We have rolled out an important backend update to optimize our LINE Bot integration, specifically resolving processing timeouts and duplicate message loops during AI data analysis for our Premium and Enterprise members.
🔍 The Issue We Found: When generating detailed AI summaries, the processing time took several seconds. This caused LINE reply tokens to expire or hit abrupt ECONNRESET disconnections. As a result, teachers didn't receive their AI summaries, and LINE kept re-delivering the same webhook over and over, causing annoying duplicate message triggers.
⚡ What's Fixed & Improved:
- Immediate 200 OK Responses: The server now instantly acknowledges LINE's webhooks, pushing the heavy AI analytical work directly into background processing. This eliminates redundant delivery loops and duplicate bot responses.
- Smart Reply-to-Push Fallback: If a free
replyMessagefails due to an expired token after a long AI generation session, the system automatically falls back to a standardpushMessage. This guarantees that teachers receive their AI reports 100% of the time, while seamlessly conserving your monthly quota during normal operations.
Your LINE AI assistant is now faster, more reliable, and ready to support your classroom workflow smoothly!