RE: Why are local serial pings always double the time of remote s erial pings?

From: Jason Sinclair (sinclairj@xxxxxxxxxxxxxxx)
Date: Sun May 12 2002 - 19:36:13 GMT-3


   
Michael,

OK - what happens in Cisco land is that the router when pinging its own
serial interface will send the ICMP packet to the remote side. This is why
you saw the first redirect. Each subsequent packet is then received at the
remote side, but due to route cache you do not see further redirects. If you
turn off ip route-cache you will see each ICMP packet redirected.

In a nutshell, your assumption is correct; the packet is sent to the remote
router and then returned, hence the double delay.

Cheers,

Jason Sinclair CCIE #9100
Manager, Network Control Centre
POWERTEL
Ground Level, 55 Clarence Street,
SYDNEY NSW 2000
AUSTRALIA
office: + 61 2 8264 3820
mobile: + 61 416 105 858
* sinclairj@powertel.com.au

                -----Original Message-----
                From: Michael Snyder [mailto:msnyder@ldd.net]
                Sent: Monday, 13 May 2002 06:44
                To: ccielab@groupstudy.com
                Subject: Why are local serial pings always double the
time of remote serial pings?

                I've noticed something. Once in a while I ping the local
(near)
                interface ip address by mistake when I really want to ping
the remote
                (far end) ip address of the serial link. Anyway while
Ethernet seems to
                have the same ping times, near or far, Serial interfaces
always seem to
                have double the ping time.
                
                I know from frame-relay that you have to map a local ip
address to a
                interface dlci in order to ping it. So when you ping an ip
of a
                frame-relay interface local to the router, your traffic is
actually
                traveling to another router, and coming back.
                
                
                Here's the rub, assuming that when I ping a local serial ip
address the
                traffic is traveling to the far end of the link and
returning, shouldn't
                a local ping have the same round trip time of a far end
ping?
                
                Why is it double? What's up with that? Surely it's not two
round trips
                for a local ping.
                
                BTW, I did some debugging on the far end of the serial link
(router b).
                I'm showing a single icmp type 8 redirect message, on my
first local
                ping, but that's it. I assume router A is caching the
redirect, but the
                ping times remain about double even after the first ping
sequence, and
                those interface lights on the far end router (b) light up
every time I
                do a local interface ping on router (a) .
                
                
                A#s
                Interface IP-Address OK? Method Status
                Protocol
                
                Serial0 10.1.1.1 YES manual up
                up
                
                A#
                A#ping 10.1.1.1
                
                Type escape sequence to abort.
                Sending 5, 100-byte ICMP Echos to 10.1.1.1, timeout is 2
seconds:
                !!!!!
                Success rate is 100 percent (5/5), round-trip min/avg/max =
68/70/80 ms
                
                A#ping 10.1.1.2
                
                Type escape sequence to abort.
                Sending 5, 100-byte ICMP Echos to 10.1.1.2, timeout is 2
seconds:
                !!!!!
                Success rate is 100 percent (5/5), round-trip min/avg/max =
32/35/36 ms
                A#
        



This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:58:55 GMT-3