Stacking + DHCP

[antliu620]

Dear all,

We are considering to use stacking feature but would like to make sure about the behavior before we go for it.

The question is, for example, when we have 6 Cisco switches stacked and DHCP client is used on the stacked switches.

What is the MAC address used when requesting IP Address from DHCP Server? How will the stack get new/renew IP adress if the master is down or another switch is elected/configured as master? Intuitively, it seems like the MAC address will change in this scenario, but it sounds like a nightmare to administrators... Does anyone ever have experience similar to this?

Any info is appreciated! Thanks in advance.

best,

anthony

[joelgooding]

Hello,

For stacked swtiches, since logically it is one switching unit rather than separate swithes, the mac address used when requesting IP's will be the same across the stack. If the master switch goes down, a new master is elected and the stack continues minus the failed switch. Remember in dhcp, when a device requests an ip, it sends out the broadcast for the dhcp server via layer 3. this reply comes back to the device with the information which the switchign stack may or may not have in its layer 2 tables. If not, an arp request is sent out to match the corresponding mac address. So its better to have these switches stacked as they perform the exact same way as if they were one big switch. I hope this answers your question.

Joel

_______________________________
Please rate helpful posts and answered questions!

[antliu620]

Hi Joel,

Thanks for you info. I can understand the stack is working as a whole and the MAC address used will be the same across the stack.

I am wondering what is this *same* MAC address? Is it a pseudo randomly generated MAC address? If it is the case, collision may occurr. If not, will it be the MAC of one of the switches in the stack?

Again, thanks for any input in advance!

best,

anthony

[Joseph W. Doherty]

Disclaimer

The  Author of this posting offers the information contained within this  posting without consideration and with the reader's understanding that  there's no implied or expressed suitability or fitness for any purpose.  Information provided is for informational purposes only and should not  be construed as rendering professional advice of any kind. Usage of this  posting's information is solely at reader's own risk.

Liability Disclaimer

In  no event shall Author be liable for any damages whatsoever (including,  without limitation, damages for loss of use, data or profit) arising out  of the use or inability to use the posting's information even if Author  has been advised of the possibility of such damage.

Posting

The MAC belongs to the host, so if you're talking about hosts connected to the stack, their MAC shouldn't change with stack master failure.

However, not related to DHCP, the stack master, by default, provides the MAC of any stack hosted IP (like for virtual interfaces).  If the stack master fails, by default, the new stack master uses a different MAC for stack IPs (not for host IPs).  The new stack master will issue a gratuitous ARP so clients will refresh their ARP cache.  If a client doesn't refresh it's ARP cache, it will be unable to send traffic to a "known" stack IP (again not a same subnet other host IP) until the ARP entry expires.  Solutions to this issue include using the command for persistent stack MACs or using HSRP for gateway IPs (which use the same virtual MAC).