DevNet 2

As we get ready for testnet launch we are introducing a new node architecture.

DevNet 2 is basically alpha testnet. Once this has proven stable for 40 days, testnet will be released

Pipe PoP Cache Node Documentation

Overview

Cache Node is a high-performance caching service that helps distribute and serve files efficiently across a network.

System Requirements

  • Linux

  • Minimum 4GB RAM (configurable), more the better for higher rewards

  • At least 100GB free disk space (configurable). 200-500GB is a sweet spot

  • Internet connectivity available 24/7

  • v0.2.6 specifically introduces support for ports 80 and 443, requiring elevated privileges.

  • A small note:

    “Port 80 and 443 require root privileges or a systemd capability fix (see ‘Systemd Service’ section). If you run the binary manually, you should use sudo or run as root.”

Basic Installation

Install

# download the compiled pop binary
curl -L -o pop "https://dl.pipecdn.app/v0.2.8/pop"
# assign executable permission to pop binary
chmod +x pop
# create folder to be used for download cache
mkdir download_cache

⚠️ IMPORTANT: to avoid formatting issues, compose your single-line curl command in a plain text editor (notepad), and then paste the single-line command into your server cli

Quick Start

sudo ./pop

Runs on port 8003, 443 and 80 with 4GB RAM and 100GB disk space.

Egress data will be on port 8003, 443 and 80. Make sure 8003, 443 and 80 are open on any firewalls/NAT's such as UFW.

If you choose not to use systemd, you must run Pipe PoP with sudo or as root to bind to ports 80 and 443

Configuration Example

sudo ./pop \
  --ram 8 \              # RAM in GB
  --max-disk 500 \       # Max disk usage in GB  
  --cache-dir /data \    # Cache location
  --pubKey <KEY>         # Solana public key

Retrieve the public key from your Solana wallet (e.g., Phantom, Backpack) and paste it here for --pubKey

Monitor

# View metrics
./pop --status

# Check points (note: Points are not active yet)
./pop --points-route

Refer / Signup by referral

# Generate referral
./pop --gen-referral-route

# Use referral
sudo ./pop --signup-by-referral-route <CODE>

Files

  • node_info.json: Node configuration

  • download_cache/: Cached content

Recommened to backup node_info.json. It is linked to the IP address that registered the PoP node. It is no recoverable if lost.

Logs

  • Output streams to stdout

  • Auto-reports errors

  • Health updates every 5 minutes

The node tracks uptime, bandwidth, cache hits, and historical metrics.

Referral System

How Referrals Work

  • Nodes can generate referral codes: ./pop --gen-referral-route

  • New nodes can sign up with referral: ./pop --signup-by-referral-route <CODE>

  • Referrer earns 10 points when referred node:

    • Stays active for 7+ days

    • Maintains reputation score > 0.5

  • The node that generated the referral needs to also maintain a good reputation score for the referrals to be counted as valid and productively adding value to the network.

  • This program will expand over time to include rewards sharing.

Checking Referral Status

  1. Browse to https://dashboard.pipenetwork.com/node-lookup

  2. Enter the referrer's node ID in the Node Lookup

  3. Scroll down to "Referral Stats" to review referred nodes

Reputation System

The node's reputation score (0-1) is calculated based on the last 7 days of node operation, using three main components:

1. Uptime Score (40% of total)

  • Reports are first grouped by hour to prevent over-weighting from frequent reporting

  • A day is considered to have "good coverage" if it has at least 75% of hours reported (18+ hours)

  • For days with good coverage, the average uptime is weighted by how complete the day's coverage was

  • The final uptime score is the weighted average daily uptime divided by seconds in a day (capped at 100%)

2. Historical Score (30% of total)

  • Based on how many days out of the last 7 had good coverage

  • Example: If 6 out of 7 days had good coverage, the historical score would be 0.857 (86%)

  • This rewards consistent reporting over time

3. Egress Score (30% of total)

  • Based on total data transferred over the 7-day period

  • Normalized against a target of 1TB per day

  • Capped at 100%

Example Calculation

If a node has:

Day 1: 24 hours reported (100% coverage)
Day 2: 22 hours reported (92% coverage)
Day 3: 23 hours reported (96% coverage)
Day 4: 24 hours reported (100% coverage)
Day 5: 20 hours reported (83% coverage)
Day 6: 12 hours reported (50% coverage - not counted)
Day 7: 24 hours reported (100% coverage)

Then:

  • 6 days have good coverage (>75% of hours)

  • Historical score would be 6/7 = 0.857

  • Uptime score would be based on the weighted average of those 6 days

  • The day with only 50% coverage is not counted in the uptime calculation

Important Notes

  1. Maintenance Windows:

    • Short gaps (< 4 hours) don't significantly impact the score

    • A day needs only 75% coverage to count, allowing for maintenance

    • Restarts don't reset your progress

  2. Score Recovery:

    • Scores are calculated over a rolling 7-day window

    • Poor performance drops out of the calculation after 7 days

    • New nodes can build up their score within a week of good operation

  3. Best Practices:

    • Regular reporting (at least hourly)

    • Plan maintenance during the same 6-hour window each day

    • Keep total downtime under 6 hours per day when possible

  4. Score Interpretation:

    • 90%+ : Excellent reliability

    • 80-90%: Good reliability

    • 70-80%: Fair reliability

    • <70%: Needs improvement

Benefits of High Reputation

  • Priority for P2P transfers (score > 0.7)

  • Eligibility for referral rewards (score > 0.5)

  • Future: Higher earnings potential

View Reputation

./pop --status

Displays detailed breakdown of reputation metrics and overall score.

