Ringlet 0, also called the outer ring, is for clockwise or west data traffic. Ringlet 1, also called the inner ring, is for counterclockwise or east data traffic.

On Ringlet 0, stations send data frames out of east ports and receive data frames from west ports. On Ringlet 1, stations send data frames out of west ports and receive data frames from east ports.

Any two adjacent stations are connected by a pair of unidirectional logical channels called links transmitting in opposite directions. These two links form a span. Multiple continuous spans and their stations form a domain. A span on which data frames are not allowed to pass is called an edge. If a ring contains at least one detected edge, it is called open ring. If it does not contain any detected edges, it is called closed ring.

Data operations on RPR

Stations on an RPR handle data frames by performing the following four types of operations:

· Insert, to place a frame on a ringlet.

· Transit, to pass a frame to the next station.

· Copy, to deliver an inbound frame from the ring to the upper layer. Copying a frame does not remove the frame from the ring.

· Strip, to remove a frame from a ringlet. The frame is not passed to the next station.

By performing these operations, stations implement unicast, broadcast, multicast, and unknown unicast transmission.

Unicast transmission

Figure 2 Unicast transmission on an RPR ring

Figure 2 shows how a unicast data frame is transmitted on an RPR ringlet:

1. The source station inserts the unicast frame into the data stream on Ringlet 0 or Ringlet 1.

2. Transit stations transit the frame.

3. The frame is copied and stripped when it reaches the destination station or when its time to live (TTL) expires.

Different from traditional ring technologies where unicast frames are removed from the ring at the source station, RPR adopts destination stripping to remove unicast frames from the ring at the destination station. This increases bandwidth utilization and spatial bandwidth reuse efficiency.

Broadcast, multicast, and unknown unicast transmission

Figure 3 Broadcast, multicast, and unknown unicast transmission on an RPR ring

Figure 3 shows how a broadcast, multicast, or unknown unicast frame is transmitted on an RPR ringlet:

1. The source station inserts the frame into the data stream on Ringlet 0 or Ringlet 1.

2. Transit and destination stations copy and transit the frame if the TTL of the frame has not expired.

3. The frame is stripped when it returns to the source station or when its TTL expires.

Topology discovery

RPR performs automatic topology discovery to collect such information as the number of stations, ring state, order of the stations on the ring to build a topology database. This database does not change after the ring topology becomes stable.

Topology database

Each RPR station maintains a topology database that describes the topology of the entire RPR ring. Based on this database, the station creates its ringlet selection table.

A topology database contains three categories of information:

· Ringlet topology information, such as the number of stations, ring state, and available bandwidth.

· Local station topology information, such as the MAC address, protection mode, protection state, station name, topology checksum of the local station and the neighbor stations.

· Topology information of other stations, such as the MAC address, validity state, reachability, protection mode, station ID, station name, and reserved bandwidth.

Topology discovery process

During the topology discovery process, RPR uses the following three types of control frames to transmit topology data: topology protection (TP), attribute discovery (ATD), and topology checksum (TC).

· TP frames convey configuration and status information of stations. The stations on an RPR ring broadcast TP frames to advertise their configuration and status information and update their own topology databases after receiving TP frames from other stations. Finally, all the stations reach an agreement on the topology of the ring.

· ATD frames convey attributes of the local station such as the MAC address and station name. These attributes will be included in the topology database.

· TC frames convey topology checksum information. They are sent between adjacent stations to check whether the topology databases on them are synchronized, thus identifying stability of the RPR ring topology.

All these control frames are sent at regular intervals, which are user configurable. For TP and TC frames, two types of interval, fast sending interval and slow sending interval, are used.

· When a station on the ring starts initializing or detects a topology change, it sends TP frames to propagate topology information throughout the network. When doing that, it sends the first nine TP frames at fast intervals and subsequent TP frames at slow intervals.

· After the RPR ring topology converges, the station starts to send TC frames. When doing that, it sends first five TC frames at fast intervals and then subsequent TC frames at slow intervals.

· Only one type of interval is available for sending ATD frames, regardless of topology change.

Fault response methods

RPR delivers strong self-recovery capability. Its protection mechanism provides event detection, quick self-recovery, and fast service recovery when faults occur to the fiber-optic or stations. The network can thus detect faults quickly and react appropriately to restore services in 50 ms.

RPR is available with two fault response methods: passthrough and protection.

Passthrough

At present, the switch does not support the passthrough approach.

The passthrough approach mainly applies to station faults handling. When a station detects an internal fault, it enters the passthrough state in which it behaves like a repeater and does not handle any services on the ring. Frames arriving at this station are forwarded directly as if it was transparent, and the station is invisible on the ring. Station B shown in Figure 4 is an example.

