Facebook
Google
GitHub
Linkedin
Hello Control Systems Experts,
I am seeking urgent assistance with a critical communication issue on a DEH (Digital Electro-Hydraulic) Controller running an aging Westinghouse WDPF system on a Sun Microsystems Ultra 10 workstation.
System Background & Problem Description
I am seeking urgent assistance with a critical communication issue on a DEH (Digital Electro-Hydraulic) Controller running an aging Westinghouse WDPF system on a Sun Microsystems Ultra 10 workstation.
System Background & Problem Description
- System Hardware: Sun Microsystems Ultra 10.
- System OS: Solaris (Version is likely older, e.g., Solaris 2.6/5.6).
- Role: DEH Controller, identified as drop210.
- Primary Issue: The system continuously displays the following critical communication error every few seconds: drop210 SHC_HWY: WDPF Highway went OFFLINE, Highway ID = 1
-
Chronology of Events and Troubleshooting Steps Taken
The issue began immediately after we attempted to connect an external SCSI tape drive via a newly installed PCI SCSI card for backup purposes. Although the SCSI card has since been removed, the system instability and communication error persist.
1. Initial Hardware Conflict & Boot Errors- SCSI Card: Installed and subsequently removed the added PCI SCSI card.
- Severe Boot Error: During the first reboot attempt after the change, the system displayed severe PCI errors in the OpenBoot PROM (ok prompt), indicating a deep hardware conflict or damage: pci108c,5000: PBM generated system error. pci10: PBM generated Target Abort.
-
2. Current Status & Actions Taken (Still Failing)
Despite removing the conflicting hardware, the following steps were performed, and the WDPF error persists:- System Reboots: The system was restarted multiple times.
- Cable Integrity Check: We have confirmed the physical cable connection for the WDPF Highway is secure and appears intact.
- Basic Network Status (Layer 3): The communication link with the PLC device (and possibly other network nodes like drop200) is established and confirmed. The PLC is visible on the network.
- PCI Configuration Check: We verified the hardware tree using prtconf and identified the WDPF card as a separate PCI device (likely a vendor ID related to the previous Target Abort error).
- Crux of the problem: The physical and IP layers seem fine (PLC is visible), but the proprietary WDPF application layer (SHC_HWY) is unstable and constantly declaring the link offline.
Question
What is the most likely cause for the persistent WDPF Highway OFF-LINE error after resolving the initial PCI conflict?- Is this a persistent resource conflict (IRQ/DMA) that requires a specific Solaris boot command (other than boot -r)?
- Is there a WDPF/SHC command-line utility in Solaris to force a protocol stack reset for the Highway (similar to ifconfig down/up)?
- Does the original Target Abort error imply permanent damage to the WDPF PCI card itself, even if the lower-level network appears active?
- Thank you in advance for any insights on this critical legacy control system.
