Add cleanup of old active jobs

- Remove active jobs older than 10 minutes to prevent accumulation
- Cleans up leftover jobs from previous runs or crashes
- Prevents duplicate job tracking issues

vidai/cluster_master.py

@@ -1400,6 +1400,16 @@ class ClusterMaster:
                         print(f"No suitable worker found for job {job['id']}, re-queuing")
                         update_queue_status(job['id'], 'queued', error='No suitable worker found, re-queued')
 
+                # Clean up old active jobs (older than 10 minutes)
+                current_time = time.time()
+                to_remove = []
+                for job_id, job_info in self.active_jobs.items():
+                    if current_time - job_info['start_time'] > 600:  # 10 minutes
+                        to_remove.append(job_id)
+                for job_id in to_remove:
+                    self.complete_job(job_id)
+                    print(f"Cleaned up old active job {job_id}")
+
                 # Check for cancelled jobs that need to be stopped
                 cancelled_jobs = []
                 for job_id, job_info in self.active_jobs.items():