Check CRC detected errors first
A flaky cable is the most common cause of ring oscillation. Check CRC detected-error counters on ring ports.
11.4 MRP Troubleshooting
The previous chapter covered MRP timing parameters and tuning for faster recovery. When the ring does not behave as expected, systematic troubleshooting identifies the root cause. MRP issues fall into a small number of categories.
Why a Systematic Approach Matters
Section titled “Why a Systematic Approach Matters”MRP ring issues share similar symptoms: devices become unreachable, the MRM reports the ring as open, or topology changes appear in the log. Without a structured approach, engineers chase symptoms instead of causes. The 4 most common root causes are: wrong VLAN, wrong Domain UUID, 2 MRMs, or a flaky cable.
Symptom: Ring is Open (No Physical Inoperability)
Section titled “Symptom: Ring is Open (No Physical Inoperability)”The MRM reports the ring as open even though cables are connected.
| Cause | Diagnosis | Fix |
|---|---|---|
| MRP not enabled on switches | Check MRP status on each switch | Enable MRP on ring switches |
| Wrong Domain UUID | Compare UUID across switches | Set the same UUID on switches |
| MRP VLAN mismatch | Check MRP VLAN on each ring port | Set the same MRP VLAN on ring ports |
| Ring port outside MRP VLAN | Check VLAN membership of ring ports | Add ring ports to MRP VLAN |
| Disabled ring port | Check port status | Enable the port |
Symptom: Ring Oscillates
Section titled “Symptom: Ring Oscillates”The ring continuously changes state. Topology change events appear in the log every few seconds.
Symptom: Failover Takes Too Long
Section titled “Symptom: Failover Takes Too Long”The ring recovers, but recovery takes longer than 200 ms.
| Cause | Fix |
|---|---|
| MRP_LinkDown disabled | Enable MRP_LinkDown |
| Test interval too long | Reduce test interval to 10 ms |
| Switch CPU overloaded | Reduce other traffic or upgrade switch |
Symptom: Traffic Lost After Failover
Section titled “Symptom: Traffic Lost After Failover”The ring recovers (MRM shows ring closed) but some devices are unreachable.
| Cause | Fix |
|---|---|
| Stale ARP entries | Clear ARP cache on affected devices |
| MAC table not flushed | Verify MRP_TopologyChange propagation |
| VLAN missing from new path | Add missing VLAN to ring ports |
Diagnostic Tools on Hirschmann HiOS
Section titled “Diagnostic Tools on Hirschmann HiOS”| Tool | Location | Use |
|---|---|---|
| MRP Status | Switching → L2-Redundancy → MRP | Ring state, port roles, domain |
| Event Log | Diagnostics → System Log | Topology change events |
| Port Statistics | Basic Settings → Port Statistics | CRC detected errors, link up/down counts |
| Port Mirroring | Diagnostics → Port Mirroring | Capture MRP frames with Wireshark |
To capture MRP frames with Wireshark, configure port mirroring on the switch. Mirror a ring port to a monitor port. Apply the Wireshark filter mrp or eth.type == 0x88e3. If no MRP_Test frames appear, then MRP is not running on the MRM or the ring port is outside the correct VLAN.
Key Takeaways
Section titled “Key Takeaways”4 causes cover most issues
Wrong VLAN, wrong Domain UUID, 2 MRMs, or a flaky cable. Check these 4 causes first.
What Comes Next
Section titled “What Comes Next”MRP runs on Hirschmann switches. The next chapter introduces the Hirschmann product families, the HiOS operating system, and configuration best practices. The next chapter delivers the practical knowledge needed to deploy and manage these switches in production.
References
Section titled “References”- IEC 62439-2:2016 — Media Redundancy Protocol
- Hirschmann. (2023). User Manual: HiOS MRP Diagnostics. Belden/Hirschmann.
- Wireshark Foundation. (2024). Wireshark User’s Guide. https://www.wireshark.org/docs/wsug_html/