HP 5500 Ei 5500 Si Switch Series Configuration Guide
Have a look at the manual HP 5500 Ei 5500 Si Switch Series Configuration Guide online for free. It’s possible to download the document as PDF or print. UserManuals.tech offer 1114 HP manuals and user’s guides for free. Share the user manual or guide on Facebook, Twitter or Google+.
240 [SwitchA-GigabitEthernet1/0/1] quit [SwitchA] interface gigabitethernet 1/0/2 [SwitchA-GigabitEthernet1/0/2] lldp enable [SwitchA-GigabitEthernet1/0/2] lldp admin-status txrx [SwitchA-GigabitEthernet1/0/2] lldp compliance admin-status cdp txrx [SwitchA-GigabitEthernet1/0/2] quit 3. Verify the configuration: # Display the neighbor information on Switch A. [SwitchA] display lldp neighbor-information CDP neighbor-information of port 1[GigabitEthernet1/0/1]: CDP neighbor index : 1 Chassis ID : SEP00141CBCDBFE Port ID : Port 1 Sofrware version : P0030301MFG2 Platform : Cisco IP Phone 7960 Duplex : Full CDP neighbor-information of port 2[GigabitEthernet1/0/2]: CDP neighbor index : 2 Chassis ID : SEP00141CBCDBFF Port ID : Port 1 Sofrware version : P0030301MFG2 Platform : Cisco IP Phone 7960 Duplex : Full As the sample output shows, Switch A has discov ered the IP phones connected to GigabitEthernet 1/0/1 and GigabitEthernet 1/0/2, and has ob tained their LLDP device information.
241 Configuring a service loopback group (available only on the 5500 EI) Overview To increase traffic redirecting throughput, you can bundle multiple Ethernet ports of a device together to increase bandwidth and implement load sharing. These ports that act as a logical link form a service loopback group. A service loopback group must contain at least one Ethernet port as its member port, called a service loopback port. For example, by assigning three Ethernet ports of the same device to a service loopback group, you can create a logical link whose bandwidth can be as high as the total bandwidth of these three ports. In addition, service traffic is load balanced among these ports. Service types of service loopback groups A service loopback group is applicable to a specified service type. The service types supported by the service loopback group are: • Tunn e l —Supports unicast tunnel traffic Requirements on service loopback ports Before you assign a port to a service loopback group, ensure the following: • The port supports the services type or types of the service loopback group. • The port is not configured with multiple spanning tree protocol (MSTP), Link Layer Discovery Protocol (LLDP), Neighbor Discovery Protocol (NDP), 802.1X, MAC address authentication, port security mode, or IP source guard, or as the member port of an isolation group. • The link type of the port is access. • The port is not a member of any Ethernet link aggregation group or service loopback group. States of service loopback ports A member port in a service loopback group is a serv ice loopback port, which can be in either of the following states: • Selected —A selected port can loop back user traffic. • Unselected —An unselected port cannot loop back user traffic. The number of selected ports is limited in a service loopback group. Setting the state of service loopback ports The system sets the state of each member port in a service loopback group to selected or unselected by using the following workflow: 1. Select the full-duplex port with the highest rate as the reference port. If two ports have the same duplex mode and speed, the one with the lower port number wins.