Figure 4 Passthrough approach

Protection

If a station is unable to forward traffic as the result of power failure or fiber cut for example, it should enter protection mode. RPR provides the following protection modes:

· Wrapping—After a span or station fails, protected traffic is directed at the point of failure to the opposing ringlet. The two ringlets (Ringlet 0 and Ringlet 1) then form a closed single ring around the point of the failure. The wrapping mode minimizes the data frame loss, because the wrapping mode allows quick switchover, However, this mode wastes bandwidth.

· Steering—The two stations at the two sides of a point of failure update their topology databases, and then send TP frames at fast intervals to the other stations on the ring. After a new topology database is stabilized, the source station directs protected frames to the ringlet that retains connectivity to their destinations. The steering mode can avoid the bandwidth waste. However, it can cause frame loss and service interruption because it requires topology reconvergence.

Figure 5 Schematic diagram before and after protection switchover

As shown in Figure 5, traffic travels from station D to station B along Ringlet 0. The transmission path is station D—station E—station A—station B. After the span between station A and station E fails, a protection switchover occurs.

· In wrapping mode, traffic that should originally travel from station E to A along Ringlet 0 is directed to Ringlet 1 to reach station A. The new transmission path is station D—station E—station D—station C—station B—station A—station B.

· In steering mode, traffic that should have traveled from station D to station B along Ringlet 0 is steered to Ringlet 1 for transmission. The new transmission path is Station D—Station C—Station B.

RPR protection switching uses the following protection hierarchy listed in the order of decreasing severity:

· Forced switch (FS)

· Signal fail (SF), related to current physical status.

· Signal degrade (SD), related to current physical status.

· Manual switch (MS)

· Wait to restore (WTR)

· Idle

Protection switching occurs only when requested by a station on the ring. The hierarchy of protection requests is the same as this protection switching hierarchy. Among protection requests, FS and MS requests are sent manually, whereas SF, SD, and WTR requests are sent automatically. If multiple protection requests appear at the same time, the higher priority protection request is processed preferentially. For example:

· The MS request sent by a station will not be processed if a higher priority protection request is present on the RPR ring.

· If an FS request is present on the current ring, the SF or SD request automatically sent as the result of a link failure will be processed only after the FS request is cleared.

Centralized RPR and distributed RPR

An RPR station is identical to an RPR logical interface bound with two RPR physical ports. The two RPR physical ports are used to transmit data to or receive data from the ring. The RPR logical interface is provided for you to make configuration.

Definitions

· If the two RPR physical ports are located on the same subcard, RPR is centralized.

· If the two RPR physical ports are located on different subcards, RPR is distributed.

Differences

The differences between distributed RPR and centralized RPR lie in that:

· In centralized RPR, when the subcard where the two RPR physical ports are located goes down, the RPR station fails.

· In distributed RPR, the down event of one subcard can cause the RPR physical port on the card to go down and edge to the RPR station. The RPR station can still work in protection mode to restore services in 50 ms.

· In distributed RPR, if two RPR physical ports are on the same RPR station, you must connect their mate ports with an optical fiber first for the RPR station to forward traffic.

RPR interface

RPR logical interface and RPR physical ports

As described earlier, an RPR station comprises a pair of RPR physical ports, one is called the “east port” and the other is called the “west port.” Initially, the two ports were fixed on the switch. The RPR station could become unusable for the impairment of either port. To address the issue, the RPR logical interface was introduced.

An RPR logical interface is identical to an RPR station. It can be bound with any two RPR physical ports. When either of the RPR physical ports fails, you can replace it with another one. Moreover, as the configuration on the RPR logical interface can be synchronized to the RPR physical ports, the configuration procedure is streamlined.

Available RPR logical interface types include RPR interface (Layer 3 RPR logical interface) and RPR-Bridge interface (Layer 2 RPR logical interface) and available RPR physical port types include RPRXGE and RPRPOS. Each RPR physical port is accompanied by a mate port.

For a complete RPR station to work, do the following:

1. Create an RPR logical interface.

2. Bind the RPR physical ports with the RPR logical interface.

· For an RPR station to work, you must bind its RPR logical interface with at least one RPR physical port. With only one RPR physical port, the station is working in protection mode.

· For the RPR station to forward traffic, before binding two RPR physical ports with the same RPR logical interface, you must connect their mate ports.

· You can bind any type of RPR physical ports to an RPR logical interface. However, the RPR physical ports bound to an RPR logical RPR interface must be of the same type.

RPR master/slave port

The first RPR physical port bound to the RPR logical interface is the master port and the second RPR physical port bound to the RPR logical interface is the slave port.

All calculation concerning the operation of RPR is done on the master port and the calculation result is synchronized to the slave port for consistency. Examples of RPR-related calculation include execution of configuration commands, protection event processing, RPR frames sending for topology discovery.

When the master port fails or leaves the binding with the RPR logical interface, the slave port takes over as the master port.

