[Application Note]                                  Parallel Redundancy Protocol for Easergy P3
relays with dual port 100Mbps Ethernet interface
Parallel Redundancy Protocol for Easergy P3Ux and
Easergy P3x3x relays with dual-port 100 Mbps Ethernet
interface
1 Introduction
An industrial, real-time Ethernet communication network must typically offer a much better
level of availability and uninterrupted operation than normal, office-type Ethernet networks.
For power automation systems, even a short loss of connectivity may result in a significant
reduction of functionality or impaired safety. To recover from a network failure, various
redundancy schemes have been considered, including Rapid Spanning Tree Protocol
(RSTP), High-availability Seamless Redundancy protocol (HSR) and Parallel Redundancy
Protocol (PRP).
PRP uses two independent networks in parallel and does not change the active topology
of the networks. Each message is transmitted by the sender node parallelly over both
networks. In the receiving node, the first received message on one of two networks
is processed while its later duplicate is discarded. Such details like duplicated message
transmission on two networks and messages reception with duplicate detection and
rejection are managed by the low-level PRP layer of the communication stack architecture,
with the two physical networks being hidden from the higher layers. Consequently, unlike
RSTP-based redundant systems, the PRP-based communication networks are continuously
available with no recovery delay in case of a single failure occurrence.
The PRP protocol implemented in Easergy P3 relays is specified in the IEC62439-3 (Clause
4) standard and is available when a dual-port, 100 Mbps Ethernet interface card is used.
1.1 PRP features
•       Ethernet redundancy method independent of any industrial Ethernet protocol
        or topology (tree, ring or mesh)
•       Seamless switchover and recovery in case of failure (no delay)
•       Continuous supervision of redundancy for better management of network devices
•       Suitable for hot swap - 24 hour/365 day operation in substations
•       Allows mixing of devices with single and dual network interfaces on the same local
        area network (LAN)
•       Allows HMI devices (laptops, workstations) to be connected to the network using
        standard Ethernet adapters
•       Particularly suited for hard real-time systems such as substation automation, high-
        speed drives and transportation
    1                                                                              P3/EN ANCOM/A04
Parallel Redundancy Protocol for Easergy P3 relays                     [Application Note]
with dual port 100Mbps Ethernet interface
2 PRP network
A PRP network is a pair of independently powered local area networks (LANs) with any
topologies (tree, ring or mesh) that may introduce different transmission delays between
nodes from other networks. The structure is presented in Figure 1.
       SAN                   SAN
                                                                                SAN
                                                      RedBox
                                                                             RedBox
                   SAN
                                               LAN_B
       SAN
                                                               LAN_A
                   DAN
                                                     DAN                  SAN
Figure 1 PRP Network Structure
A PRP network can have various nodes such as:
       Single attached nodes (SANs) that represent devices with a single Ethernet interface.
        SANs are able to communicate with devices connected to the same LAN and they
        are responsible for non-critical system functions.
       Double attached nodes (DANs) represent devices with double Ethernet interfaces
        and PRP protocol support. DANs are able to communicate with all devices from both
        LANs. These are normally critical devices whose connection needs to be redundant.
       A redundancy box (RedBox) is used to create a virtual DAN connected to both LANs
        from devices with a single Ethernet interface. The RedBox allows SANs
        to communicate with devices from both LANs because it has its own unique IP
        address and acts as a gateway to the PRP network.
P3/EN ANCOM/A004                                                                         2
 [Application Note]                                     Parallel Redundancy Protocol for Easergy P3
                                                    relays with dual port 100Mbps Ethernet interface
3 PRP operation
PRP assumes that all devices demanding high a availability of their communication are
equipped with dual-port Ethernet interfaces and connected to two separate and
independent local area networks, LAN_A as primary and LAN_B as redundant. Each
message is then transmitted by the sender node on both LAN_A (as “A”-frame) and on
LAN_B (as “B”-frame). The receiving node captures eventually both frames but only the
first received is further processed while the second received is recognized as a duplicate
and rejected from further processing. This principle of operation allows for seamless fault
tolerance of any single network link failure.
