Faulted: Broken Pipe

Ignition
1

I'm running Ignition 8.1.53, i received this message this morning '....Faulted: Broken pipe'.

<- incoming local='xxmasterxx' remote='xxbackupxx' method=onClose: Connection xxbackupxx|5bxx-xxxxx-xxxxx has been set to Faulted:Broken pipe

We are experiencing network timeouts quite often, our IT team states it's not the network.

In addition we see this in the logs:

WebSocketConnection 20Jul2026 06:22:55 -> outgoing local='xxbackupxx' remote='xxmasterxx' method=handleConnectException: Connection attempt to wss://xxx.xxx.xxx.005:8060/system/ws-control-servlet?name=hq-backup&uuid=5baxxxx-xxxxx-xxxxx&url=https://xxx.xxx.xxx.006:8060/system&**pingpong=true failed**! Verify that your host and port settings are correct. Error message=java.net.ConnectException: Connection refused: getsockopt

Log Properties
gan-remote-connection-id https://xxx.xxx.xxx.005:8060/system

2

This AI response seems alright.

This error indicates a network-level or service-level rejection between your backup gateway (xxbackupxx) and your master gateway (xxmasterxx).
Specifically, java.net.ConnectException: Connection refused: getsockopt means the backup machine reached target IP ending in .005 on port 8060, but nothing was actively listening on that port, or a firewall actively rejected the TCP SYN packet.

Step-by-Step Troubleshooting Guide

  1. Verify Ignition Service Status on the Master Node
    Target: xxx.xxx.xxx.005
    Check if the master node's Ignition service is actually running. If the Ignition service crashed or is restarting, port 8060 (the Gateway Network / Redundancy WebSocket port) will not respond.
  2. Verify Port 8060 Listening State
    On the Master Host Machine
    Confirm that Ignition is bound to port 8060. Open a terminal/command prompt on the master host (.005) and run:
    Windows: netstat -ano | findstr 8060
    Linux: sudo netstat -tulnp | grep 8060 or ss -tulnp | grep 8060
    If nothing is listening, check the Master Gateway logs (wrapper.log) to see why the Gateway Network service failed to initialize.
  3. Test Network Connectivity & Firewalls
    From Backup Host (.006) -
    Master Host (.005)">
    Test if TCP port 8060 is open between the backup and master machines using PowerShell or Telnet from the backup host:

If TcpTestSucceeded returns False:
Check local OS firewalls (firewalld, Windows Defender) on both machines.
Ensure any corporate firewalls or VLAN routing rules permit bidirectional TCP traffic on port 8060.
4. Check Gateway Network Approval Status
In Master Gateway Web Interface
Navigate to Config > Gateway Network > Incoming Connections on the Master Gateway.
Look for a pending connection request from hq-backup / xxbackupxx.
If present, click Approve.
If the status is stuck in Faulted or Denied, delete the connection entry and restart the Gateway Network connection on the backup side to re-trigger the handshake.

Common Edge Cases to Check

  • Duplicated Gateway UUID / Cloned VMs: If the backup gateway was created by cloning the VM or restoring a full backup without clearing unique identifiers, both gateways might share the exact same internal UUID. This causes immediate connection drops on port 8060.

    Fix: Stop the Backup Gateway, delete <Ignition_Install_Dir>/data/.uuid and <Ignition_Install_Dir>/webserver/metro-keystore, then restart the service.

  • SSL Requirement Mismatch: Redundancy on port 8060 requires SSL (wss://). Ensure Require SSL settings match across both nodes under Redundancy Settings.