Protocols and standards

The RPR implementation follows the following document:

IEEE802.17: Resilient packet ring (RPR) access method and physical layer specifications

RPR configuration task list

Complete the following tasks to configure RPR:

Task		Remarks
Configuring basic RPR functions	Configuring an RPR physical port	Optional
	Creating and configuring an RPR logical interface	Required
	Binding an RPR logical interface with RPR physical ports	Required
	Enabling RPR mate port smart connection	Optional
	Changing the RPR physical port type	Optional
	Configuring the RPR station name	Optional
Configuring RPR protection	Configuring the RPR protection mode	Optional
	Configuring RPR protection reversion mode	Optional
	Manually sending a protection request	Optional
Configuring the ringlet selection table	Adding a static ringlet selection entry	Optional
Configuring the ringlet selection table	Configuring default ringlet selection	Optional
Configure the RPR fairness algorithm	Configuring reserved bandwidth or rate limiting	Optional
Configure the RPR fairness algorithm	Configuring station weight	Optional
Configuring RPR timers	Configuring the ATD timer	Optional
	Configuring the hold off timer	Optional
	Configuring the keepalive timer	Optional
	Configuring the topology stability timer	Optional
	Configuring the TC timers	Optional
	Configuring the TP timers	Optional
	Configuring the WTR timer	Optional
Testing connectivity between RPR stations		Optional

Configuring basic RPR functions

Configuring an RPR physical port

NOTE:

RPR physical ports include RPRXGE interfaces and RPRPOS interfaces.

To configure an RPRXGE interface:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPRXGE interface view.	interface rprxge interface-number	N/A
3. Configure the description for the interface.	description text	Optional. By default, the description of an interface is interface-name Interface.
4. Shut down the interface.	shutdown	Optional. Enabled by default.
5. Restore the default settings of the interface.	default	Optional.

To configure an RPRPOS interface:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPRPOS interface view.	interface rprpos interface-number	N/A
3. Configure the description for the interface.	description text	Optional. By default, the description of an interface is interface-name Interface.
4. Set the overhead bytes of the interface.	flag { c2 \| { j0 \| j1 } { sdh \| sonet } } flag-value	Optional. By default, the C2 byte value of an RPRPOS interface is 0x16, and the SDH J0 and J1 byte values are null.
5. Set the framing format on the interface.	frame-format { sdh \| sonet }	Optional. By default, the framing format is SDH.
6. Set the SD or SF alarm threshold for the interface.	threshold { sd \| sf } value	Optional.
7. Shut down the interface.	shutdown	Optional. Enabled by default.
8. Restore the default settings of the interface.	default	Optional.

Creating and configuring an RPR logical interface

Creating and configuring a Layer 2 RPR logical interface

To create and configure a Layer 2 RPR logical interface:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter Layer 2 RPR logical interface view.	interface rpr-bridge interface-number	N/A
3. Configure the description for the interface.	description text	Optional. By default, the description of an interface is interface-name interface.
4. Shut down the RPR logical interface.	shutdown	Optional. Enabled by default.
5. Restore the default settings of the interface.	default	Optional.

Creating and configuring a Layer 3 RPR logical interface

To create and configure a Layer 3 RPR logical interface:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter Layer 3 RPR logical interface view.	interface rpr interface-number	N/A
3. Configure the description for the interface.	description text	Optional. By default, the description of an RPR interface is interface-name Interface.
4. Configure the MTU of the interface.	mtu size	Optional. 1500 bytes by default.
5. Shut down the interface.	shutdown	Optional. Enabled by default.
6. Restore the default settings of the interface.	default	Optional.

Binding an RPR logical interface with RPR physical ports

To create an RPR station, you must create an RPR logical interface first and then bind RPR physical ports with the interface.

You may bind RPR physical ports with an RPR logical interface in RPR logical interface view or RPR physical port view. In either case, you must create the RPR logical interface first.

Binding RPR physical ports with an RPR logical interface in RPR logical interface view

To bind two RPR physical ports with an RPR logical interface:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPR logical interface view.	interface { rpr \| rpr-bridge } interface-number	N/A
3. Bind an RPR physical port with the RPR logical interface.	rpr bind { interface-type interface-number } { ringlet0 \| ringlet1 }	By default, no RPR physical ports are bound with an RPR logical interface.

Binding RPR physical ports with an RPR logical interface in RPR physical port view

To bind RPR physical ports with an RPR logical interface in RPR physical port view:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPR physical port view.	interface { rprpos \| rprxge } interface-number	N/A
3. Bind the RPR physical port with an RPR logical interface.	rpr bind { { rpr \| rpr-bridge } interface-number } { ringlet0 \| ringlet1 }	By default, no RPR physical ports are bound with an RPR logical interface.

Enabling RPR mate port smart connection

For an RPR station to forward traffic, before binding its two RPR physical ports with the same RPR logical interface, you must connect their mate ports.

When the two RPR physical ports of distributed RPR are on the same subcard, you can enable the RPR mate port smart connection function. With the function enabled, RPR automatically connects the two mate ports without you having to connect them manually.

