Task Execution Monitoring & Result Parsing
After a workflow dispatches tasks to scanning nodes, you can monitor them via task details, real-time logs, and result parsing.
1. Task Details & Status
Click "Detail" on any row in the Run Center's task list to inspect task status:
Task Status
| Status | Icon | Description |
|---|---|---|
| WAITING | 🕗 | Waiting for upstream dependencies to finish |
| PENDING | ⏳ | Task queued, waiting for an online scanning node |
| ASSIGNED | 📌 | Task picked up by a scanning node |
| RUNNING | 🔄 | Node is executing the task, streaming logs |
| PARSING | 🧩 | Command finished; the output parser pipeline is processing results |
| COMPLETED | ✅ | Command exited successfully, output parsed and saved |
| FAILED | ❌ | Non-zero exit code or execution error |
| PARSE_FAILED | ⚠️ | Command finished but output parsing failed |
| TIMEOUT | ⏱️ | Execution exceeded the configured timeout |
| CANCELLED | 🚫 | Cancelled by the user |
| SKIPPED | ⏭️ | Skipped due to empty upstream output or unmet condition |
Heartbeats & Auto-Retries
- Heartbeat Tracking: The system continuously monitors node health. Disconnected or crashed nodes trigger automatic task handling.
- Automated Retries: If the workflow specifies
maxRetries, transient failures automatically re-queue the task.
Manual Interventions
- Force Kill: Click "Kill" on a stuck
RUNNINGtask to immediately stop and clean up the container. - Single Task Re-run: Click "Re-run" on a failed task to re-dispatch execution.
2. Real-Time Logs
Without SSH access to nodes, you can view real-time colored terminal logs directly in the console:
- Incremental Loading: Task output is refreshed near-real-time in the log panel via incremental polling;
- Colored Rendering: Terminal color codes are automatically rendered as highlighted text;
- Search & Filter: Keyword search with hit counting and highlighting helps quickly locate errors.
3. Result Parsing
The system supports automatic parsing of multiple result formats including JSON, JSONL, CSV, REGEX, XML, and LINE, converting raw scanner output into structured assets. Parsers are configured by tool DSL — no manual selection required.
Parsed assets and vulnerabilities are saved automatically: new subdomains, IPs, and web apps go to the asset library, while vulnerability findings generate vulnerability records and trigger notifications.
Related Documentation
- Task Lifecycle — the task state machine and task-workflow relationship
- Run Center — the unified monitoring entry for tasks and workflow runs