|
Home > Archive > CCNP > January 2002 > Packet Switching Question.
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Packet Switching Question.
|
|
| Captain_Miller 2002-01-22, 10:08 am |
| While reading Lammle's CCNP Routing Study Guide I came across a section packet switching section that stated when a router reads a packet and recognizes the logical address it discards the frame and puts the packet in the exit port's buffer. Seems logical. In the same section it stated that if a router recognizes the hardware address it discards the frame and forwards the packet to Main memory.
Why does this happen? Is this because the destination may be on the same segment from which it was recieved?
Also, a bit of insight into this would be quite helpful. I'm assuming the router looks at the hardware address first and then the logical. Could someone explain this topic in a bit of depth. Any help will be much appreciated.
Best Regards,
Miller | |
|
| Once router receive any packet, it will determine the layer2 address if frame destination layer2 address is router's. It so, then this frame is destined to router. If not, it will discard the frame.
Then, router will remove layer2 header and determine layer3 destination address. For process switch, router will consult its routing table. If there is a match in routing entry, router will know the destination interface to send this packet out. If there is no match, the packet will be dropped and ICMP unreachable will be sent back to the source of the packet.
After router know the outgoing interface, the packet will be sent to interface buffer, waiting for encapsulation with layer2 address according to data-link type of the interface. Once the encapsulation is completed, the frame will be sent through the physical to the next hop.
I would not mention about NAT, access-list, etc since I think your question focus about routing issue.
In case of other switching method e.g. fast switching, silicon switching, CEF switching. The first packet of the flow will follow the same way as process switch. But router will cache the destination address and interface. Router will not consult routing table for next packet in the same flow but this cache. This will increase performace of switching of router.
Hope the help, | |
| Captain_Miller 2002-01-23, 11:29 am |
| Thanx for the explanation! Cleared everything up  |
|
|
|
|