To enable RPR mate port smart connection:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPR logical interface view.	interface { rpr \| rpr-bridge } interface-number	N/A
3. Enable RPR mate port smart connection.	rpr mate smart-connect	Disabled by default. This command takes effect only on the logical interface bound with two 2.5GPOS RPR ports. It cannot take effect on the logical interface bound with 10GPOS and 10GE RPR physical ports.

Changing the RPR physical port type

To change the RPR physical port type on an RPR logical interface:

Step	Command
1. Enter system view.	system-view
2. Enter RPR physical port view.	interface { rprpos \| rprxge } interface-number
3. Change the RPR physical port type.	rpr port-type { 10gpos \| 10ge }

CAUTION:

· After you change the RPR physical port type, the subcard will reboot to have the change take effect, and the previous configuration of the RPR physical port will be lost.

· The rpr port-type command takes effect only on 10G POS and 10-GE RPR physical ports. It cannot take effect on 2.5G POS ports.

Configuring the RPR station name

To assign a name to the RPR station:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPR logical interface view.	interface { rpr \| rpr-bridge } interface-number	N/A
3. Configure a station name.	rpr station-name station-name	No station name is configured by default.

Configuring RPR protection

Configuring the RPR protection mode

To configure the protection mode on the RPR station:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPR logical interface view.	interface { rpr \| rpr-bridge } interface-number	N/A
3. Configure the protection mode.	rpr protect-mode { steer \| wrap }	By default, the steering mode applies.

CAUTION:

All the stations on an RPR ring must use the same protection mode for the ring to operate normally.

Configuring RPR protection reversion mode

Two protection reversion modes are available: revertive and non-revertive.

· In revertive mode, a station resumes the idle protection state once the WTR timer expires.

· In non-revertive mode, the station remains in automatic protection state upon expiration of the WTR timer and does not resume the idle state until a higher protection request is present on the ring.

To configure the RPR protection reversion mode on the RPR station:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPR logical interface view.	interface { rpr \| rpr-bridge } interface-number	N/A
3. Configure the protection reversion mode.	rpr reversion-mode { revertive \| non-revertive }	Revertive mode applies by default.

Manually sending a protection request

You can send FS or MS protection requests to trigger protection switchover or send idle protection requests to clear manually sent protection requests on the RPR station.

To send a protection request:

Step	Command
1. Enter system view.	system-view
2. Enter RPR logical interface view.	interface { rpr \| rpr-bridge } interface-number
3. Send an RPR protection request on a ringlet.	rpr admin-request { fs \| ms \| idle } { ringlet0 \| ringlet1 }

Configuring the ringlet selection table

Each RPR station maintains a ringlet selection table for making ringlet selection decisions for frame transmission. A ringlet selection table entry contains such information as the MAC address of a station, the ringlet through which frames are transmitted.

The RPR ringlet selection table comprises a static ringlet selection table, a dynamic ringlet selection table, a default ringlet selection table, and an overall ringlet selection table.

· Static ringlet selection table is administratively configured to specify the ringlet for delivering a frame to a destination station.

· Dynamic ringlet selection table is built dynamically based on the topology database.

· Default ringlet selection table specifies the default ringlet for delivering frames.

· Overall ringlet selection table is created from the above three tables. Static ringlet selection table has the highest priority. If a static ringlet selection entry is available for reaching a destination station, it is added into the overall ringlet selection table. If no static entry is available, the system looks at the dynamic ringlet selection table. If two shortest paths are available, default ringlet selection applies to pick up either of them. The selected entry is then added into the overall ringlet selection table. When sending frames, the station consults this table to identify the path to the destination MAC address.

Adding a static ringlet selection entry

To add an entry to the static ringlet selection table:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPR logical interface view.	interface { rpr \| rpr-bridge } interface-number	N/A
3. Add a static ringlet selection entry.	rpr static-rs mac-address { ringlet0 \| ringlet1 }	No static ringlet selection entry is created by default.

CAUTION:

Static ringlet selection entries take effect only when the RPR ring is closed.

Configuring default ringlet selection

To configure the default ringlet selection:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPR logical interface view.	interface { rpr \| rpr-bridge } interface-number	N/A
3. Configure default ringlet selection.	rpr default-rs { ringlet0 \| ringlet1 }	Optional. By default, the default ringlet is Ringlet 0.

CAUTION:

When a span is faulty and unable to forward data on the default ringlet you configured, another ringlet works as the default ringlet.

Configure the RPR fairness algorithm

By configuring the RPR fairness algorithm, you can better guarantee the transmission quality over RPR rings.

Configuring reserved bandwidth or rate limiting

RPR traffic fall into three classes: class A, class B, and class C, with decreasing priorities.