242 2. Set the state of each member port in the service loopback group. Figure 78 Setting the state of each member port in a service loopback group NOTE: Every time a new port is assigned to a service loo pback group, the system resets the state of the member ports in the service loopback group accord ing to the process described previously. Configuration restrictions and guidelines • A service loopback group may be referenced by mu ltiple features once it is created, and can process service traffic only after it is referenced. • You can change the service type of an existing service loopback group. For the change to be successful, make sure that the service group has not been referenced, the attributes of all member ports (if any) do not conflict with the target service type, and no service loopback group has been created for the target service type, because only one service loopback group is allowed for a service type. • You cannot remove a service loopback group that other features reference. Configuration procedure Step Command Remarks 1. Enter system view. system-view N/A 2. Create a service loopback group and specify its service type. service-loopback group number type tunnel N/A
243 Step Command Remarks 3. Enter Layer 2 Ethernet interface view. interface interface-type interface-number N/A 4. Assign the Ethernet interface to the specified service loopback group. port service-loopback group number By default, a port does not belong to any service loopback group. You can perform this command on different ports to assign multiple ports to a service loopback group. Displaying and maintaining service loopback groups Task Command Remarks Display information about the specified service loopback group or all service loopback groups. display service-loopback group [ number ] [ | { begin | exclude | include } regular-expression ] Available in any view Service loopback group configuration example Network requirements All the Ethernet ports of Device A support the tu nnel service. Assign GigabitEthernet1/0/1 through GigabitEthernet1/0/3 to a service loopback group to increase bandwidth and achieve load sharing for tunnel traffic. Configuration procedure # Create service loopback group 1, and specify the service type as tunnel (unicast tunnel service). system-view [DeviceA] service-loopback group 1 type tunnel # Disable MSTP, LLDP, and NDP on GigabitEthernet1/0/1 through GigabitEthernet1/0/3 and then assign them to service loopback group 1. [DeviceA] interface gigabitethernet 1/0/1 [DeviceA-GigabitEthernet1/0/1] undo stp enable [DeviceA-GigabitEthernet1/0/1] undo lldp enable [DeviceA-GigabitEthernet1/0/1] undo ndp enable [DeviceA-GigabitEthernet1/0/1] port service-loopback group 1 [DeviceA-GigabitEthernet1/0/1] quit [DeviceA] interface gigabitethernet 1/0/2 [DeviceA-GigabitEthernet1/0/2] undo stp enable [DeviceA-GigabitEthernet1/0/2] undo lldp enable [DeviceA-GigabitEthernet1/0/2] undo ndp enable [DeviceA-GigabitEthernet1/0/2] port service-loopback group 1 [DeviceA-GigabitEthernet1/0/2] quit
244 [DeviceA] interface gigabitethernet 1/0/3 [DeviceA-GigabitEthernet1/0/3] undo stp enable [DeviceA-GigabitEthernet1/0/3] undo lldp enable [DeviceA-GigabitEthernet1/0/3] undo ndp enable [DeviceA-GigabitEthernet1/0/3] port service-loopback group 1 [DeviceA-GigabitEthernet1/0/3] quit # Create logical interface Tunnel 1 and reference service loopback group 1 on Tunnel 1. [DeviceA] interface tunnel 1 [DeviceA-Tunnel1] service-loopback-group 1
245 Configuring MVRP Overview Multiple Registration Protocol (MRP) is an attribute registration protocol and transmits attribute messages. Multiple VLAN Registration Protocol (MVRP) is a ty pical MRP application. MVRP propagates and learns VLAN configuration among devices. MVRP enab les a device to propagates the local VLAN configuration to the other devices, receive VLAN configuration from other devices, and dynamically update the local VLAN configuration (including the active VLANs and the ports through which a VLAN can be reached). MVRP makes sure that all MVRP-enabled devices in a LAN maintain the same VLAN configuration, and reduces the VLAN configuration workload. When the network topology changes, MVRP can propagate and learn VLAN configuration information again according to the new topology, and real-time synchronize the network topology. MRP is an enhanced version of Generic Attribute Registration Protocol (GARP) and improves the declaration efficiency. MVRP is an enhanced version of GARP VLAN Registration Protocol (GVRP). MVRP delivers the following benefits over GVRP: • GVRP does not support the multiple spanning tree instance (MST I). MVRP runs on a per-MSTI basis, and implements per-VLAN redundant link calculation and load sharing. • MVRP decreases the number of packets transmitted for the same amount of VLAN configuration, and improves the declaration efficiency. For more information about GVRP, see Configuring GVRP. F or more information about MSTI, see Configuring spanning tree protocols . Introduction to MRP MRP allows participants in the same LAN to declare, propagate, and register information (for example, VLAN information) on a per Multiple S panning Tree Instance (MSTI) basis. MRP implementation Each port that participates in an MRP application (for example, MVRP) is called an MRP participant. Similarly, a port that participates in an MVRP application is called an MVRP participant. As shown in Figure 79, an MRP participant registers and deregi sters its attribute values on other MRP participants by sending declarations and withdrawals, and registers and deregisters the attribute values of other participants according to the received declarations and withdrawals. MRP rapidly propagates the configuration information of an MRP participant throughout the LAN. Figure 79 MRP implementation
