OPC_UA TAsk takes up 100% system load on Ip change

[hammarbytp]

We have a vxworks 1.7.x OPC_UA server

When we change the IP address of one of the connected ports, one OPC_UA task takes 100% of the available system load

This is the log output

** urn:192.168.1.100/GE/P80iServer: start trace
** Product version: GE Power Conversion OPC UA Server for HPCi (V8.0.0) 1.7.7.549
** C++ SDK Version: 1.7.7.549
** C Stack Version: Version:1.4.22 327
** Platform Layer: PlatformName:vxWorks
** Date: 2024-08-28
**
15:37:14.239Z|2|AE4C6918* [uastack] OpcUa_Socket_HandleEvent: OpcUa_Socket_HandleAcceptEvent failed 0x80050000!
15:37:14.239Z|2|AE4C6918* [uastack] OpcUa_Socket_HandleAcceptEvent: accepting failed with 22
15:37:14.239Z|2|AE4C6918* [uastack] OpcUa_Socket_HandleEvent: OpcUa_Socket_HandleAcceptEvent failed 0x80050000!
15:37:14.239Z|2|AE4C6918* [uastack] OpcUa_Socket_HandleAcceptEvent: accepting failed with 22
15:37:14.239Z|2|AE4C6918* [uastack] OpcUa_Socket_HandleEvent: OpcUa_Socket_HandleAcceptEvent failed 0x80050000!
15:37:14.239Z|2|AE4C6918* [uastack] OpcUa_Socket_HandleAcceptEvent: accepting failed with 22
15:37:14.239Z|2|AE4C6918* [uastack] OpcUa_Socket_HandleEvent: OpcUa_Socket_HandleAcceptEvent failed 0x80050000!
15:37:14.239Z|2|AE4C6918* [uastack] OpcUa_Socket_HandleAcceptEvent: accepting failed with 22

It looks like it is in a tight loop after the change in IP and causing system overload

[Support Team]

Hi,

this indeed looks like an endless loop, and because the select is driving the whole communication loop, it is natural that it goes into 100% load.

Your best option is to not change the IP address.

A second guess would be to ask the manufacturer "GE Power Conversion" on how to change IP in their system (without breaking it). You should contact the GE help desk, but check user manual first.