· Class A is further divided into two subclasses: A0 and A1. RPR can reserve bandwidth for subclass A0. When congestion occurs, the unused reserved bandwidth cannot be used by lower priority traffic. Subclass A1 is rate limited. When congestion occurs, subclass A1 allows lower priority traffic to use its unused bandwidth and the bandwidth allocation in this case is regulated by the fairness algorithm.

· Class B can also be divided into two subclasses: B-CIR and B-EIR. For class B traffic, the portion within the predefined rate limit belongs to class B-CIR and the portion exceeding the limit belongs to class B-EIR. Similar to traffic of class C, traffic of class B-EIR is regulated by the fairness algorithm.

· Traffic of class C is regulated by the fairness algorithm and takes the lowest priority.

To configure reserved bandwidth or rate limit for a certain service class:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPR logical interface view.	interface { rpr \| rpr-bridge } interface-number	N/A
3. Configure reserved bandwidth or rate limit for a class on a certain ringlet.	rpr rate-limiter { high \| low \| medium \| reserved } { ringlet0 \| ringlet1 } value	By default, no bandwidth is reserved for subclass A0; the rate limit is 2‰ for subclass A1, 0‰ for class B-CIR, and 1000‰ for B-EIR and class C.

NOTE:

The total bandwidth reserved for subclass A0 by the stations on a ringlet must be less than the ringlet bandwidth.

Configuring station weight

On an RPR ring, bandwidth resources are shared among stations. When traffic size is small, RPR can handle bandwidth requests of all stations on the ring well. When traffic size gets heavy, congestion may occur. To prevent some stations from taking advantage of their spatially or chronologically favorable position on the ring to occupy excessive bandwidth, RPR adopts a fairness algorithm to ensure bandwidth allocation fairness.

The fairness algorithm of RPR mainly regulates class B-EIR and class C traffic on the RPR ring. You can assign a weight to a station to adjust the ratio of its available service bandwidth to the total non-reserved bandwidth on a ringlet. The fairness weight of a station is configured as an exponent of 2.

To configure the fairness weight of an RPR station on a ringlet:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPR logical interface view.	interface { rpr \| rpr-bridge } interface-number	N/A
3. Configure the weight of the station.	rpr weight { ringlet0 \| ringlet1 } value	By default, the weight is 2⁰=1.

Configuring RPR timers

These are the RPR timer configuration tasks:

· Configuring the ATD timer

· Configuring the hold off timer

· Configuring the keepalive timer

· Configuring the topology stability timer

· Configuring the TC timers

· Configuring the TP timers

· Configuring the WTR timer

Configuring the ATD timer

The ATD timer defines the interval at which attribute discovery frames are sent.

To configure the ATD timer:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPR logical interface view.	interface { rpr \| rpr-bridge } interface-number	N/A
3. Set the ATD timer.	rpr timer atd atd-value	1 second by default

Configuring the hold off timer

The hold off timer defines the delay for the physical layer to report a protection request after detecting a link failure.

To configure the hold off timer:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPR logical interface view.	interface { rpr \| rpr-bridge } interface-number	N/A
3. Set the hold off timer.	rpr timer holdoff holdoff-value	0 milliseconds by default

Configuring the keepalive timer

A station sends single choke fairness frames (SCFFs) periodically to notify the neighbor stations of its normal operation state. When a station fails to receive an SCFF, a keepalive timer starts. If no SCFF frame is received after the timer expires, the station sends an SF protection request.

To configure the keepalive timer:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPR logical interface view.	interface { rpr \| rpr-bridge } interface-number	N/A
3. Set the keepalive timer.	rpr timer keepalive keepalive-value	3 milliseconds by default

Configuring the topology stability timer

When a station detects a topology change on the ring, it starts the topology stability timer and begins to collect topology information to update its topology database. After the timer expires, the station checks the validity of received topology information. If the information is valid, the station enters topology valid state; if not, the station re-starts the timer.

To configure the topology stability timer:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPR logical interface view.	interface { rpr \| rpr-bridge } interface-number	N/A
3. Set the topology stability timer.	rpr timer stability stability-value	40 milliseconds by default

Configuring the TC timers

The TC fast/slow timer defines the fast/slow interval at which TC frames are sent. When the ring topology checksum changes, five TC frames are sent at the fast interval; when the ring topology is stable, TC frames are sent at the slow interval.

To configure the TC timers:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPR logical interface view.	interface { rpr \| rpr-bridge } interface-number	N/A
3. Set the TC fast timer.	rpr timer tc-fast tc-fast-value	10 milliseconds by default
4. Set the TC slow timer.	rpr timer tc-slow tc-slow-value	100 milliseconds by default

Configuring the TP timers

The TP fast/slow timer defines the fast/slow interval at which TP frames are sent. When a station starts initializing or detects a topology change, nine TP frames are sent at the fast interval. When the ring topology is stable, the station begins to send TP frames at the slow interval.