246 MVRP registers and deregisters VLAN attributes as follows: • When a port receives the declaration of a VLAN attribute, the port registers the VLAN and joins the VLAN. • When a port receives the withdrawal of a VLAN at tribute, the port deregisters the VLAN and leaves the VLAN. Figure 79 s hows a simple MVRP implementation on an MSTI. In a network with multiple MSTIs, VLAN registration and deregistration are performed on a per-MSTI basis. MRP messages MRP exchanges information among MRP participants by advertising MRP messages, including Join, New, Leave, and LeaveAll. Join and New messages are declarations, and Leave and LeaveAll messages are withdrawals. • Join message { An MRP participant sends Join messages when it wishes to declare the attribute values configured on it and receives Join messages from other MRP participants. { When receiving a Join message, an MRP partici pant sends a Join message to all participants except the sender. Join messages fall into the following types: { JoinEmpty —An MRP participant sends JoinEmpty messages to declare attribute values that it has not registered. For example, when a static VLAN exists on a device, the attribute of the VLAN on the device is not changed even if the device learns the VLAN again through MRP. In this case, the Join message for the VLAN attribute is a JoinEmpty message, because the VLAN attribute is not registered. { JoinIn —An MRP participant sends JoinIn messages to declare attribute values that it has registered. For example, when the device learns a VLAN through MRP messages, and dynamically creates the VLAN, the Join message for the VLAN attribute is a JoinIn message. • New message Similar to a Join message, a New message enables MRP participants to register attributes. { When the Multiple Spanning Tree Protocol (MSTP) topology changes, an MRP participant sends New messages to declare the topology change. { On receiving a New message, an MRP participant sends a New message out of each port except the receiving port. • Leave message { An MRP participant sends Leave messages when it wishes other participants to deregister the attributes that is has deregistered. { When receiving a Leave message, an MRP participant sends a Leave message to all participants except the sender. • LeaveAll message { Each MRP participant is configured with an individual LeaveAll timer. When the timer expires, the MRP participant sends LeaveAll messages to the remote participants, so that the local participant and the remote participants deregister all attributes and re-register all attributes. This process periodically clears the useless attributes in the network. { On receiving a LeaveAll message, MRP determines whether to send a Join message to request the sender to re-register these attributes according to attribute status.
247 MRP timers The implementation of MRP uses the following timers to control MRP message transmission. • Pe rio d ic t i m e r On startup, an MRP participant starts its own Peri odic timer to control MRP message transmission. The MRP participant collects the MRP messages to be sent before the Periodic timer expires, and sends the MRP messages in as few packets as po ssible when the Periodic timer expires and meanwhile restarts the Periodic timer. This mechanism reduces the number of MRP protocol packets periodically sent. You can enable or disable the Periodic timer at th e CLI. When you disable the Periodic timer, MRP will not periodically send MRP messages. • Join timer The Join timer controls the transm ission of Join messages. To make sure that Join messages can be reliably transmitted to other participants, an MRP participant waits for a period of the Join timer after sending a Join message. If the participant re ceives JoinIn messages from other participants and the attributes in the JoinIn messages are the same as the sent Join messages before the Join timer expires, the participant does not re-send the Join message. When both the Join timer and the Periodic timer expire, the participan t re-sends the Join message. • Leave timer The Leave timer controls the deregistration of attributes. When an MRP participant wishes other participants to deregister its attributes, it send s a Leave message. On receiving a Leave message, MRP starts the Leave timer, and deregisters the attributes if it does not receive any Join message for the attributes before the Leave timer expires. Wh en an MRP participant sends