                             „B“-
                            frame           „A“-
                                           frame
Figure 2 PRP Redundancy network
The basic Ethernet frame to be sent to the network is composed of the following fields:
         preamble   dest_addr   src_addr     type      LSDU                   FCS
        dest_addr - destination MAC address
        src_addr - source MAC address
        type - Ethertype
        LSDU - Link Service Data Unit (data)
        FCS - Frame Check Sequence (checksum characters for error detection and
         correction)
 3                                                                                     P3/EN AN/A004
  Parallel Redundancy Protocol for Easergy P3 relays                        [Application Note]
  with dual port 100Mbps Ethernet interface
 PRP extends the basic Ethernet frame of a so-called redundancy control trailer to the
 following structure transmitted on both networks:
preamble      dest_addr    src_addr     type    LSDU     Seq     LAN id    Size     PRP suffix       FCS
          Seq - 16-bit sequence number
          LAN id - 4-bit LAN Identifier: 1010 (0xA) for LAN_A and 1011 (0xB) for LAN_B
          Size - 12-bit LSDU size
          PRP suffix - 16-bit PRP suffix (0x88 0xFB)
 3.1 Duplicate detection and discarding algorithm
 It is assumed that each node sends PRP frames with increasing sequence numbers, but it
 is possible to receive frames out of sequence if layer 2 prioritization is applied in the network.
 The source MAC address and current sequence number are used to keep track of the
 frames that have been sent and that have already been received. Thus, it allows to detect
 and discard incoming duplicates for a configurable time limit.
 For every node on both networks, the algorithm also provides:
          Fast detection of frame reception from a wrong LAN, which indicates physical network
           connection problems.
          Message counts to show how many messages have been received.
          Error counts to show whether frames were not received or were sent/received on the
           wrong LAN.
  P3/EN ANCOM/A004                                                                               4
[Application Note]                                          Parallel Redundancy Protocol for Easergy P3
                                                        relays with dual port 100Mbps Ethernet interface
4 Hardware options
Two hardware options are provided to implement a dual-port Ethernet interface:
        Interface card with two RJ-45 ports
        Interface card with two fiber-optic LC ports
     Easergy P3 Advanced (P3x3x)                  Easergy P3 Standard (P3Ux)
4.1 Port assignment to networks
        ETH1 Port is assigned to PRP LAN_A
        ETH2 Port is assigned to PRP LAN_B
 5                                                                                         P3/EN AN/A004
 Parallel Redundancy Protocol for Easergy P3 relays                   [Application Note]
 with dual port 100Mbps Ethernet interface
5 Technical details
Product details related to the operation in PRP networks:
      Easergy P3x3x and Easergy P3UX devices offer hardware configuration with two
       redundant Ethernet ports and PRP support.
      Two hardware versions, with RJ ports and with LC optical fibre ports, are available.
      Easergy P3 device is able to work as a double attached node (DAN) in PRP networks.
      Configuration of the PRP is available in the Easergy Pro setting and configuration
       tool (later called Easergy Pro) and transparent for any Ethernet-based protocols.
      Loss of a single LAN connection causes no losses in application data transmitted
       over Ethernet.
      Easergy P3 device supports cyclic transmission of PRP supervision frame with a
       configurable interval (LifeCheckInterval), default 5 s (+/- 10 ms).
      There is a single MAC address for the two redundant physical interfaces.
      There is a single IP address used by both physical interfaces.
      The performance of Easergy P3 Ethernet interface is not limited by using PRP.
      PRP operation in an Easergy P3 device does not require any limitation in the number
       of connected DANs in PRP networks.
      The default value of the duplicate discard timeout parameter (duration that the
       received message Sequence number will be held to detect and discard a duplicate
       message) is 400 ms. Because of the buffering capacity, it is recommend to configure
       a value not greater than 500 ms.
 P3/EN ANCOM/A004                                                                      6
[Application Note]                                    Parallel Redundancy Protocol for Easergy P3
                                                  relays with dual port 100Mbps Ethernet interface
6 PRP configuration with Easergy Pro
6.1 Redundancy protocol selection
In Communication > Protocol configuration in Easergy Pro, PRP can be selected as the
redundancy protocol for Ethernet.
There are three redundancy protocol options:
        None
        RSTP
        PRP
PRP is available only for selected Easergy P3 types (3xx series and Easergy P3UX) when a
dual-port Ethernet interface card is used.
6.2 PRP operation control and monitoring
After configuring PRP as the redundancy protocol for Ethernet, the Protocol configuration
view in Easergy Pro shows the list of configuration parameters of the PRP and the list of
diagnostic parameters of the PRP performance for each of the two networks.
6.3 PRP configuration parameters
Table 1 PRP configuration parameters
 Parameter       Possible values       Defaults                Description
 Enable for      True / False          False                   Enabling PRP protocol
 PRP
 Duplicate       400 [ms]              2-65535 [ms]            Duration that the received
 Discard                                                       message sequence number
 Timeout                                                       is kept, used by the duplicate
                                                               discard algorithm to detect
                                                               PRP duplicates
 Enable Superv True / False            False                   Enable cyclic transmission of
 for PRP                                                       the supervision frame
 Supervision     5 [s]                 2-65535 [s]             Interval of sending the
 duration                                                      supervision frame
 7                                                                                   P3/EN AN/A004
Parallel Redundancy Protocol for Easergy P3 relays                       [Application Note]
with dual port 100Mbps Ethernet interface
 Supervision       01:15:4E:00:01:00        01:15:4E:00:01:00   Multicast address 01-15-4E-
 Destination                                -                   00-01-XX reserved for this
 Address                                                        protocol.