FAQ

File Size Limits & Performance

  • Default RAM usage: 4GB

  • Default disk cache: 100GB

  • Chunk size: 64MB per transfer

  • Max concurrent downloads controlled by RAM allocation

Network Requirements

  • Stable internet connection required

  • Ports 8003, 443 and 80 must be accessible

  • Supports IPv4 and IPv6

  • Auto-fallback on network errors

Monitoring & Diagnostics

  • Auto-reports errors to central servers

  • Logs in stdout

  • Health checks every 5 minutes

  • Built-in metrics for CPU, RAM, disk, network usage

Automatic Updates

  • Checks for updates on startup

  • Auto-notifies when new version available

  • Shows download URL for latest version

Cache Behavior

  • Files expire after 4 hours if unused

  • Auto-cleanup of expired files

  • Automatic peer discovery

  • Geographic-based peer selection

Security Notes

  • IP-based rate limiting

  • Node ID verification

  • Geographic distribution tracking

  • IP address validation on registration

Systemd Service

Here is an example systemd config setup submitted by the community

Please customize based on your preferences 

# Create service user and directories
sudo useradd -r -m -s /sbin/nologin pop-svc-user -d /home/pop-svc-user 2>/dev/null || true

# Create required directories
sudo mkdir -p /opt/pop
sudo mkdir -p /var/lib/pop
sudo mkdir -p /var/cache/pop/download_cache

# Move pop binary from home directory to installation directory
sudo mv -f ~/pop /opt/pop/
sudo chmod +x /opt/pop/pop 2>/dev/null || true

# Handle any existing node_info.json from quick-start
sudo mv -f ~/node_info.json /var/lib/pop/ 2>/dev/null || true

# Set ownership
sudo chown -R pop-svc-user:pop-svc-user /var/lib/pop
sudo chown -R pop-svc-user:pop-svc-user /var/cache/pop
sudo chown -R pop-svc-user:pop-svc-user /opt/pop

# Create systemd service file
sudo tee /etc/systemd/system/pop.service << 'EOF'
[Unit]
Description=Pipe POP Node Service
After=network.target
Wants=network-online.target

[Service]
AmbientCapabilities=CAP_NET_BIND_SERVICE
CapabilityBoundingSet=CAP_NET_BIND_SERVICE
User=pop-svc-user
Group=pop-svc-user
ExecStart=/opt/pop/pop \
    --ram=12 \
    --pubKey 7ugorwsqWvT6fwHoZhejAUPGWxK4fJtE9W8f8vebVh2S \
    --max-disk 175 \
    --cache-dir /var/cache/pop/download_cache \
    --no-prompt
Restart=always
RestartSec=5
LimitNOFILE=65536
LimitNPROC=4096
StandardOutput=journal
StandardError=journal
SyslogIdentifier=pop-node
WorkingDirectory=/var/lib/pop

[Install]
WantedBy=multi-user.target
EOF

# Set config file symlink and pop alias: prevent dup configs/registrations, convenient pop commands.
ln -sf /var/lib/pop/node_info.json ~/node_info.json
grep -q "alias pop='cd /var/lib/pop && /opt/pop/pop'" ~/.bashrc || echo "alias pop='cd /var/lib/pop && /opt/pop/pop'" >> ~/.bashrc && source ~/.bashrc

# Reload systemd, check and enable service
sudo systemctl daemon-reload
sudo systemd-analyze verify pop.service && sudo systemctl enable pop.service && sudo systemctl start pop.service

echo "Example pop.service setup is complete"
echo "1. Update values in pop.service based on your preferences"
echo "2. Use the 'pop' command (the alias) to run any pop commands"
echo "3. Check service status with: sudo systemctl status pop"

⚠️ IMPORTANT: When running pop commands while using the systemd service configuration, you must either:

  • Change to the working directory first: cd /var/lib/pop && /opt/pop/pop [command]

  • OR use the provided alias: pop [command] (recommended) This ensures node_info.json is found and prevents duplicate node registration.

Backup your node

Take a backup copy of the node_info.json and store it offline. Example of taking a backup copy:

If running as a service: cp /var/lib/pop/node_info.json ~/node_info.backup2-4-25

If running quickstart: cp ~/node_info.json ~/node_info.backup2-4-25

You can download the backup file off your machine using SCP or other means.

Upgrade your Node

The nodes logs our --status output will show the URL of a newly released version

  1. Download new version: curl -L -o pop "<DOWNLOAD-URL>"

  2. chmod +x ./pop

  3. Move pop binary to permanent location as needed, for example mv ./pop /opt/pop/pop

  4. Change directory 'cd' into the WorkingDirectory containing the node_info.json, for example cd /var/lib/pop

  5. pop --refresh or /opt/pop/pop --refresh

  6. You may need to add these to the [SERVICE] section of your .service if you are using systemd [Service] AmbientCapabilities=CAP_NET_BIND_SERVICE CapabilityBoundingSet=CAP_NET_BIND_SERVICE

    1. If you prefer not to modify your systemd unit, you can also run: sudo setcap 'cap_net_bind_service=+ep' /path/to/pop This lets a non-root user bind to ports 80/443.”

  7. Open necessary ports:

    sudo ufw allow 8003/tcp
sudo ufw allow 80/tcp
sudo ufw allow 443/tcp
sudo ufw reload

Decommission old DevNet 1 Node

If using same computer, this is best done before activating DevNet 2 Node

sudo systemctl disable dcdnd.service
sudo systemctl stop dcdnd.service
PreviousPerformance and fraud detectionNextTroubleshooting

Last updated