Baystack 470pwr stack - Is this normal?

[Malcybood]

Hi,

We re-allocated a stack of 3 baystack 470pwr switches from a site that was closing down and just changed the IP address and default gateway etc to match the new site.

We have a data, voice and mgmt VLAN configured on the switch stack and when we plugged the switch in I was unable to ping the Baystack's management IP address until I had pinged the router sub-interface.

During the time I couldn't ping the switch mgmt address, I could ping devices on the voice and data VLAN without any issues so I knew dot1q trunking was configured correctly.

When I pinged the management VLAN default g/w on the router this appeared to "wake up" the LAN switch and I could then ping the switch remotely.

Is this normally required to ping the switch default gateway on the router before being able to ping the switch?

Also I couldn't ping the default gateway from the switch initially.

Another question slightly off topic when I go to renumber stacks the stack is still seeing 3 switches. If I select renumber the stacks will this resolve itself and does it cause any down time to users doing this?

Cheers

 

[DaddyOfThree]

It is not normal operation... you should not have to ping the default gateway from the switch. What is acting as the default gateway? The ES470 is only a layer 2 switch so you need something else to route between the VLANs.

If you wish to renumber the stack you will need to reboot the entire stack to complete the renumbering process.

Cheers!

 

[Malcybood]

Hi DaddyOfThree,

Yeah I probably should have put that in.

We have dual separate WAN links with WAN routers (Cisco 2800's) which perform doy1q trunking.

The voice and data sub interfaces on the router, IP addresses are .2 primary router, .3 backup router and we use .1 for the HSRP running between the 2 routers and acts default gateway address.

I guess the PC and IP phones generate DHCP requests (IP helper used so they hit the default gateway) which wakes up the sub interfaces on the routers when generating IP requests, but the L2 switch which is on a completely different subnet doesn't generate any IP traffic requests.

Not quite sure, it's a strange one I think it is probably to do with using sub-if's and HSRP between the 2 cisco routers and as mentioned the 470 being L2 switch. I think when I've used sub-ifs on Cisco routers and L2 cisco switches I had to ping the default gw before I could telnet to the mgmt address.

The same if your Cisco switches are connected but not stacked so if you had 2 switches newly installed with separate mgmt addresses "switch a" with IP 10.1.1.10 and "switch b" 10.1.1.11 I know that you need to ping 10.1.1.10 before you can ping 10.1.1.11 and the first few replies time out.

Ah well one to keep in the scrapbook for future reference!