Stacking ports in 6602 switches

[a.kumar]

We have two edge 6602 switches on the same rack , I want to access both the switches using same ip address, also these switches are connected using stack ports. How can enable both the switch same ip and configuring stack for access both the switches using the same ip address ?

[neo77]

If they are already connected with stacking cables then they will act as a virtual chassis. You can use any one vlan for management and assign an ip for that vlan.

Using that ip you can access the entire stack.

[a.kumar]

Physically it is connected using staking cable, but it is configured to connect as One switch. Could you please help me the steps for COnfiguring Stacking ?

Thanks

Anil

[benny]

Moved the topic to the correct forum...

---

Regarding the stacking it seems that I have to clear up some things.

Basically the Alcatel-Lucent stacking builds a virtual chassis for you. The chassis has eight (8) slots for NIs.
The LED digit on the chassis indicates the NI number. It is not a good idea to run the same number twice.
Also the units should run the same AOS version before you stack them. All units will download the version which is on the primary to have the same timestamp.

As in a physical chassis you need a primary CMM and a secondary CMM. Those are "virtually" attached to the NIs in a stack.

Let's consider the following example:
_____________
| NI | Role |
|-----|--------|
| NI 1 | CMM A |
| NI 2 | CMM B |
| NI 3 | IDLE |
| NI 4 | IDLE |
| NI 5 | IDLE |
|____________|

In the above case unit 1 == NI1 == CMM A. This is the primary CMM and is managing ALL the other units in the stack.

If you want to configure the unit 5 (which is NI 5 and IDLE role) you connect to the IP address of the primary unit (the IP is not actually attached to a unit, but you will understand in a second) and then you configure e.g. vlan 11 for port 5/11 (as on a OS7800).

Understand the redundancy:
- If unit 1 (in this case the primary CMM) fails e.g. power-cut
-> Unit 2 will become the CMM A (primary)
-> Unit 3 will become the CMM B (secondary)
-> When unit 1 comes back after the reboot it will become IDLE

Idle does not mean that it is bad, but it is just an unit which is not actively managed by the user. (The user managed it through the primary).

All units in the stack keep and receive the configuration when the admin is doing "write memory && copy working certified flash-syncro"

The "flash-syncro" is actually the important part here. It tells the primary to send the config to the other units in the stack.

Only when the stack is properly synchronized the redundancy works!

Hope this is clear now.

-benny