Cannot reach the gateway on the switch (3750X)

[hereim]

Hi there,


I have a stack "A" of 2 3750X newly installed having data vlan and a voice vlan. Just for temporary usage, we had connected a 3750 standalone switch "B" on the port G1/0/27 of the stack.

B was programmed to be an access port carrying just the Data vlan. Switch A's port was also prgrammed the same.


But whats happening is that local users on the stack cannot ping their own data vlan gateway which locally on the stack switch. Whereas the users in the same vlan but on the standaone switch can reach this gateway.



Ports Config on the stack:

Switch_A#sh run | begin GigabitEthernet1/0/25

interface GigabitEthernet1/0/25

description ***** Standard Interface configuration to Client PC & Phone *****

switchport access vlan 240

switchport trunk encapsulation dot1q

switchport trunk native vlan 240

switchport mode access

switchport voice vlan 651

no logging event link-status

speed 1000

duplex full

srr-queue bandwidth share 10 10 60 20

srr-queue bandwidth shape 10 0 0 0

queue-set 2

priority-queue out

no snmp trap link-status

mls qos trust cos

auto qos voip trust

no mdix auto

flowcontrol receive on

spanning-tree portfast

service-policy input PM-SET-DSCP

!

interface GigabitEthernet1/0/26

description ***** Standard Interface configuration to Client PC & Phone *****

switchport access vlan 240

switchport trunk encapsulation dot1q

switchport trunk native vlan 240

switchport mode access

switchport voice vlan 651

no logging event link-status

srr-queue bandwidth share 10 10 60 20

srr-queue bandwidth shape 10 0 0 0

queue-set 2

priority-queue out

no snmp trap link-status

mls qos trust cos

auto qos voip trust

no mdix auto

flowcontrol receive on

spanning-tree portfast

service-policy input PM-SET-DSCP

!

interface GigabitEthernet1/0/27

description ***** Temporary Uplink*****

switchport access vlan 240

switchport trunk encapsulation dot1q

switchport trunk native vlan 240

switchport mode access

switchport voice vlan 651

no logging event link-status

srr-queue bandwidth share 10 10 60 20

srr-queue bandwidth shape 10 0 0 0

queue-set 2

priority-queue out

no snmp trap link-status

mls qos trust cos

auto qos voip trust

no mdix auto

flowcontrol receive on

spanning-tree portfast

service-policy input PM-SET-DSCP


I tried doing debug on arp frames and could see the following:

000612: Aug 23 15:07:34.809 WST: IPARP_TRACK: ARP tableid(0) 10.42.64.43 Vlan240 mode(1) modified.

000613: Aug 23 15:07:46.771 WST: IPARP_TRACK: ARP tableid(0) 10.42.64.43 Vlan240 mode(1) modified.

000614: Aug 23 15:07:49.799 WST: IPARP_TRACK: ARP tableid(0) 10.42.64.43 Vlan240 mode(1) modified.

000615: Aug 23 15:07:55.830 WST: IPARP_TRACK: ARP tableid(0) 10.42.64.43 Vlan240 mode(1) modified.

000616: Aug 23 15:08:03.120 WST: IPARP_TRACK: ARP tableid(0) 10.42.64.43 Vlan240 mode(1) modified.

000617: Aug 23 15:08:04.328 WST: IPARP_TRACK: ARP tableid(0) 10.42.64.43 Vlan240 mode(1) modified.

000618: Aug 23 15:08:06.089 WST: IPARP_TRACK: ARP tableid(0) 10.42.64.43 Vlan240 mode(1) modified.

000619: Aug 23 15:08:06.291 WST: IPARP_TRACK: ARP tableid(0) 10.42.64.43 Vlan240 mode(1) modified.

000620: Aug 23 15:08:07.314 WST: IPARP_TRACK: ARP tableid(0) 10.42.64.43 Vlan240 mode(1) modified.


Any suggestions?

 

[VinceWhirlwind]

What a horrible mess this is:

switchport access vlan 240
switchport trunk encapsulation dot1q
switchport trunk native vlan 240
switchport mode access
switchport voice vlan 651

That is *not* the way to configure a switchport.

It should be either:

switchport mode access
switchport access vlan 240
switchport voice vlan 651

OR

switchport mode trunk
switchport trunk encapsulation dot1q
switchport trunk native vlan 240
switchport trunk vlan allow 240,651

 

[hereim]

Hi VinceWhirlwind

I appreciate your time in this. And I also understand that we need not have trunk encapsulation put in if we are using it as an access port.However, it does not apply it is technically not going to work.

The fact is that we are telling the switch the encap should be dot1Q whenever the port goes into trunking mode.

I am sure thats not the cause of the issue we are having.But as i said earlier, thanks for your time looking at it. Thats more important that we excahnge ideas amongst each other and fix problems.

I shall keep you posted how i go with my troubleshooting.

Cheers!

 

[brianinms]

I agree with Vince, that configuration is a mess. Instead of troubleshooting your issue you should just change it to a standard configuration.

 

[VinceWhirlwind]

It's much easier to troubleshoot a clear, readable, sensible config. I agree with you - I *think* I know how your switchports will behave with the way they are configured. However, if you're having problems, then you need to address the obvious misconfigurations first.

First, fix up all the "edge" ports.

Then, think about the uplink ports - you can configure your uplink to Switch B as either just an access port in VLAN240, or make it a dot1q trunk.

Then, where is the router/gateway address for VLAN 240?
Is it the VLAN 240 interface address, or is it a separate device patched to the switch?
If it's a separate device, then configure its switchport properly - it's either an access port with just VLAN240, or it might be a dot1q trunk with both VLANs.

 

[nelsonw]

run the following command:

sho cdp neighbour
this will give you all the cdp neighbours that are connected to the swithch. If you tag "detail" to the end of it, it will also give you information such as neighbours ip addresses and vlans.
check to make sure that you dont have a vilan mismatch.
To fix vlan mismatches, type on your interface (both switches):
switchport trunk native vlan 240

 

[hereim]

Hi guys,

I have fixed the issue.Technically all the configured things were correct. But yes, i agree its not that easy to read thru. but no way that was causing the issue.

the thing was it was one application on that user PC, which whenever starting, used to block all the network connectivity. after all the troubleshooting, it just came down to port speed issue. If we ran the application on the default settings which was 1G, it gave us issues. The moment we shrunk it to 100M, everything started working normally.

I hope that helps all of us in future, just in case you see that type of symptom.