Deployment Center

The Deployment Center is your hub for provisioning new agents, managing updates, and maintaining your fleet. Access it from the Deployment tab in the navigation bar.

If no agents are connected yet, the dashboard will show a shortcut card directing you to this page.

Fleet Overview

At the top, three metric cards give you a quick snapshot of your fleet:

Metric	Description
Total Agents	Number of registered agents across all systems
Online Now	Currently connected agents
Outdated	Agents running an older version than what’s staged on the Hub

---

Official Installers

Direct download links for agent installers:

• Windows Agent: Download the MSI installer from the latest GitHub release. Supports Windows 10/11 (x86_64). See Windows Agent for setup details.
• Linux Agent: Use the script-based installation in the Deployment AIO section below. See Linux Agent for manual setup.

---

Agent Downloads (Hub Staging)

Before you can deploy or update Linux agents remotely, you need to stage an agent binary on your Pankha Hub (your server). This avoids agents downloading directly from GitHub, keeping everything on your local network.

How to Stage a Version

1. Select a version from the dropdown (grouped into Stable and Unstable releases).
2. Click the Stable or Unstable button to download that version to your Hub.
3. Wait for the download to complete. The status will change to “vX.X.X Ready”.

---
title: Hub-and-Spoke Binary Distribution
---
graph LR
    GitHub([GitHub Releases]) -->|Stage Once| Hub[Pankha Hub]
    Hub -->|Fast LAN| Agent1([Agent 1])
    Hub -->|Fast LAN| Agent2([Agent 2])
    Hub -->|Fast LAN| Agent3([Agent 3])

Why stage? Instead of every agent downloading from the internet, you download once to your server. Agents then pull the binary over your fast local network. This also works in air-gapped environments with no internet access on agent machines.

The Clear Cache button removes all cached binaries from the Hub.

---

Deployment AIO (All-In-One)

This section generates a ready-to-run install command for deploying Linux agents. Configure your settings visually, then copy the command and run it on the target machine.

Step 1: Installation Mode

Mode	Install Path	Log Path
Standard (/opt/)	/opt/pankha-agent/	/var/log/pankha-agent/
Portable (CWD)	Current working directory (e.g., ~/pankha/)	Same directory as the agent binary

Both modes are functionally identical. They install the same systemd service and support the same features. The only difference is where files are stored.

Step 2: Connection Settings

Choose how the agent will connect to your Hub:

• Internal: Uses your server’s LAN IP (e.g., http://192.168.1.100:3143). Best for agents on the same network.
• External: Uses the URL you’re currently browsing from. Best for remote agents outside your LAN.

You can also manually edit the Hub URL if needed.

Step 3: Agent Configuration

Configure the agent’s behavior before deployment:

Setting	Options	Description
Log Level	DEBUG, INFO, WARN, ERROR	Verbosity of agent logs
Emergency Temp	70-85 C	All fans go to 100% if any sensor reaches this temperature
Failsafe Speed	30-100%	Fan speed when the agent loses connection to the Hub
Agent Rate	1-10 seconds	How often the agent reads sensors and reports data
Fan Step	1-15%	Maximum fan speed change per update cycle (smooth transitions)
Hysteresis	1-5 C	Minimum temperature change before the fan adjusts

See Advanced Settings for detailed explanations of each setting.

Step 4: Copy and Run

Two terminal blocks display your install command in both curl and wget format. Click Copy and run it on the target Linux machine:

# Example (your command will have a unique token)
curl -sSL "http://192.168.1.100:3143/api/deploy/linux?token=abc123" | bash

The command downloads the agent binary from your Hub (not the internet), creates the configuration, installs the systemd service, and starts the agent. It will appear on your dashboard within seconds.

Token expiry: Each generated command contains a unique token that expires after 24 hours. Generate a new command if the token has expired.

---

Fleet Maintenance

This section shows all connected agents in a table and lets you manage updates.

Agent Table

Column	Description
Hostname	Agent name with platform icon (Windows or Linux)
Agent ID	Unique agent identifier
Version	Current agent version (with update badge if outdated)
Status	Online, Offline, or Updating
Maintenance	Update action (platform-specific)

Updating Linux Agents

1. Stage a version in the Agent Downloads section (see above).
2. Open Fleet Maintenance and look for agents with the “NEW vX.X.X” badge.
3. Click Update Now. The agent downloads the new binary from the Hub, performs an atomic swap, and restarts automatically.
4. The agent reconnects within a few seconds and shows the new version.

---
title: Remote Agent Update Flow
---
sequenceDiagram
    participant User as Dashboard
    participant Hub as Pankha Hub
    participant Agent as Linux Agent

    User->>Hub: Click "Update Now"
    Hub->>Agent: selfUpdate command (WebSocket)
    Agent->>Hub: Download binary (LAN)
    Agent->>Agent: Verify SHA256 checksum
    Agent->>Agent: Atomic swap (old → .old, new → current)
    Agent->>Agent: Restart service
    Agent->>Hub: Reconnect with new version
    Hub->>User: Agent shows updated version

Downgrade: If the staged Hub version is older than the agent’s current version, the button will show “Downgrade” instead.

Updating Windows Agents

Windows agents cannot be updated remotely. The table provides a Download MSI link to the latest release. Download and install it manually on the Windows machine.

---

Next Steps

• Advanced Settings: Fine-tune agent behavior after deployment.
• Fan Profiles & Logic: Set up fan curves for your newly deployed agents.
• Troubleshooting: If an agent doesn’t appear after deployment.