FAQ

Account & Login

Forgot Password?

Method 1: Contact admin to set a new password

The admin goes to "System Management" → "User Management", edits the user and sets a new password.

Method 2: Use deployment script to reset admin password

cd testnet-deploy
./testnet.sh reset-password
# Password after reset is Admin@123456

Account Disabled?

Contact the system admin to re-enable your account in "User Management".

---

Workflows & Tasks

Task remains in PENDING state after workflow execution

Possible Cause 1: No online scanning nodes

• Check the "Scanning Nodes" list to confirm there are 🟢 online nodes
• If there are no nodes, you need to deploy a scanning probe first

Possible Cause 2: The node's tool whitelist excludes the tool

• Open "Scanning Nodes" → "Configure" to review the node's tool whitelist (empty means unrestricted)
• Add the tool to the whitelist, or clear the whitelist

How to troubleshoot workflow execution failures?

1. Go to the "Run Center" and find the failed task
2. Click "Logs" to view detailed error information
3. Common errors:
   • Docker image not found: Tool Docker image not pulled, check node network
   • Input asset is empty: Input asset is empty, check if upstream nodes have output
   • Timeout: Execution timeout, increase timeout in workflow node configuration

How to retry a failed task?

Find the failed task in the "Task List" and click "Re-run" to execute it again.

How do I cancel a running task?

In the "Run Center", click "Cancel Task" in the task row actions. The server immediately notifies the probe over WebSocket to terminate the corresponding container or subprocess.

How do I restrict certain nodes to specific tools only?

Configure a node tool whitelist: open "Scanning Nodes" → "Configure" and pick the allowed tools under "Tool Execution Whitelist" (empty means unrestricted). Once saved, the node only pulls tasks for whitelisted tools — see Node Tool Whitelist.

---

Asset Management

Excel import failed?

1. Download the latest import template (list page "Import" → "Download Template")
2. Check if required fields are filled (fields marked with *)
3. Check if data format meets specifications (e.g., IP address format)
4. File size must not exceed 20MB

Assets blocked by rules?

Asset rules can configure validation rules. Assets that do not meet the conditions will be rejected (BLOCKED).

• View blocked records in the import results
• Go to "Asset Configuration" → "Asset Rules" to check currently active rules
• Modify rules or adjust data and re-import

Too many items selected during batch operation?

Click "Clear Selection" in the batch operation bar to clear current selection, or refresh the page to reset state.

---

Search Engine

API Key test failed after configuration

• Check if the Key is correct (no extra spaces)
• Check if the Key has expired or been banned
• Check if the server can access the corresponding engine's API (network connectivity)

Search results are empty

• Try simplifying the search syntax to rule out syntax errors
• Some engine free accounts have query limits, try reducing the result count
• Check if API quota has been exhausted

---

System & Permissions

Menu shows "No Permission"

Contact admin to check your role permission configuration:

1. Go to "Role Management"
2. Find your role
3. Confirm that the view permission for the corresponding module is enabled

WebSocket connection disconnected, notifications cannot be received in real-time

• Check if network connection is stable
• Refresh the page to re-establish WebSocket connection
• If using a reverse proxy, check if the proxy configuration supports WebSocket long connections

---

AI & MCP

Where do I get the MCP API Key?

Sign in and open "AI" → "MCP Access" to view and copy the API Key; for setup steps see Claude / Cursor Setup. In production, configure a dedicated key via the TESTNET_MCP_API_KEY environment variable.

---

Performance & Stability

System response is slow

• Check server CPU/memory usage
• Check if PostgreSQL connection count has reached the limit
• Temporary slowdown may occur during high concurrent task execution, wait a moment

Task logs cannot load

• Logs are pushed in real-time via WebSocket, check WebSocket connection status
• When tasks are completed, logs are loaded from the database, confirm the task was actually executed

---

Deployment & Environment Troubleshooting

Docker File Descriptor Limit Error (out of memory / ulimit)

Symptom:

When running high-concurrency scan tasks or starting Docker containers on Linux distributions (such as Kali Linux, Debian, or older CentOS releases), you encounter:

library initialization failed - unable to allocate file descriptor table - out of memory

Solution:

1. Edit the system's docker.service configuration file:

   sudo vim /usr/lib/systemd/system/docker.service

2. Append --default-ulimit nofile=65536:65536 to the ExecStart=/usr/bin/dockerd line:

   ExecStart=/usr/bin/dockerd --default-ulimit nofile=65536:65536

3. Reload systemd daemon and restart Docker:

   sudo systemctl daemon-reload
   sudo systemctl restart docker

DNS Resolution Failure Inside Docker Containers

Symptom:

Scan probe nodes report network errors when pulling dependencies or resolving target domains:

Temporary failure resolving 'mirrors.aliyun.com'

Solution:

1. Configure global DNS resolvers in Docker's daemon configuration:

   sudo mkdir -p /etc/docker
   sudo vim /etc/docker/daemon.json

2. Add reliable public DNS servers:

   {
     "dns": ["223.5.5.5", "114.114.114.114", "8.8.8.8"]
   }

3. Restart Docker service:

   sudo systemctl restart docker

Permission Denied When Mounting Docker Sock or Volumes

Symptom:

Probe nodes fail to launch scanner containers when mounting /var/run/docker.sock or local data volumes, throwing Permission denied.

Solution:

• Temporarily disable: Run sudo setenforce 0
• Permanently disable: Edit /etc/selinux/config and set SELINUX=disabled, then reboot the host.
• Volume Policy (Recommended): Append :z or :Z flags to the volume mount path (e.g., -v /var/run/docker.sock:/var/run/docker.sock:z).

Slow Dependency Downloads in Restricted Networks

Symptom:

Dependencies (e.g., Python pip packages or GitHub repository artifacts) time out during online probe setup.

Solution:

1. PyPI Mirror: Add domestic mirror flags to your Python dependency installation:

   pip3 install -r requirements.txt -i https://mirrors.aliyun.com/pypi/simple/

2. GitHub Acceleration: Prepend acceleration proxy URLs (e.g., https://gh-proxy.com/) in node tool configurations or environment variables.

Manual Debugging & Diagnostic Steps for Failed Scan Tasks

If a scan task fails on a probe node and UI logs are insufficient, follow these 5 diagnostic steps:

1. Extract Command: Go to the "Run Center", view task logs, and copy the exact executed command.
2. Access Probe Container: Access the probe client container on the host server:

   docker exec -it testnet-client /bin/bash

3. Execute Manually: Paste and execute the copied tool command directly inside the container to inspect raw stdout and stderr output.
4. Fix Dependencies:
   • For binary permission errors, execute chmod +x <tool>.
   • For missing Python / Go dependencies, run pip install or adjust environment variables.
5. Re-verify: Ensure the command exits with code 0 inside the container, then click "Re-run" in the web console.

---

Contact Support

If the above content cannot resolve your issue, please:

1. Check the Deployment Guide to confirm correct installation configuration
2. Collect server logs (docker logs testnet-server)
3. Contact platform admin or submit an Issue