ndt-users list created

[Richard Carlson <>]

Peter;

The NDT server usually reports much longer RTT's that tools like 
ping.  This is due to the server pumping data into the NIC faster than the 
NIC can send it out on the wire.  This results in a standing output queue 
on the server.  TCP measures the RTT from the time it passes a packet down 
to the IP layer until it receives an ACK for that packet (assuming that no 
loss occurred in that time window).  So you are seeing 48 msec of buffer or 
~600 KB of data sitting in output buffers waiting to get onto the wire.

Rich

At 05:28 PM 8/24/2006, Peter Van Epp wrote:
>         We are seeing a long rtt reported on 5.4.8. The rtt shouldn't be
> anywhere near 49 Msec, it should be closer to 1.
>
> ping -c 100 -s 1452 sniffer1.ucs.sfu.ca
> PING sniffer1.ucs.sfu.ca (142.58.200.253): 1452 data bytes
> 1460 bytes from 142.58.200.253: icmp_seq=0 ttl=62 time=1.057 ms
> 1460 bytes from 142.58.200.253: icmp_seq=1 ttl=62 time=0.999 ms
> 1460 bytes from 142.58.200.253: icmp_seq=2 ttl=62 time=1.008 ms
> ...
> --- sniffer1.ucs.sfu.ca ping statistics ---
> 100 packets transmitted, 100 packets received, 0% packet loss
> round-trip min/avg/max = 0.976/1.417/34.343 ms
>
>
> WEB100 Enabled Statistics:
> Checking for Middleboxes . . . . . . . . . . . . . . . . . .  Done
> checking for firewalls . . . . . . . . . . . . . . . . . . .  Done
> running 10s outbound test (client-to-server [C2S]) . . . . . 90.25Mb/s
> running 10s inbound test (server-to-client [S2C]) . . . . . . 87.67Mb/s
>
>         ------  Client System Details  ------
> OS data: Name = Mac OS X, Architecture = ppc, Version = 10.3.9
> Java data: Vendor = Apple Computer, Inc., Version = 1.4.2_09
>
>         ------  Web100 Detailed Analysis  ------
> 100 Mbps FastEthernet link found.
> Link set to Full Duplex mode
> No network congestion discovered.
> Good network cable(s) found
> Normal duplex operation found.
>
> Web100 reports the Round trip time = 49.48 msec; the Packet size = 1460 
> Bytes; and
> No packet loss - but packets arrived out-of-order 0.08% of the time
> C2S throughput test: Packet queuing detected: 0.19%
> S2C throughput test: Packet queuing detected: 8.21%
> This connection is sender limited 100.0% of the time.
>
> Web100 reports TCP negotiated the optional Performance Settings to:
> RFC 2018 Selective Acknowledgment: OFF
> RFC 896 Nagle Algorithm: OFF
> RFC 3168 Explicit Congestion Notification: OFF
> RFC 1323 Time Stamping: OFF
> RFC 1323 Window Scaling: OFF
>
> Server 'sniffer1.ucs.sfu.ca' is not behind a firewall. [Connection to the 
> ephemeral port was successful]
> Client is probably behind a firewall. [Connection to the ephemeral port 
> failed]
> Information: Network Middlebox is modifying MSS variable
> Server IP addresses are preserved End-to-End
> Client IP addresses are preserved End-to-End
>
> Peter Van Epp / Operations and Technical Support
> Simon Fraser University, Burnaby, B.C. Canada

------------------------------------



Richard A. Carlson                              e-mail: 

Network Engineer                                phone:  (734) 352-7043
Internet2                                       fax:    (734) 913-4255
1000 Oakbrook Dr; Suite 300
Ann Arbor,  MI  48104