Media Redundancy Protocol

Media Redundancy Protocol

The Media Redundancy Protocol (MRP) is a network protocol that:

  • Ensures fast convergence in ring network topologies for industrial automation,

  • Operates under the IEC standard 62439-2, and

  • Supports recovery times of 30 ms, 200 ms, and 500 ms.

List of Cisco Catalyst IE9300 Rugged Series Switches that support MRP:

  • IE-9310-26S2C-E and IE-9310-26S2C-A

  • IE-9320-26S2C-E and IE-9320-26S2C-A

  • IE-9320-22S2C4X-E and IE-9320-22S2C4X-A

  • IE-9320-24T4X-E and IE-9320-24T4X-A

  • IE-9320-24P4X-E and IE-9320-24P4X-A

  • IE-9320-16P8U4X-E and IE-9320-16P8U4X-A

  • IE-9320-24P4S-E and IE-9320-24P4S-A

  • IE-9310-16P8S4X-E and IE-9310-16P8S4X-A

Table 1. Feature history table

Feature

Release information

Feature description

Enhanced ring convergence profiles for standalone IE9300 switches

Release 17.18.1

From Cisco IOS-XE 17.18.1 onwards, 12 rings of 30ms is also supported on a standalone IE9300 switch.

PTP over MRP

Release 17.18.1

From Cisco IOS-XE 17.18.1 onwards, PTP support over Media Redundancy Protocol (MRP) is introduced.

MRP on an IE9300 Stack

Release 17.17.1

From Cisco IOS-XE 17.17.1 onwards, stack support for Media Redundancy Protocol (MRP) is introduced.

Media Redundancy Protocol (MRP)

Release 17.13.1

From Cisco IOS-XE 17.13.1 onwards, 12 rings of 200ms or 500ms convergence profiles are supported on a standalone IE9300 switch.

MRP modes

MRP modes are operating configurations that define how the switch interacts with the MRP network. These modes have distinct management interfaces and compatibility with external frameworks.

The switch supports two MRP modes, with only one mode being enabled at a time.

  • PROFINET MRP mode: Designed for integration with Siemens' Totally Integrated Automation (TIA) framework. This mode is primarily managed through the TIA framework and does not utilize the CLI or WebUI for MRP configuration. This mode is the default when the MRP is activated via the web interface or command line.

  • MRP CLI mode: Managed using Cisco IOS XE CLI or WebUI. This mode is incompatible and does not support downloading MRP configurations from Siemens TIA.

Roles and states of MRP

The roles and states of the MRP ring are a set of operational configurations and behaviours in Cisco Catalyst IE9300 Rugged Series Switches that define how MRP manages network redundancy in a ring topology. These roles and states are distinguished by their functionality in preventing loops, managing failures, and ensuring network recovery.

Roles:

  • MRA (Media Redundancy Automanager): Default role for all nodes, uses a voting protocol to select one MRM based on priority, while the remaining nodes transition to the MRC role.

  • MRM (Media Redundancy Manager): Manages the ring by monitoring its status, blocking ports to prevent loops, and coordinating recovery during failures.

  • MRC (Media Redundancy Client): Operates under the MRM's control, forwarding or blocking traffic as instructed.

States of MRA ring ports:

  • Disabled : Ports drop all received frames.

  • Blocked : Ports drop all frames except MRP control frames and some standard frames like LLDP.

  • Forwarding : Ports forward all received frames.

  • Not Connected: Ports are physically disconnected or down (distinct from Disabled , which is manually configured).

Network operation in normal (Ring-Closed) state:

  • During normal operation, the network operates in a Ring-Closed state.

  • One of the MRM’s ring ports is blocked, while the other forwards traffic to prevent loops.

  • MRCs typically have both ports in the Forwarding state, resulting in a logical stub topology.

Network behavior during failures (Ring-Open state):

  • When a failure occurs, the network transitions to a Ring-Open state.