To configure the TP timers:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPR logical interface view.	interface { rpr \| rpr-bridge } interface-number	N/A
3. Set the TP fast timer.	rpr timer tp-fast tp-fast-value	10 milliseconds by default
4. Set the TP slow timer.	rpr timer tp-slow tp-slow-value	100 milliseconds by default

Configuring the WTR timer

When a protection switching event occurs on a station due to link failure, the station enters automatic protection state; after the link recovers, the station enters the idle protection state. The WTR timer defines the delay that a station transits to the idle protection state after entering the automatic protection state.

To configure the WTR timer:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPR logical interface view.	interface { rpr \| rpr-bridge } interface-number	N/A
3. Set the WTR timer.	rpr timer wtr wtr-value	10 milliseconds by default

Testing connectivity between RPR stations

You may use Echo Request/Echo Response messages to test the connectivity between two stations and locate failure points if any. The connectivity between the current station and the destination station is considered as normal if the connection between the two stations is normal on both the sending ringlet and the receiving ringlet. In this case, the destination station should be able to receive the Echo Request messages that the current station sends along the specific ringlet and the current station should be able to receive the Echo Response messages that the destination station sends along the specific ringlet. Otherwise, the connectivity between the current station and the destination station is considered as having failed.

To test the connectivity to a specific RPR station:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enter RPR logical interface view.	interface { rpr \| rpr-bridge } interface-number	N/A
3. Test the connectivity between the current station and a specific station.	rpr echo mac mac-address [ -c c-value \| -r { ringlet0 \| ringlet1 \| reverse } \| -s { ringlet0 \| ringlet1 } \| -t t-value ] *	By default, five Echo Request packets are sent; it is the default ringlet that sends Echo Response packets and Echo Request packets; the timeout time is 10 milliseconds.

CAUTION:

If neither the sending ringlet nor the receiving ringlet is specified, the default ringlet currently in use will transmit Echo Request packets and Echo Response.

Displaying and maintaining RPR

Task	Command	Remarks
Display information about an RPR physical port.	display interface { rprxge \| rprpos } interface-number [ \| { begin \| exclude \| include } regular-expression ]	Available in any view
Display information about an RPR logical interface.	display interface { rpr \| rpr-bridge } interface-number [ \| { begin \| exclude \| include } regular-expression ]	Available in any view
Display RPR logical interface-to-RPR physical port bindings.	display rpr bind-info [ { rpr \| rpr-bridge } interface-number ] [ \| { begin \| exclude \| include } regular-expression ]	Available in any view
Display RPR defect information.	display rpr defect [ { rpr \| rpr-bridge } interface-number ] [ \| { begin \| exclude \| include } regular-expression ]	Available in any view
Display RPR fairness settings.	display rpr fairness [ { rpr \| rpr-bridge } interface-number ] [ \| { begin \| exclude \| include } regular-expression ]	Available in any view
Display RPR protection information.	display rpr protection [ { rpr \| rpr-bridge } interface-number ] [ \| { begin \| exclude \| include } regular-expression ]	Available in any view
Display RPR ringlet selection table information.	display rpr rs-table { default \| dynamic \| overall \| static } [ { rpr \| rpr-bridge } interface-number ] [ \| { begin \| exclude \| include } regular-expression ]	Available in any view
Display the settings of all configurable RPR timers.	display rpr timers [ { rpr \| rpr-bridge } interface-number ] [ \| { begin \| exclude \| include } regular-expression ]	Available in any view
Display RPR topology information.	display rpr topology { all \| local \| ring \| stations } [ summary ] [ { rpr \| rpr-bridge } interface-number ] [ \| { begin \| exclude \| include } regular-expression ]	Available in any view
Display RPR MAC address table information.	display rpr mac-address [ static \| dynamic ] [ destination mac-address1 ] [ vlan vlan-id ] [ ring mac-address2 ] [ rpr-bridge interface-number ] [ count ] [ \| { begin \| exclude \| include } regular-expression ]	Available in any view
Clear the statistics about protection events on the RPR station.	reset rpr protection statistics [ { rpr \| rpr-bridge } interface-number ]	Available in user view
Clear the statistics of the specified RPR physical ports.	reset counters interface [ { rprpos \| rprxge } [ interface-number ] ]	Available in any view

RPR configuration examples

This section provides two examples:

· Layer 3 RPR interface binding configuration example

· Layer 2 RPR interface binding configuration example

· RPR protection mode/static ringlet selection configuration example

Layer 3 RPR interface binding configuration example

Network requirements

As shown in Figure 6, stations A through E form an RPR ring. The system working mode of the switches is hybrid.

Bind the RPR physical ports to the RPR logical interface on each station.

Figure 6 Network diagram

Configuration procedure

1. Create an RPR logical interface and bind two RPR physical ports to the RPR logical interface

# On Station A, create Layer 3 RPR logical interface RPR1, and bind RPR physical ports RPRPOS 2/1/1 (the west port) and RPRPOS 2/1/2 (the east port) to RPR 1.

