[LONG] Performance of an 877. How to increase the buffer?

First of all I'm sorry for the long post. I ran some tests with ping and you can see the output below.

Router#ping 195.210.91.83 size 1400 repeat 50 Sending 50, 1400-byte ICMP Echos to 195.210.91.83, timeout is 2 seconds: !!.!!.!!.!!.!!.!!.!!.!!.!!.!!.!!.!!.!!.!!.!!.!!.!! Success rate is 68 percent (34/50), round-trip min/avg/max = 64/65/72 ms

Router#ping 195.210.91.83 size 1100 repeat 50 Sending 50, 1100-byte ICMP Echos to 195.210.91.83, timeout is 2 seconds: !.!!.!!.!!.!!.!!.!!.!!.!!.!!.!!..!.!!.!!.!.!!.!!.! Success rate is 64 percent (32/50), round-trip min/avg/max = 56/56/64 ms

Router#ping 195.210.91.83 size 900 repeat 50 Sending 50, 900-byte ICMP Echos to 195.210.91.83, timeout is 2 seconds: !!!.!!!.!!!.!!!.!!!.!!!.!!!.!!!.!!!.!!!.!!!.!!!.!! Success rate is 76 percent (38/50), round-trip min/avg/max = 48/50/56 ms

Router#ping 195.210.91.83 size 700 repeat 50 Sending 50, 700-byte ICMP Echos to 195.210.91.83, timeout is 2 seconds: !!!!.!!!!.!!!!.!!!!.!!!!.!!!!.!!!!.!!!!.!!!!.!!!!. Success rate is 80 percent (40/50), round-trip min/avg/max = 40/41/52 ms

Router#ping 195.210.91.83 size 500 repeat 50 Sending 50, 500-byte ICMP Echos to 195.210.91.83, timeout is 2 seconds: !!!!!!.!!!!!!.!!!!!!.!!!!!!.!!!.!!!.!!!!!!.!!!!!!. Success rate is 84 percent (42/50), round-trip min/avg/max = 32/36/40 ms

Router#ping 195.210.91.83 size 300 repeat 50 Sending 50, 300-byte ICMP Echos to 195.210.91.83, timeout is 2 seconds: !!!!!!!!!!.!!!!!!!!!!.!!!!!!!!!!.!!!!!!!!!!.!!!!!! Success rate is 92 percent (46/50), round-trip min/avg/max = 28/28/36 ms

Router#ping 195.210.91.83 size 150 repeat 50 Sending 50, 150-byte ICMP Echos to 195.210.91.83, timeout is 2 seconds: !!!!!!!!!!!!!!!!!.!!!!!!!!!!!!!!!!!!.!!!!!!!!!!!!! Success rate is 96 percent (48/50), round-trip min/avg/max = 20/24/32 ms

As you can see the less the ICMP size is the less losses the router experiences. Can I work on the buffer on the ATM0.1 interface?

Which could be the option?

I ran the test in the opposite direction and I got no problems with any size of packets, hence I will choose the buffer for the packets leaving the router from ATM0.1 These are the values for the Vlan1 and ATM0.1 interfaces

interface ATM0.1 point-to-point ip address X.X.X.X 255.255.255.0 ip access-group 1 in ip mtu 1460 ip nat outside ip virtual-reassembly ip tcp adjust-mss 1440 pvc 8/35 oam-pvc manage encapsulation aal5snap ! end

interface Vlan1 ip address 10.10.10.254 255.255.255.224 ip access-group 199 in ip inspect ethernetin in ip nat inside ip virtual-reassembly ip tcp adjust-mss 1440 hold-queue 100 out end

I got the same results using MTU=1492 mss=1452 on ATM0.1

The ACL 1 cut off packets with not allowed sources (almost private addresses)

TIA

Alex

Reply to
AM
Loading thread data ...

Alex,

I doubt you have a buffering problem, your pings will be very slow in reality, I mean the interval between each is very long compared to normal traffic, they will not be filling the output queue regardless of their size. Most likely you have an erroring line or interface, and the longer the packet the better the chance it will get corrupted in transmission:

FYI:

formatting link
Check your interface stats for errors as well as queue drops:

Show interface AT0 and ATM0.1

also, show dsl int ATM0

hope that helps!

Phil Dotchon CCIE 6632

Reply to
Phil Dotchon

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.