    For example: If a link between two MRCs fails, both MRM ports switch to Forwarding , while MRCs adjacent to the failure have one port Disabled and the other Forwarding .

MRP control frames:

  • Test Frames : Sent by the MRM to monitor ring status.

  • TopoChange Frames : Sent by the MRM upon failure or recovery detection.

  • LinkChange Frames : Sent by MRCs to the MRM during local port failure or recovery (Linkdown or Linkup events).

Recovery performance

MRP ensures efficient switchover during failures to avoid loss of Layer 2 Ethernet frames. This loop avoidance results in the physical ring topology becoming a logical stub topology. Recovery time profiles (for example: 200 ms) dictate the maximum recovery time during topology convergence.

The figure displays the two rings: the Left and Right rings.

  • Left Ring: The connection (small blue square, top) on the MRM is in a blocked state (as shown by the two parallel lines) because no ports are disconnected.

  • Right Ring: Two MRC connections (left and center small white squares) are in the disabled state because the link between them is broken, as marked by a red “x”.

Figure 1. MRP Ring States

MRP-STP interoperation

MRP-STP interoperation is a network feature that enables MRP and Spanning Tree Protocol (STP) to work together to maintain network stability and prevent broadcast loops. This interoperation is designed to ensure seamless integration between MRP and STP without requiring additional configuration.

The interoperability between MRP and STP is managed in MRP CLI mode. No additional CLI configuration is required for MRP and STP to interoperate.

Loop prevention and protocol behaviour is achieved as follows:

  • MRP works with STP to prevent broadcast loops caused by devices that do not participate in the MRP ring but are accidentally connected to the network.

  • STP Bridge Protocol Data Units (BPDUs) are not sent on MRP-enabled ports. When ports are unconfigured from the MRP ring, they are automatically added back to the STP network.

Guidelines and limitations

General guidelines and limitations

This section outlines MRP support, compatibility, port and ring configurations, convergence profiles, scalability, interface behavior, and access port setup.

  • MRP Support and Compatibility:

    • Starting Cisco Catalyst IE9300 Rugged Series Switches release 17.13.1, MRP is supported.

    • Support for multiple MRP rings is available only through the CLI or WebUI.

    • You cannot configure MRP on the port running these features: Resilient Ethernet Protocol (REP), Spanning Tree Protocol (STP), Flex Links, MACsec, Dot1x, or TrustSec.

  • Ring and port configurations:

    • Each IE9300 node must include two interfaces as members of an MRP ring. Partial rings or ring segments are not supported.

    • MRP ports cannot be configured as SPAN destination ports, Private VLAN ports, Tunnel ports, or on EtherChannels (or individual ports in EtherChannels).

    • Each MRP ring can have one Admin VLAN (by default) and can be overridden by a configurable CLI. The Admin VLAN must be unique across rings in the same device to prevent traffic flooding. VLANs can only include interfaces from a single MRP ring.

  • Convergence profiles and scalability:

    • The switch supports up to 12 rings with standard convergence profiles of 30 ms, 200 ms, or 500 ms in a standalone IE9300 switch.

    • Each ring can support up to 50 Media Redundancy Clients (MRCs).

  • MRP interface behavior:

    • MRP interfaces start in a forwarding state and remain so until notified it is safe to block. The ring state transitions to Ring-Closed during normal operation.

  • Access port configuration:

    • For MRP interfaces configured as access ports, use the switchport mode access command and switchport access vlan x command to assign them to the appropriate VLAN.

Guidelines and limitations for MRP CLI mode

This section focuses on key configuration requirements, such as ensuring interface mode consistency, handling VLAN mismatches, reconfiguring rings, and supported convergence profiles.

  • After using the CLI to configure the MRP ring, you must attach it to a pair of ports that support MRP.

  • Both MRP ports must have the same interface mode, either access or trunk.

  • To change the MRP ring configuration or to switch the interface mode of the ring ports between access and trunk, you must delete the existing ring configuration and recreate it with the desired settings.