<StationA> system-view

[StationA] interface rpr 1

[StationA-RPR1] rpr bind rprpos 2/1/1 ringlet0

[StationA-RPR1] rpr bind rprpos 2/1/2 ringlet1

[StationA-RPR1] quit

The configuration steps on Station B, C, D, and E are similar to those on Station A.

2. Verify the configuration

# Display RPR interface binding information on Station A.

[StationA-RPR1] display rpr bind-info

Bind information on interface: RPR1

PHY-Interface Ringlet-ID Role Mate-Port

---------------------------------------------------

RPRPOS2/1/1 Ringlet0 Master Up

RPRPOS2/1/2 Ringlet1 Slave Up

# Display the RPR topology summary on Station A.

[StationA] display rpr topology all summary

Topology information items

Psw:protection state, west Pse:protection state, east

Esw:edge state, west Ese:edge state, east

Wc:wrap protection configured Jp:jumbo frame preferred

Ring-level topology information on interface: RPR1

Ringlet0 Ringlet1 Ring Jumbo-Prefer Topology-Type

-------------------------------------------------

4 4 5 Regular Closed ring

Local station topology information on interface: RPR1

MAC-Address Psw Pse Esw Ese Wc Jp IP-Address Station-Name

---------------------------------------------------------------------

000f-e257-0001 Idle Idle 0 0 0 0 0.0.0.0

Station topology information on interface: RPR1

Station entry on ringlet0

MAC-Address Psw Pse Esw Ese Wc Jp IP-Address Station-Name

---------------------------------------------------------------------

000f-e257-0005 Idle Idle 0 0 0 0 0.0.0.0

000f-e257-0004 Idle Idle 0 0 0 0 0.0.0.0

000f-e257-0003 Idle Idle 0 0 0 0 0.0.0.0

000f-e257-0002 Idle Idle 0 0 0 0 0.0.0.0

Station entry on ringlet1

MAC-Address Psw Pse Esw Ese Wc Jp IP-Address Station-Name

---------------------------------------------------------------------

000f-e257-0002 Idle Idle 0 0 0 0 0.0.0.0

000f-e257-0003 Idle Idle 0 0 0 0 0.0.0.0

000f-e257-0004 Idle Idle 0 0 0 0 0.0.0.0

000f-e257-0005 Idle Idle 0 0 0 0 0.0.0.0

The topology information shows that the RPR ring is closed.

Layer 2 RPR interface binding configuration example

Network requirements

As shown in Figure 7, stations A through E form an RPR ring. The system working mode of the switches is hybrid.

Bind the RPR physical ports to the RPR logical interface on each station.

Figure 7 Network diagram

Configuration procedure

1. Create a Layer 2 RPR logical interface RPR-Bridge 1 and bind two RPR physical ports to the RPR logical interface

# On Station A, create Layer 2 RPR logical interface RPR-Bridge 1, and bind RPR physical ports RPRPOS 3/1/1 (the west port) and RPRPOS 5/1/2 (the east port) to RPR-Bridge 1.

<StationA> system-view

[StationA] interface RPR-Bridge 1

[StationA-RPR-Bridge1] rpr bind rprpos 3/1/1 ringlet0

[StationA-RPR-Bridge1] rpr bind rprpos 5/1/2 ringlet1

# A Layer 2 RPR logical interface is similar to a Layer 2 Ethernet port. Configure it as a trunk port, and assign it to VLAN 100. Create VLAN-interface 100, and assign an IP address to it.

[StationA-RPR-Bridge1] port link-type trunk

[StationA-RPR-Bridge1] port trunk permit vlan 100

[StationA-RPR-Bridge1] undo port trunk permit vlan 1

[StationA-RPR-Bridge1] port trunk pvid vlan 100

[StationA-RPR-Bridge1] quit

[StationA] interface Vlan-interface 100

[StationA-Vlan-interface100] ip address 10.1.1.1 24

[StationA-RPR-Bridge1] quit

The configuration steps on Station B, C, D, and E are similar to those on Station A. Configure the IP addresses of VLAN interfaces on Station B, C, D, and E as 10.1.1.2, 10.1.1.3, 10.1.1.4, and 10.1.1.5, respectively.

