beautypg.com

Stack failure recovery – Allied Telesis AlliedWare Plus Operating System Version 5.4.4C (x310-26FT,x310-26FP,x310-50FT,x310-50FP) User Manual

Page 2127

background image

VCStack Introduction

Software Reference for x310 Series Switches

C613-50046-01 REV A

AlliedWare Plus

TM

Operating System - Version 5.4.4C

82.13

1.

If a backup member becomes the

Disabled Master

it will first disable all its switch

ports, then activate any triggers specified with the

type stack disabled-master

command that have been configured.

2.

The stack member assumes the role of stack master. In specific situations this
condition could result in a stack containing two masters. This would present problems
with network management and the control of links that were previously aggregated.

3.

The following situation will apply to a switch that has been removed from the stack by
the using the “no stack enable” command:

«

The switch will be unaware of further faults within the stack it was previously a
member of.

«

Should this switch then be powered down, all links previously shutdown (as a
result of running the no stack enable command) will remain down.

4.

If the backup member’s ports are still up, this may cause downstream switches with
trunked ports to operate incorrectly.

Stack Failure Recovery

If the stack master either fails, or is removed, the other stack members will re-elect a new
stack master. The stack members automatically determine which of them becomes the
new stack master. See the

Disabled Master Monitoring (DMM)

and the

Disabled Master

sections for more information. Alternatively, you can manually configure a trigger with the

type stack disabled-master command on page 77.26

to activate on a stack member if it

becomes the disabled master.

Table 82-1

shows how the stack backup members would respond to various problems

occurring on the stack master stack.

Run the no stack enable
command

3

No change

Disabled Master

Disabled Master

Software application failover
(lock-up or continual crashes)

No change

Re-boot as backup
member

Reboot as stack member

software crash or lock-up

No change

Frozen

4

Frozen

4

Power-down / PSU failure

No change

Powered Down

Powered down

Table 82-1: State Change Table

Note

When VCStack is used with EPSR, the EPSR failovertime must be set to at least 5
seconds to avoid any broadcast storms during failover. Broadcast storms may
occur if the switch cannot failover quickly enough before the EPSR failovertime
expires. See the

epsr

command for further information about the EPSR

failovertime.
See the

reboot rolling

command for further information about stack failover.