  • Access VLAN matching:

    • When both MRP ports are in access mode, their access VLANs must match.

    • If the configured MRP VLAN does not match the ports' access VLAN, the MRP VLAN is automatically updated to match the access VLAN of the MRP ports.

  • VLAN mismatch error handling:

    • If the ports in an MRP ring do not belong to the same access VLAN during ring creation or if the VLAN of one port is changed after the ring is created, the MRP ring operation is suspended.

    • A system error message is displayed:
      ERROR% The ring 1 ports don't belong to the same access VLAN. The MRP ring will not function until the issue has been fixed.

    • To resolve this, ensure that both ring ports have the same access VLAN.

Guidelines and limitations to MRA

Media Redundancy Automanager (MRA) is an administrative role in Cisco Catalyst IE9300 Rugged Series Switches that facilitate the selection of the Media Redundancy Manager (MRM) through a voting protocol during the startup phase of a Media Redundancy Protocol (MRP) ring. This role is temporary and transitions to either the MRM or Media Redundancy Client (MRC) role after the voting process.

MRA can be activated through the CLI. For configuration details, see the section MRP CLI configuration parameters in this configuration guide.

  • MRA is not an operational role. It is used specifically at device startup to facilitate the MRM selection process.

  • All nodes begin as MRAs at power-up. After the MRM is selected, the remaining MRAs automatically transition to the MRC role.

  • Manual configuration of an MRM and MRA in the same ring is not supported.

  • MRAs use a voting protocol based on a configured priority value to select the MRM. It sends MRP_Test frames on both ring ports.

    Manual configuration of an MRC role is not allowed. It is automatically determined after the voting process.

  • MRAs compare their own priority with the priority values in the received MRP_Test frames:

    • If the MRA's priority is higher, it sends an MRP_TestMgrNAck (negative acknowledgment) frame with the remote manager's MAC address.

    • If the MRA receives an MRP_TestMgrNAck containing its own MAC address, it transitions to the MRC role.

  • Once the MRM is selected, the remaining MRAs transition to the MRC role. MRP_TestPropagate frames are sent to inform other MRAs in the client role about the role change and the new higher-priority MRM.

    These frames ensure that clients monitor the higher-priority manager and remain in the client role if the manager's role changes.

Guidelines and limitations to PROFINET MRP mode configuration

Before configuring the Cisco switch with PROFINET MRP through Siemens TIA or STEP7, ensure the following:

  • PROFINET MRP feature doesn't support MRC role.

  • Use the TIA portal to configure or modify MRA role.

  • Avoid using CLI for configuration when TIA is in use. MRP CLI mode and PROFINET MRP mode are mutually exclusive.

  • If the switch is connected to the PROFINET PLC, the output of show profinet status | include Connected should display Yes . If it displays No , the switch is not connected to the PROFINET PLC.

  • Ensure that the GSD file version matches the Cisco IOS release to avoid compatibility issues. For detailed configuration steps, refer to the PROFINET Protocol Configuration Guide.

MRP on an IE9300 stack

From Cisco IOS-XE 17.17.1 onwards, stack support for the Media Redundancy Protocol (MRP) is introduced as a means to ensure system stability and seamless integration across stacked switches.

The functionality and behavior of MRP remain consistent in both standalone and stacked configurations, where the MRP stack setup supports up to 4-member stacking. PROFINET MRP mode does not support stack configuration.

For more information on Switch Stacks, see Managing Switch Stacks.

Benefits of MRP on an IE9300 stack

Deploy MRP in a stacked setup to introduce node level redundancy within the network. Here are the benefits:

  • Supports up to 16 rings with 200 ms and 500 ms convergence profiles in IE9300 stack setups.

    The 30ms convergence profile is not supported in IE9300 stack setups.

  • Reduces the risk of single points of failure by ensuring the system stays operational, even if a stack member or the active switch fails.

  • Provides seamless integration and reliability across the network.

