For Week 5, I’m ready to dive in and:

• Fix the scheduler pod’s CrashLoopBackOff by resolving the PostgreSQL schema error and making database updates work perfectly.

• Get the upload and copyright features working on the FOSSology UI so everything runs smoothly.

• Clean up all my changes in files, making them neat and work on Mentor’s feedback.