2. Verify the configuration

# Display RPR interface binding information on Station A.

[StationA-RPR-Bridge1] display rpr bind-info

Bind information on interface: RPR-Bridge1

Smart-connection: Disabled

PHY-Interface Ringlet-ID Role Mate-Port

---------------------------------------------------

RPRPOS3/1/1 Ringlet0 Master Up

RPRPOS5/1/2 Ringlet1 Slave Up

# Display the RPR topology summary on Station A.

[StationA] display rpr topology all summary

Topology information items

Psw:protection state, west Pse:protection state, east

Esw:edge state, west Ese:edge state, east

Wc:wrap protection configured Jp:jumbo frame preferred

Ring-level topology information on interface: RPR-Bridge1

Ringlet0 Ringlet1 Ring Jumbo-Prefer Topology-Type

-------------------------------------------------

4 4 5 Jumbo Closed ring

Local station topology information on interface: RPR-Bridge1

MAC-Address Psw Pse Esw Ese Wc Jp IP-Address Station-Name

---------------------------------------------------------------------

000f-e257-0001 Idle Idle 0 0 0 1 10.1.1.1

Station topology information on interface: RPR-Bridge1

Station entry on ringlet0

MAC-Address Psw Pse Esw Ese Wc Jp IP-Address Station-Name

---------------------------------------------------------------------

000f-e257-0005 Idle Idle 0 0 0 1 10.1.1.5

000f-e257-0004 Idle Idle 0 0 0 1 10.1.1.4

000f-e257-0003 Idle Idle 0 0 0 1 10.1.1.3

000f-e257-0002 Idle Idle 0 0 0 1 10.1.1.2

Station entry on ringlet1

MAC-Address Psw Pse Esw Ese Wc Jp IP-Address Station-Name

---------------------------------------------------------------------

000f-e257-0002 Idle Idle 0 0 0 1 10.1.1.2

000f-e257-0003 Idle Idle 0 0 0 1 10.1.1.3

000f-e257-0004 Idle Idle 0 0 0 1 10.1.1.4

000f-e257-0005 Idle Idle 0 0 0 1 10.1.1.5

The topology information shows that the RPR ring is closed.

RPR protection mode/static ringlet selection configuration example

Network requirements

As shown in Figure 8, Stations A through E form an RPR ring and work in steering protection mode by default. The system working mode of the switches is hybrid.

Change the protection mode of the ring to wrapping, and configure a static ringlet selection entry to allow frames from Station A to Station B to travel Ringlet 1 when no edge occurs on the RPR ring (the ring remains closed).

Figure 8 Network diagram

Configuration procedure

1. Configure the protection mode

# Create Layer 3 RPR logical interface RPR 1 on Station A, and configure the protection mode as wrapping on the interface.

<StationA> system-view

[StationA] interface rpr 1

[StationA-RPR1] rpr protect-mode wrap

The configuration steps on Station B through Station E are similar to those on Station A.

2. Configure static ringlet selection

# On Layer 3 RPR logical interface RPR 1 of Station A, configure transmitting frames destined for Station B through Ringlet 1.

[StationA-RPR1] rpr static-rs 000f-e257-0002 ringlet1

[StationA-RPR1] quit

3. Verify the configuration

# Display ring topology information on Station A.

[StationA] display rpr topology ring

Ring-level topology information on interface: RPR1

Stations on ringlet0: 4

Stations on ringlet1: 4

Total stations on ring: 5

Jumbo preference: regular

Ring topology type: closed ring

# Display the static ring selection table information on Station A.

[StationA] display rpr rs-table static

Static ringlet selection table on interface: RPR1

MAC-Address Ringlet-ID Status

-----------------------------------

000f-e257-0002 Ringlet1 Valid

--- Total entrie(s): 1 ---

# Display the information about the overall ring selection table on Station A.

[StationA] display rpr rs-table overall

Overall ringlet selection table on interface: RPR1

MAC-Address Ringlet-ID TTL Type IP-Address Station-Name

-------------------------------------------------------------------------------

000f-e257-0002 Ringlet1 4 static 0.0.0.0

000f-e257-0003 Ringlet0 2 dynamic 0.0.0.0

000f-e257-0004 Ringlet1 2 dynamic 0.0.0.0

000f-e257-0005 Ringlet1 1 dynamic 0.0.0.0

--- Total entrie(s): 4 ---

12-High Availability Configuration Guide

Configuring RPR

Overview

Ring structure of RPR

Data operations on RPR

Unicast transmission

Broadcast, multicast, and unknown unicast transmission

Topology discovery

Topology database

Topology discovery process

Fault response methods

Passthrough

Protection

Centralized RPR and distributed RPR

Definitions

Differences

RPR interface

RPR logical interface and RPR physical ports

RPR master/slave port

Protocols and standards

RPR configuration task list

Creating and configuring a Layer 2 RPR logical interface

Binding RPR physical ports with an RPR logical interface in RPR logical interface view

Binding RPR physical ports with an RPR logical interface in RPR physical port view

Manually sending a protection request

Configure the RPR fairness algorithm

Configuring station weight

Configuring RPR timers

Testing connectivity between RPR stations

RPR configuration examples

Network requirements

Configuration procedure

Network requirements

Configuration procedure

Network requirements

Configuration procedure

Intelligent Terminal Products

Product Support Services

Technical Service Solutions

Resource Center

Policy

Online Help

Become a Partner

Partner Policy & Program

Global Learning

Partner Sales Resources

Service Business

News & Events

Contact Us