  • Greater port density, which is crucial for accommodating more devices without compromising performance.

PTP over Media Redundancy Protocol

From IOS XE 17.18.1 onwards, Precision Time Protocol (PTP) is supported over Media Redundancy Protocol (MRP).

PTP is a time synchronization protocol for nodes distributed across a network. It is now supported over MRP according to IEC standard 62439-2 to achieve fast convergence in ring network topology for Industrial Automation networks.

PTP over MRP is supported on:

  • IE9300-Manager, having 12 rings, with each ring consisting of 20 nodes spanning over a length of 20kms.

  • Network using QoS (COS 6 / DSCP EF 46).

  • Copper and SFP ports.

  • Time error deviation of less than 10 microseconds at 50% link utilization between the head end and the host in case of an MRP ring break.

The PTP over MRP supported configurations are:

Device

PTP clock

PTP profile

MRP mode

MRP convergence

IE9300

Boundary/Grand Master Boundary (GMC-BC)

Default (IEEE 1588 v2)

Manager

  • 200ms

  • 500ms

Configure PROFINET MRP

This task guides you through configuring PROFINET MRP to ensure proper network operation and redundancy.

Before you begin

Disconnect an MRP Ethernet port from the ring (open ring) to discover all neighboring devices using the LLDP protocol. Perform this step before deploying PROFINET MRP to the network. This approach prevents unnecessary flooding if configuration issues occur.

  • Use the show lldp neighbor command to confirm all neighbor devices are correctly discovered before continuing with PROFINET MRP setup.

  • Check that the PROFINET status indicates a connected-state .

Procedure

Step 1

Access the PROFINET Device Discovery (DCP) window.

Open the PROFINET DCP window to identify and manage devices in the network.

Figure 2. PROFINET Device Discovery (DCP) window before configuring MRP

Step 2

Assign PROFINET MRP manager role and domain name on the device.

Figure 3. PROFINET MRP manager role and MRP domain name

Step 3

Define the PROFINET MRP client and MRP domain name on client devices.

Figure 4. PROFINET MRP and MRP domain on client

Step 4

When using MRA mode, configure all devices and domain details.

Step 5

Configure the PROFINET MRP interfaces on all devices participating in the ring.

Figure 5. PROFINET MRP inerfaces

Step 6

Compile the configuration and then download it to the PLC device.

Step 7

Verify that all devices are connected and the MRP ring is closed.

Figure 6. PROFINET MRP network configuration diagram

Prerequisites to MRP CLI mode configuration

To prevent network storms when configuring or unconfiguring the MRP feature in a physical ring topology, follow these steps:

  • Ensure that one physical connection between two nodes in each ring remains open.

  • Issue the shut command on the connecting interfaces or physically disconnect the cable.

  • After completing the configuration of all MRMs, issue the no shut command on the port or reconnect the cable between the nodes.

  • MRP CLI is disabled by default.

    Default mode: PROFINET MRP

    Default VLAN: VLAN 1

  • If using a non-default VLAN for MRP, create the VLAN before assigning it to MRP Ring 1 .

MRP CLI configuration parameters

The MRP CLI configuration parameters are a category for configuring the MRP on a network device. These optional parameters enable redundancy and resiliency in network communication.

Optional configuration parameters

  • domain-id:

    A unique identifier represents the MRP ring, ensures the correct recognition and communication within the ring.

  • domain-name:

    A logical name assigned to the MRP domain-ID for easier management and identification.

  • profile:

    Define the timing parameter, with the default set to 200 milliseconds, which controls how quickly the ring responds to topology changes.

  • vlan-id:

    The VLAN used for transmitting MRP frames, separating MRP traffic from other network data.

Key components of MRP CLI configuration

  • Node setup:

    You must configure the device as an MRA (Media Redundancy Auto Manager) node and associate with two MRP ports.

  • Ring count limits:

    • Up to 16 rings can be configured in a stack setup.

    • Up to 12 rings can be configured on a standalone node.