or receives LeaveAll messages, it starts the Leave time r. MRP deregisters the attributes in the LeaveAll messages if it does not receive any Join message for the attributes before the Leave timer expires. • LeaveAll timer On startup, an MRP participant starts its own LeaveAll timer. When the LeaveAll timer expires, MRP sends out a LeaveAll message and restarts the Leav eAll timer. On receiving the LeaveAll message, other participants re-register all the attrib utes and re-start their LeaveAll timer. When the LeaveAll timer of an MRP participan t expires, the MRP participant sends LeaveAll messages to the remote parti cipants. On receiving a LeaveA ll message, a remote participant restarts its LeaveAll timer, and stops sending ou t LeaveAll messages. This mechanism effectively reduces the number of LeaveAll messages in the network. To avoid the case that the LeaveAll timer of a fi xed participant always first expires, the switch randomly changes the LeaveAll timer within a cert ain range when the MRP participant restarts its LeaveAll timer. MVRP registration modes The VLAN information propagated by MVRP includes no t only locally, manually configured static VL AN information but also dynamic VLAN information from other devices. VLANs created manually, locally are called static VLANs, and VLANs learned through MVRP are called dynamic VLANs. The following MVRP registration modes are available. • Normal An MVRP participant in normal registration mo de performs dynamic VLAN registrations and deregistrations, and sends declarations and withdrawals for dynamic and static VLANs. • Fixed
248 An MVRP participant in fixed registration mode disables deregistering dynamic VLANs, sends declarations for dynamic VLANs and static VLANs, and drops received MVRP protocol packets. As a result, an MVRP participant port in fixed regi stration mode does not deregister or register dynamic VLANs. • Forbidden An MVRP participant in forbidden registration mo de disables registering dynamic VLANs, sends declarations for dynamic VLANs and static VLANs, and drops received MVRP protocol packets. As a result, an MVRP participant in forbidden regi stration mode does not register dynamic VLANs, and does not re-register a dynamic VLAN when the VLAN is deregistered. Protocols and standards IEEE 802.1ak IEEE Standard for Local and Metropolitan Area Networks: Virtual Bridged Local Area Networks – Amendment 07: Multiple Registration Protocol MVRP configuration task list Task Remarks Configuration prerequisites Required. Enabling MVRP Required. Configuring the MVRP registration mode Optional. Configuring MRP timers Optional. Enabling GVRP compatibility Optional. Configuration prerequisites Before configuring MVRP, perform the following tasks: • Make sure that all MSTIs in the network are effective and each MSTI is mapped to an existing VLAN on each device in the network, because MVRP runs on a per-MSTI basis. • Configure the involved ports as trunk ports, beca use MVRP is available only on trunk ports. Enabling MVRP Configuration restrictions and guidelines • MVRP can work with STP, RSTP, or MSTP, but not other link layer topology protocols, including service loopback, PVST, RRPP, and Smart Link. Ports blocked by STP, RSTP, or MSTP can receive and send MVRP protocol packets. For more in formation about service loopback, see Configuring a se rvice loopback group (available only on the 5500 EI) . For more information about STP, RSTP, MSTP, and PVST, see Configuring spanning tree protocols . F or more information about RRPP and Smart Link, see High Availability Configuration Guide . • Do not enable both MVRP and remote port mirror ing on a port. Otherwise, MVRP may register the remote probe VLAN to incorrect ports, which wo uld cause the monitor port to receive undesired
249 duplicates. For more information about port mirroring, see Network Management and Monitoring Configuration Guide . • Enabling MVRP on a Layer 2 aggregate interface enables both the aggr egate interface and all Selected member ports in the link aggregation gr oup to participate in dynamic VLAN registration and deregistration. Configuration procedure To enable MVRP: Step Command Remarks 1. Enter system view. system-view N/A 2. Enable MVRP globally. mvrp global enable By default, MVRP is globally disabled. To enable MVRP on a port, first enable MVRP globally. 3. Enter interface view. • Enter Layer 2 Ethernet interface view or Layer 2 aggregate interface view: interface interface-type interface-number • Enter port group view: port-group manual port-group-name Use one of the commands. 4. Configure the port to permit the specified VLANs. port trunk permit vlan { vlan-list | all } By default, a trunk port permits only VLAN 1. Make sure that the trunk port permits all registered VLANs. For more information about the port trunk permit vlan { vlan-list | all } command, see Layer 2 —LAN Switching Command Reference . 5. Enable MVRP on the port. mvrp enable By default, MVRP is disabled on a port. Configuring the MVRP registration mode To configure the MVRP registration mode: Step Command Remarks 1. Enter system view. system-view N/A