                                            01:15:4E:00:01:FF
                                                                By default XX is “00”, but if
                                                                conflicts arise, XX can be
                                                                configured to take any value
                                                                between 0x00 and 0xFF.
6.4 Diagnostic parameters of PRP performance
Table 2 Diagnostic parameters of PRP performance
 Parameter                Description                                         Data Type
 Tx Sequence              Sequence number incremented by 1 for each sent      Unsigned16
 Number                   PRP frame
 Buffered Duplicates      Number of frames that are received and currently    Unsigned32
                          processed by the duplicate discard algorithm
 Buffer Hash Conflict     Total number of hash conflicts in the duplicate     Unsigned32
                          discard algorithm that affects the detection of
                          duplicate frames
 Frames from other        Number of frames that were received with the        Unsigned32
 LAN                      wrong LAN identifier on LAN_A and LAN_B
 Missing Duplicates       Number of PRP frames that were received only        Unsigned32
                          with single LAN, without duplicate on the second
                          LAN
 Non PRP frame            Number of frames that were received without         Unsigned32
 received                 PRP suffix on LAN_A and LAN_B
 Current State            The current state of the PRP network interface    NoLink,
                                                                            ConnOK,
                                                                            WrongLAN
 PRP frame received       Number of frames received over LAN_A and Unsigned32
                          LAN_B
 PRP frame sent           Number of frames sent over LAN_A and LAN_B Unsigned32
 Ethernet packet          Number of frames sent to upper layers after being Unsigned32
 received                 received over the ETH1 and ETH2 ports and not
                          discarded as duplicates
P3/EN ANCOM/A004                                                                            8
 [Application Note]                                   Parallel Redundancy Protocol for Easergy P3
                                                  relays with dual port 100Mbps Ethernet interface
6.4.1 Interpretation of diagnostic parameters
Table 3 Interpretation of diagnostic parameters
 Parameter       Value        Interpretation
 Tx Sequence     >0           Current frame sequence number might be used for PRP
 Number                       traffic analysis i.e. in Wireshark in case of frame loss.
 Buffered        >10          Increasing number of stored and processed frames might
 Duplicates                   indicate communication problems i.e. broken redundant
                              network connection that cause duplicate loss.
 Buffer Hash     >0           The number of hash conflicts reflects the total amount of
 Conflict                     valid PRP duplicates that were not handled by the duplicate
                              discard algorithm and were sent directly to the upper
                              protocol layer.
 Frames from     >0           Increasing number of frames that were received with the
 other LAN                    wrong LAN identifier might indicate wrong cable connection
                              or switching loops in network topology.
 Missing         >0           Increasing number of missing duplicates might indicate
 Duplicates                   broken network path or invalid cable connection.
 Non PRP         >0           Increasing number of non PRP frames that were received
 frame                        indicates the presence of SAN connected directly to PRP
 received                     network without RedBox.
 Current State   NoLink       There is no physical connection with LAN network
                 ConnOk       There is a valid connection with LAN network and PRP LAN
                              identifiers of received frames belong to ports LAN_A and
                              LAN_B.
                 WrongLAN     There is valid physical LAN connection, but the received
                              PRP frames do not belong to the port’s LAN identifier. It
                              might be caused by connecting device port assigned to
                              LAN_A with redundancy network from LAN_B and vice
                              versa. This state also indicates switching loops in the
                              network topology.
 PRP frame       >0           Number of sent and received PRP frames might be used to
 received                     detect packet loss between DAN in PRP network caused by
 PRP frame       >0           invalid connection, network reconfiguration or firewall
 sent                         filtering.
 Ethernet        >0           Number of frames that were filtered by the duplicate
 packet                       discarding algorithm and sent to upper protocol layers.
 received
 9                                                                                   P3/EN AN/A004
Parallel Redundancy Protocol for Easergy P3 relays                                        [Application Note]
with dual port 100Mbps Ethernet interface
Schneider Electric
35 rue Joseph Monier
92500 Rueil-Malmaison
                                                     © 2017 Schneider Electric Industries SAS - All rights reserved.
FRANCE
                                                     Electrical equipment should be installed, operated, serviced,
Phone: +33 (0) 1 41 29 70 00                         and maintained only by qualified personnel. No responsibility is
Fax: +33 (0) 1 41 29 71 00                           assumed by Schneider Electric for any consequences arising
                                                     out of the use of this material.
easergy.schneider-electric.com
Publishing: 10/2017
P3/EN ANCOM/A004                                                                                                        10