Diagnosed and Troubleshot ActivityWatch on Linux

  • Day: 2026-03-29
  • Time: 09:30 to 09:55
  • Project: Dev
  • Workspace: WP 2: Operational
  • Status: Completed
  • Priority: MEDIUM
  • Assignee: Matías Nehuen Iglesias
  • Tags: Activitywatch, Linux, Troubleshooting, SQL, Database

Description

Session Goal

The session aimed to diagnose and troubleshoot issues with the ActivityWatch service on Linux, focusing on data integrity, API responses, and database verification.

Key Activities

  • Searched for ActivityWatch documentation and troubleshooting resources, focusing on starting the service and resolving issues with the aw-qt interface using official documentation and GitHub.
  • Executed troubleshooting steps to recover the ActivityWatch application, including checking the web UI, killing stale processes, and manually starting components.
  • Diagnosed potential data loss by verifying database integrity and UI issues, using specific commands to check database file size, bucket contents, and event data.
  • Verified API responses and database file existence to address potential data loss and redirect handling issues.
  • Reflected on the ActivityWatch database status, confirming the existence of historical data despite misleading UI indications.
  • Inspected the ActivityWatch database schema to identify table structures and correct SQL query assumptions.
  • Compiled a comprehensive guide for operating and troubleshooting ActivityWatch on Ubuntu, detailing system architecture, diagnosis steps, and recovery procedures.

Achievements

  • Successfully identified and corrected misunderstandings in SQL queries related to the ActivityWatch database schema.
  • Developed a structured guide for diagnosing and resolving ActivityWatch service issues, including SQL corrections for historical data inspection.

Pending Tasks

  • Further verification of older historical data existence in the SQLite database through specific queries.
  • Continuous monitoring and refinement of the ActivityWatch troubleshooting guide to incorporate new insights as they arise.

Evidence

  • source_file=2026-03-29.sessions.jsonl, line_number=1, event_count=0, session_id=2dfdbe6ca79f6a023857ed40817743ea4ca080d0bf17ed0fe0ba4941d9e2de6e
  • event_ids: []