You will have to post the whole config of the etherswitch module so we can look for a definitive answer. The way it is now they have separate default gateways so you need a couple of statics pointing back and forth across the channel if you are not using a routing protocol or just turn on something like ospf and it will probably fix it . The way it is now pca sends to 192.168.1.1 , he does not have a route listing for the pc's on the other side so he doesn't know what to do with them . PCA-pcd works because it is talking to him via layer 2 and does not have to be routed to talk, same for pcb and pcc . Add 2 statics on each side.I don't know what you used for subnet masks so i just used a /24 as an example. The other option is to get rid of the layer 3 SVI's on "one" side set the pc nic cards default gateway to point to the layer 3 SVI address on the other side . Not a great design either way.
From both your diagram and orginal post that notes:
work fine, I would expect
PCA-PCC and PCB-PCD to also work. As long as IP routing in enabled between the VLANs on both 3640s, shouldn't matter which one you use as a gateway. I.e. each router should forward packet from the one VLAN to the other.
Sorry for my late response and thank you all for your valuable input.
I want you to know, this is not a production network. This is a lab environment in which I simulate the inter-vlan routing between trunked switches"
PCA-PCB is successful, because ip routing is enabled in switch1(R0) and it has directly connected routes via SVIs. When I sh ip route, I see the connected routes. switch1(R0) SVIs take care of that inter-vlan routing. Handled in L3
PCA-PCD is successfull, because VLAN information is carried via trunk Etherchannel, no routing required since they are in same VLAN. Handled in L2
Here it comes. PCA-PCC results in "Request timed out" not "Destination unreachable". Because switch1(R0) has a directly connected route for that segment, but given IP is not found. Because it is in the segment which is connected to other switch(R1). So traffic should be routed to other switch. Here are options
Etherchannel is in trunking mode. If I assign IP address to po1 interfaces, I would have to remowe the switchport capability to switch the port to routed port, thus trunk will be down. Ok lets say that his is the best practice, and I formed another etherchannel for trunking.
A routing protocol's route wont appear in routing table since switches R0 and R1 has a route with better admin distance, which is connected route. As glen stated, static routes can be added, but would switch prefer a static route while it has a connected route to that segment? And how efficent would that be? DHCP may be enabled , or thousands of hosts exist. Cisco says that use 1 subnet per VLAN and this is the best practise.
So according to above information, what is the best practice?
"This is a lab environment in which I simulate the inter-vlan routing between trunked switches"
When you write simulate, are you using real equipment and have you actually tried this for real?
Although you write "Here it comes. PCA-PCC results in "Request timed out" not "Destination unreachable". Because switch1(R0) has a directly connected route for that segment, but given IP is not found. Because it is in the segment which is connected to other switch(R1). So traffic should be routed to other switch.", doesn't make sense since routing between R0 and R1 shouldn't be necessary. Logically you should have what my attachment shows. If both VLANs are properly trunked, all hosts on each VLANs should be visible to each router. I.e. shouldn't matter which router they are connected to.
Your 2 options are don't route those vlans on "1" of the 3640's and just let it ride the etherchannel trunk to the other 3640 which would then do the routing for the whole setup and adjust your nic default gateways accordingly. Otherwise make separate subnets on each side and route between the 3640's either via routing protocol or statics.
Question We run asr9001 with XR 6.1.3, and we have a very long delay to
login w/ SSH 1 or 2 to the device compare to IOS device. After
investigation, the there is 1s delay between the client KEXDH_INIT and
the server (XR) KEXDH_REPLY. After debug ssh serv...
Introduction The purpose of this document is to demonstrate the Open
Shortest Path First (OSPF) behavior when the V-bit (Virtual-link bit) is
present in a non-backbone area. The V-bit is signaled in Type-1 LSA only
if the router is the endpoint of one or ...
Hi, I am seeing quite a few issues with patch install and wanted to
share my experience and workaround to this. Login to admin via CLI, then
access root with the “shell” command Issue “df –h” and you’ll probably
see the following directory full or nearly ...