AOSSIE-Org / AOSSIE-Org/Devr.AI

BUG: Silent Exceptions in Backend Initialization Are Not Logged

Offen
#259 3 Kommentare 0 Reaktionen 0 zugewiesene Personen Auf GitHub ansehen
Vorherrschende Sprache
Python
Sterne
102
Forks
137
PR-Merge-Kennzahlen
Keine gemergten PRs in 30 T.

Beschreibung

### Is there an existing issue for this?

- [x] I have searched the existing issues

### What happened?

## 📌 Issue Overview
There are several places in the backend where exceptions are being "swallowed" (ignored) by using `except Exception: pass`. This pattern is used, for example, when setting up database indices or initializing service components. When an error occurs within these `try` blocks, no message is printed or logged, and the backend continues running as if nothing is wrong—even though some features may break or behave unpredictably.

**Location:**
Most notably found in files under:
- `backend/app/database/falkor/code-graph-backend/api/graph.py`

## 🔍 Steps to Reproduce

1. Open the Python file in backend/app/database/falkor/code-graph-backend/ api/graph.py
2. Search for except Exception: pass
3. Trigger or simulate an error inside one of these try blocks (e.g., fail a database/index creation)
4. Check the logs — notice that no error is recorded and the application continues running silently

## 🎯 Expected Behavior
All exceptions during initialization and setup should be logged using Python’s logging module with descriptive messages and context

## 🚨 Actual Behavior
Exceptions raised during backend initialization or setup are silently ignored due to `except Exception: pass`.
No logs or error messages are generated, allowing failures to go unnoticed.

## 📷 Screenshot

## 💡 Suggested Improvements
try:
self.g.create_node_range_index("File", "name", "ext")
except Exception as exc:
logging.error(f"Failed to create node range index: {exc}")

### Record

- [x] I agree to follow this project's Code of Conduct
- [x] I want to work on this issue

Beitragsleitfaden

Für dieses Repository ist kein Beitragsleitfaden indexiert

Bewertung

Dieses Issue wurde noch nicht bewertet.

Neue Issues direkt in Ihr Postfach

Eine kurze Übersicht über anfängerfreundliche GitHub-Issues.