Cluster management is implemented through Huawei
Group Management Protocol version 2 (HGMPv2). By employing HGMPv2, a network
administrator can manage multiple switches using the public IP address of one
switch in a cluster. The switch that is configured with a public address and
performs the management function is known as the management device; and other
managed switches are called member devices, which are managed and maintained by
the management device. The management device and member devices together form a
cluster.
In a complex network environment, a network
administrator uses HGMP to manage switches inside the network remotely, which
is timesaving and easy to operate.
The network administrator can manage member
devices through the management device only by performing the following
operations:
l
Create a cluster on the management device
l
Enable Neighbor Discovery Protocol (NDP) and
Neighbor Topology Discovery Protocol (NTDP) on member switches
l
Enable HGMP
Figure
1-1 Network diagram for HGMP
l
13 switches form a cluster
l
Switch A serves as the management device, and
the other switches are the member devices
l
The management VLAN of the cluster is VLAN 10
l
The IP address of the management switch is
163.172.55.1
l
The IP address of the TFTP server is 63.172.55.1
l
The IP address of the SNMP network management
(NM) host is 69.172.55.4
l
All the devices in the cluster use the same FTP
server and SNMP host.
Management device:
l
Port GigabitEthernet 1/0/1 belongs to VLAN 2,
whose interface IP address is 163.172.55.1
l
The management device connects to two member
devices through ports GigabitEthernet 1/0/2 and GigabitEthernet 1/0/3
respectively.
Member devices:
l
Switch B (member device) connects with Switch D,
E and F through ports GigabitEthernet 1/0/2, GigabitEthernet 1/0/3 and GigabitEthernet
1/0/4 respectively
l
Switch C (member device) connects with Switch G,
H and I through ports GigabitEthernet 1/0/2, GigabitEthernet 1/0/3 and GigabitEthernet
1/0/4 respectively.
l
Switch E (member device) connects with Switches
J and K through ports GigabitEthernet 1/0/2 and GigabitEthernet 1/0/3
respectively
l
Switches A, B and C are usually low-end Ethernet
switches, like S5500-SI series, and S5500-EI series.
l
Switches D, E and F are usually low-end Ethernet
switches, like S3610 series, and S5500 series.
|
Product series
|
Software version
|
Hardware version
|
|
S3610 Series Ethernet Switches
|
Release 5301
|
All versions
|
|
S5510 Series Ethernet Switches
|
Release 5301
|
All versions
|
|
S5500-SI Series Ethernet Switches
|
Release 1207
|
All versions except S5500-20TP-SI
|
|
Release 1301
|
S5500-20TP-SI
|
|
S5500-EI Series Ethernet Switches
|
Release 2102
|
All versions
|
1)
Configure a member device (taking Switch B as an
example)
# Enable NDP and NTDP on Switch B
<Sysname>
system-view
[Sysname] ndp
enable
[Sysname] ntdp
enable
# Enable NDP and NTDP
for ports GigabitEthernet 1/0/1, GigabitEthernet 1/0/2,
GigabitEthernet 1/0/3, and GigabitEthernet 1/0/4.
[Sysname]
interface GigabitEthernet 1/0/1
[Sysname-GigabitEthernet1/0/1]
ndp enable
[Sysname-GigabitEthernet1/0/1]
ntdp enable
[Sysname-GigabitEthernet1/0/1]
quit
[Sysname]
interface GigabitEthernet 1/0/2
[Sysname-GigabitEthernet1/0/2]
ndp enable
[Sysname-GigabitEthernet1/0/2]
ntdp enable
[Sysname-GigabitEthernet1/0/2]
quit
[Sysname]
interface GigabitEthernet 1/0/3
[Sysname-GigabitEthernet1/0/3]
ndp enable
[Sysname-GigabitEthernet1/0/3]
ntdp enable
[Sysname-GigabitEthernet1/0/3]
quit
[Sysname]
interface GigabitEthernet 1/0/4
[Sysname-GigabitEthernet1/0/4]
ndp enable
[Sysname-GigabitEthernet1/0/4]
ntdp enable
[Sysname-GigabitEthernet1/0/4]
quit
l
NDP and NTDP functions should be enabled for
member switch ports connecting with the devices in the cluster.
l
The management device may collect topology
information of a device that does not need to join the cluster and add this
device into the cluster. To avoid this, you are recommended to disable NDP on
the port which connects with this device.
l
The management device may collect topology
information of a device that does not need to join the cluster and add this
device into the cluster. To avoid this, you are recommended to disable NTDP on
the port which connects with this device.
2)
Configure the management device
# Enable the cluster
function.
[Sysname] cluster
enable
# Disable NDP on the
uplink port GigabitEthernet 1/0/1.
<Sysname>
system-view
[Sysname] ndp
enable
[Sysname] undo
ndp enable intferface GigabitEthernet 1/0/1
# Enable NDP for ports
GigabitEthernet 1/0/2 and GigabitEthernet 1/0/3.
[Sysname]
interface GigabitEthernet 1/0/2
[Sysname-GigabitEthernet1/0/2]
ndp enable
[Sysname-GigabitEthernet1/0/2]
quit
[Sysname]
interface GigabitEthernet 1/0/3
[Sysname-GigabitEthernet1/0/3]
ndp enable
[Sysname-GigabitEthernet1/0/3]
quit
# Set the period for
the receiving device to keep NDP packets to 300 seconds.
