mirror of
https://github.com/Significant-Gravitas/AutoGPT.git
synced 2026-02-10 14:55:16 -05:00
198b3d9f45
Graph execution that fails due to interruption or unknown error should be enqueued back to the queue. However, swallowing the error ends up not marking the execution as a failure. ### Changes 🏗️ * Avoid keep updating the graph execution status on each node execution step. * Added a guard rail to avoid completing graph execution on non-completed execution status. * Avoid acknowledging messages from the queue if the graph execution is not yet completed. ### Checklist 📋 #### For code changes: - [x] I have clearly listed my changes in the PR description - [x] I have made a test plan - [x] I have tested my changes according to the test plan: <!-- Put your test plan here: --> - [x] Run graph execution, kill the process, re-run the process --------- Co-authored-by: Swifty <craigswift13@gmail.com>