Configure the Switch as MRA in MRP CLI mode

To configure a switch as an MRA in MRP CLI mode, ensure proper ring and interface configurations for redundancy, as this is the default mode.

Before you begin

  • Ensure the switch is connected and accessible via CLI.

  • If the device is connected to a PLC module, select the option no device in the ring for MRP.

Procedure

Step 1

Enter the global configuration mode and enable MRP on the switch.

Example:

Switch#configure terminal

Configure the MRP ring with the specified ring ID using the mrp ring <mrp_id> command.

Example:

Switch(config-mrp)#mrp ring 1

Step 2

Set the switch to auto-manage mode.

Use the mode auto-manager command.

Example:

Switch(config-mrp)#mode auto-manager

Step 3

(Optional) Configure the recovery profile.

Use the profile <value> command.

Choose a value for maximum recovery time: 30, 200, or 500 ms. Default: 200 ms.

Example:

Switch(config-mrp-auto-manager)#profile 30

Step 4

(Optional) Configure the MRA priority

Use the priority <value> command.

Set a priority within 36864–61440 , default is 40960 .

Example:

Switch(config-mrp-auto-manager)#priority 40955

Step 5

(Optional) Configure the VLAN ID:

Use the vlan-id vlan command. Use within 1–4094 , default is 1 .

Example:

Switch(config-mrp-auto-manager)#vlan-id 2025

Step 6

(Optional for single MRP ring) Configure the domain name.

Use the domain-name <name> command. Provide a name that can be up to 32 characters long.

Example:

Switch(config-mrp-auto-manager)#domain-name Cisco MRP Ring 1

Step 7

(Optional for single MRP ring) Configure the domain ID.

Use the domain-id <value> command.

where:

<value> is a UUID string of 32 hexadecimal digits in five groups separated by hyphens.

domain-id for MRP ring 1 is FFFFFFFF-FFFF-FFFF-FFFF-FFFFFFFFFFFE (by default).

Example:

Switch(config-mrp-auto-manager)#domain-id FFFFFFFF-FFFF-FFFF-FFFF-FFFFFFFFFFFE

Assign a unique domain ID to each ring:

  • The default domain ID for ring 1 is FFFFFFFF-FFFF-FFFF-FFFF-FFFFFFFFFFFE.

  • The default domain ID for ring 2 is FFFFFFFF-FFFF-FFFF-FFFF-FFFFFFFFFFFD.

The domain ID as FFFFFFFF-FFFF-FFFF-FFFF-FFFFFFFFFFFF should be avoided.

Step 8

(Optional) Set the Interval.

Use the interval <interval> command.

Specify the interval options in milliseconds:

  • 3 : Default interval for 30 ms recovery profile.

  • 20 : Default interval for 200 ms recovery profile.

  • 50 : Default interval for 500 ms recovery profile.

Note

 

The interval field is not displayed in WebUI for MRP.

Example:

Switch(config-mrp-auto-manager)#interval 3

Step 9

Exit global configuration mode.

Example:

Switch(config-mrp-auto-manager)#end

Configure the ring ports and modes for MRP

To configure the ring ports and modes for MRP on network switches.

Procedure

Step 1

Enter the global configuration mode and configure the interface mode.

Example:

Switch#configure terminal

Specify the interface where you want to configure the ring ports and modes.

Example:

Switch(config)#interface GigabitEthernet1/0/22

Step 2

Configure the interface mode.

Use the switchport mode { access | trunk } command.

Example:

For default VLAN ID 1 : 

Switch(config-if)#switchport mode access

For other VLAN IDs:

Switch(config-if)#switchport mode access VLAN <ID>

Step 3

Associate the interface to the MRP ring.

Use the mrp ring 1 command.

Example:

Switch(config-if)#mrp ring 1

Step 4

Return to the Global Configuration mode.

Use the exit command.

Example:

Switch(config-if)#end

Step 5

(For multiple rings) Repeat steps 1 to 4 for each additional ring.

