Stacking Avaya ERS 3500 using Rear-Ports

Task

Create the stack of two Avaya ERS 3500 switches using Rear Ports.


Scenario

There is a Avaya ERS 3524GT switch working as Stand-Alone device. The customer need to connect the 2nd switch with 24 PoE ports and use both devices as single stack. The existing switch will be MASTER (BASE) unit in new scenario and the new 3524GT-PWR+ just the SLAVE (NON-BASE) unit. For stacking I will use the HP Direct Attach 1m J9281B cables (x2). I tried it with Avaya 10GBaseCX SFP+ 3.0M Active AA1403019-E6 cables (x2), too. It works with both types without any problems.


Hardware | Software

  • Avaya ERS 3524GT with BOSS v5.3.1.003 (x1)
  • Avaya ERS 3524GT-PWR+ with BOSS v5.3.1.003 (x1)
  • HP Direct Attach 1m J9281B cables (x2)
  • Avaya 10GBaseCX SFP+ 3.0M Active AA1403019-E6 (x2)

Solution | Overview

Step 1: Change the Rear Port mode.

Step 2: Prepare the stack BASE / MASTER unit.

Step 3: Prepare the NON-BASE / SLAVE unit.

Step 4: Stack both switches using SFP+ cables.

Step 5: Verify the stack.


Solution | Detailed

Step 1: Change the Rear Port mode.

Change it from Standalone to Stacking in global configuration on both switches. Save the config and restart both devices.

rear-ports mode stacking
show rear-ports mode
********************

Current rear-ports mode: Standalone Mode
Next rear-ports mode: Stacking Mode
Next mode will be applied after reset

———

Step 2: Prepare the stack BASE / MASTER unit.

a) Configure the stack parameters in global configuration.

stack-monitor trap-interval 30
stack-monitor stack-size 2
stack-monitor enable
stack reboot-on-failure 
stack forced-mode
stack auto-unit-replacement enable
stack auto-unit-replacement-image enable
  • stack-monitor trap-intervalSets the trap interval value, default is 60.
  • stack-monitor stack-sizeSets the stack size to be monitored. Values range from 2 to 8, default is 2.
  • stack-monitorConfigures the monitoring of stack switch units to detect problems within the stack.
  • stack reboot-on-failureEnables or disables rebooting of stack units when stack communication fails.
  • stack forced-modeEnables or disables Stack Forced Mode for switch units that are part of a stack.
  • stack auto-unit-replacementEnables or disables Auto Unit Replacement (AUR) for switch units that are part of a stack.
  • stack auto-unit-replacement-imageEnables or disables Agent Auto Unit Replacement (AAUR) for switch units that are part of a stack.

b) Switch right the BASE button on stack master’s rear panel to BASE position.

c) Remove switch & stack ip addresses, configure and verify new stack and switch ip address.

no ip address switch
no ip address stack 
ip address stack 10.0.0.1
ip address switch 10.0.0.2
ip address netmask 255.0.0.0
show ip address
***************

                    Configured        In Use Last       BootP/DHCP
                    ---------------   ---------------   --------------------
Stack IP Address:   10.0.0.1                            0.0.0.0
Switch IP Address:  10.0.0.2          10.0.0.2          0.0.0.0
Stack Subnet Mask:  255.0.0.0         255.0.0.0         0.0.0.0

d) Verify the version of BOSS and Diagnostic Tools.

sh boot
*******

Unit    Agent Image   Active Image   Diag Image   Active Diag
-----   -----------   ------------   ----------   -----------
1       5.3.1.003     5.3.1.003      5.3.0.6      5.3.0.6

* - Unit requires reboot for new Active Image to be made operational.
# - Unit requires reboot for new Diag to be made operational.

———

Step 3: Prepare the NON-BASE / SLAVE unit.

a) Configure the stack parameters in global configuration.

stack-monitor trap-interval 30
stack-monitor stack-size 2
stack-monitor enable
stack reboot-on-failure 
stack forced-mode
  • stack-monitor trap-intervalSets the trap interval value, default is 60.
  • stack-monitor stack-sizeSets the stack size to be monitored. Values range from 2 to 8, default is 2.
  • stack-monitorConfigures the monitoring of stack switch units to detect problems within the stack.
  • stack reboot-on-failureEnables or disables rebooting of stack units when stack communication fails.
  • stack forced-modeEnables or disables Stack Forced Mode for switch units that are part of a stack.

b) Leave the BASE button on stack slave’s rear panel. Do not switch it.

c) Remove switch & stack ip addresses, configure and verify new stack and switch ip address.

no ip address stack
no ip address switch 
ip address stack 10.0.0.1
ip address switch 10.0.0.3
ip address netmask 255.0.0.0
show ip address
***************

                    Configured        In Use Last       BootP/DHCP
                    ---------------   ---------------   --------------------
Stack IP Address:   10.0.0.1                            0.0.0.0
Switch IP Address:  10.0.0.3          10.0.0.3          0.0.0.0
Stack Subnet Mask:  255.0.0.0         255.0.0.0         0.0.0.0

d) Check the version of BOSS and Diagnostic Tools.

If it’s different then image on master unit, you can make optional an upgrade/downgrade. If you don’t want to do this, slave unit should make it alone during stacking process, without user interaction. It takes a couple of minutes and requires 2 reboot of slave unit. During this process all LEDs are blinking. Be patient.

sh boot
*******

Unit    Agent Image   Active Image   Diag Image   Active Diag
-----   -----------   ------------   ----------   -----------
1       5.3.1.003     5.3.1.003      5.3.0.6      5.3.0.6

* - Unit requires reboot for new Active Image to be made operational.
# - Unit requires reboot for new Diag to be made operational.

———

Step 4: Stack both switches using SFP+ cables.

a) Power off the slave unit.

b) Connect both cables between stack members using Cascade Down configuration.

c) Power on the slave unit and wait for stacking.

———

Step 5: Verify the stack.

show stack-info
***************

Unit#   Switch Model   Pluggable   Pluggable   Pluggable   Pluggable   SW Version
                       Port        Port        Port        Port
-----   ------------   ---------   ---------   ---------   ---------   ----------
1       3524GT         (21) None   (22) None   (23) None   (24) None   v5.3.1.003
2       3524GT-PWR+    (21) None   (22) None   (23) None   (24) None   v5.3.1.003
show stack health
*****************
----------------------------------------------------------
UNIT#      Switch Model   Cascade Up   Cascade Down
----------------------------------------------------------
1 (Base)   3524GT         OK           OK
2          3524GT-PWR+    OK           OK
----------------------------------------------------------

Switch Units Found = 2 
Stack Health Check = OK - RESILIENT
Stack Diagnosis = Stack in full resilient mode.
show stack-monitor
******************

Status: enabled
Stack size: 2
Trap interval: 30
show stack forced-mode
**********************

Forced-Stack Mode: Enabled
Device is not currently running in forced stack mode.