[Sysname] ndp
timer aging 300
# Set the interval to send NDP packets to
100 seconds.
[Sysname] ndp
timer hello 100
# Enable NTDP on the device and for ports
GigabitEthernet 1/0/2 and GigabitEthernet 1/0/3.
[Sysname] ntdp
enable
[Sysname]
interface GigabitEthernet 1/0/2
[Sysname-GigabitEthernet1/0/2]
ntdp enable
[Sysname-GigabitEthernet1/0/2]
quit
[Sysname]
interface GigabitEthernet 1/0/3
[Sysname-GigabitEthernet1/0/3]
ntdp enable
[Sysname-GigabitEthernet1/0/3]
quit
# Configure the hop
count to collect topology as 2.
[Sysname] ntdp
hop 2
# Configure the delay time for
topology-collection request packets to be forwarded on member devices as 180
ms.
[Sysname]
ntdp timer hop-delay 180
# Configure the delay time for
topology-collection request packets to be forwarded through the ports of member
devices as 20 ms.
[Sysname] ntdp
timer port-delay 20
# Configure the interval to collect
topology information as 3 minutes.
[Sysname] ntdp
timer 3
# Configure the
management VLAN of the cluster as VLAN 10.
[Switch] vlan
10
[Switch-vlan10]
quit
[Switch] management-vlan
10
# Configure the port
connecting the management device to candidate devices as a trunk port and allow
packets from the management VLAN to pass.
[Switch] interface
GigabitEthernet 1/0/1
[Switch-GigabitEthernet1/0/1]
port link-type trunk
[Switch-GigabitEthernet1/0/1]
port trunk permit vlan 10
[Switch-GigabitEthernet1/0/1]
quit
[Switch] interface
GigabitEthernet 1/0/3
[Switch-GigabitEthernet1/0/3]
port link-type trunk
[Switch-GigabitEthernet1/0/3]
port trunk permit vlan 10
[Switch-GigabitEthernet1/0/3]
quit
# Enable the cluster function.
[Sysname] cluster
enable
# Enter cluster view.
[Sysname]
cluster
[Sysname-cluster]
# Configure an IP
address pool for the cluster. The IP address pool contains 14 IP addresses,
starting from 172.16.0.1.
[Sysname-cluster]
ip-pool 172.16.0.1 255.255.255.240
# Specify a name for the cluster and create
the cluster.
[Sysname-cluster]
build aaa
[aaa_0.Sysname-cluster]
# Enable mangement VLAN auto-negotiation.
[aabbcc_0.Sysname-cluster]
management-vlan synchronization enable
# Set the holdtime of the member device
information to 100 seconds.
[aaa_0.Sysname-cluster]
holdtime 100
# Set the interval to send handshake
packets to 10 seconds.
[aaa_0.Sysname-cluster]
timer 10
# Configure the FTP
Server, TFTP Server and SNMP host for the cluster.
[aaa_0.Sysname-cluster]
tftp-server 63.172.55.1
[aaa_0.Sysname-cluster]
snmp-host 69.172.55.4
3)
Operations on the cluster members
Perform the following operations after the
management device adds the connected device into the cluster.
# The member device downloads the file aaa.txt
from the TFTP server of the cluster.
<aaa_1.Sysname>
tftp cluster get aaa.txt
# The member device uploads the file bbb.txt
to the TFTP server of the cluster.
<aaa_1.Sysname>
tftp cluster put bbb.txt
Configuration on switch A
#
ntdp hop 2
ntdp timer
port-delay 20
ntdp timer
hop-delay 180
ntdp timer 3
#
ndp timer
hello 100
ndp timer
aging 300
#
cluster
ip-pool
172.16.0.1 255.255.255.248
build aaa
holdtime 100
tftp-server 63.172.55.1
snmp-host 69.172.55.4
community-string read public write private
#
Follow these guidelines when configuring
cluster management:
l
Upon completion of the above configurations, you
can execute the cluster switch-to { member-number | mac-address
H-H-H } command on the management device to switch to the operation
interface of a member device to maintain and manage it. You can then execute
the cluster switch-to administrator command to return to the operation
interface of the management device.
l
You can also reboot a member device by executing
the reboot member { member-number | mac-address H-H-H
} [ eraseflash ] command on the management
device.
l
You can also receive at the SNMP host the logs
and SNMP traps of all the member devices.
l
The switches cannot act as a TFTP server.
l
The management VLAN of a cluster defaults to 1.
For security purpose, you are recommended to isolate management packets from
other packets and configure another VLAN as the management VLAN for a cluster.
l
If the port of an access NM device (including
FTP/TFTP server, NM host and log host) does not allow the packets from the
management VLAN to pass, the NM device cannot manage the devices in a cluster
through the management device. In this case, on the management device, you need
to configure the VLAN interface of the access NM device (including FTP/TFTP
server, NM host and log host) as the NM interface.
l
You are recommended to set the timers to their
default values.
l
After you execute the display cluster
current-topology command, the device connecting with a blacklist device
will not be displayed.