Configuration examples

The following example shows configuring MRP auto-manager:

Configuration examples for MRP CLI mode

The following example shows configuring auto-manager for MRP CLI mode:

Switch#configure terminal
Switch# no profinet mrp
Enter configuration commands, one per line. End with CNTL/Z.
Switch(config)#mrp ring 1
Switch(config-mrp)#mode auto-manager
Switch(config-mrp-auto-manager)#domain-id FFFFFFFF-FFFF-FFFF-FFFF-FFFFFFFFFFFE
Switch(config-mrp-auto-manager)#priority 40960
Switch(config-mrp-auto-manager)#end

The following example shows configuring interface mode for MRP CLI mode:


Switch#configure terminal
Enter configuration commands, one per line. End with CNTL/Z.
Switch(config)#interface GigabitEthernet1/0/22
Switch(config-if)#switchport mode trunk
Switch(config-if)#mrp ring 1
WARNING% Enabling MRP automatically set STP FORWARDING. It is recommended to shutdown all interfaces which are not currently in use to prevent potential bridging loops.
Switch(config-if)#exit

Verification examples

A verification procedure is a systematic process that

  • confirms whether a configuration has been applied as intended,

  • detects discrepancies between the expected and actual system state, and

  • ensures that operational objectives are met after changes or deployments.

Verify the MRP CLI mode configuration:

Switch#show  mrp ring  1

MRP ring 1

Profile         : 30 ms
Mode            : Auto-Manager
Priority        : 40955
Operational Mode: Manager
From            : CLI
License         : Not Applicable
Gateway         :
Status         : Disabled
Best Manager    :
MAC Address    : 70:DA:48:3B:B3:95
Priority       : 40955

Network Topology: Ring
Network Status  : CLOSED
Port1:                                     Port2:
MAC Address    :70:DA:48:3B:B3:96          MAC Address    :70:DA:48:3B:B3:95
Interface      :Tw1/0/22                   Interface      :Tw1/0/21
Status         :Forwarding                 Status         :Blocked

VLAN ID     : 2025
Domain Name : Cisco MRP Ring 1
Domain ID   : FFFFFFFF-FFFF-FFFF-FFFF-FFFFFFFFFFFE

Topology Change Request Interval        : 0ms
Topology Change Repeat Count            : 3
Short Test Frame Interval               : 1ms
Default Test Frame Interval             : 3ms
Operational Test Frame Interval         : 3ms
Test Monitoring Interval Count          : 3
Test Monitoring Extended Interval Count : N/A

Displays the summary of the MRP CLI mode configuration:

Switch#show  platform mrp summary

MRP ring Summary

Total number of rings configured    : 1
Total number of rings Closed        : 1
Total number of rings Open          : 0
Total number of rings Idle          : 0
Total number of rings Undefined     : 0
Ring instance 1

status       : CLOSED
port 1       : Tw1/0/21, SECONDARY
port 2       : Tw1/0/22, PRIMARY

Verify the PROFINET MRP mode configuration:

Switch#show profinet mrp ring 1
MRP ring 1

Mode            : Client
From            : Profinet
License         : Not Applicable
Gateway         :
 Status         : Disabled
Best Manager    :
 MAC Address    : 28:63:36:A9:F8:08
 Priority       : 40960

Network Topology: Ring
Network Status  : CLOSED
PNPORT 0:(126/53540)                            PNPORT 0:(126/53540)
 MAC Address    :70:DA:48:E4:86:01          MAC Address    :70:DA:48:E4:86:10
 Interface      :Gi1/0/1                    Interface      :Gi1/0/16
 Status         :Forwarding                 Status         :Forwarding

VLAN ID     : 1
Domain Name : mrpdomain-1
Domain ID   : C3D687FE789E3A1ACDBE5BFCBBC27B6
Link Down Timer Interval        : 20 ms
Link Up Timer Interval          : 20 ms
Link Change (Up or Down) count  :  4