HP3000-L Archives

December 2002, Week 3

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
James Hofmeister <[log in to unmask]>
Reply To:
Date:
Thu, 19 Dec 2002 21:11:21 GMT
Content-Type:
text/plain
Parts/Attachments:
text/plain (97 lines)
Hello James, et all,

RE: FTP Question #2.


1) Run nettool to see if you have any network resource constraints:

:nettool.net.sys;info="res;di;quit"

looking for any :(


2) Run nettool to see if you have any TCP level retransmissions,
this should be run before and after a transfer to see any change.

:nettool.net.sys;info="status;tcpstat;tcpglobal @;quit"
...
    . Total re-transmitted segments :       38  <<<<<=====


3) Check your TCP Timers:

:nmmgr - [open config], [ns], [unguided config], [goto netxport],
         [goto gprot], [goto tcp]:

------------------------------------------------------------
Path:  NETXPORT.GPROT.TCP

[1  ]     Retransmission Interval Lower Bound (Secs)
[360  ]   Maximum Time to Wait For Remote Response (Sec)
[2  ]     Initial Retransmission Interval (Secs)
[8  ]     Maximum Retransmissions per Packet
------------------------------------------------------------

I am currently running over a IFITL/DSL link which has serious bouts
of packet loss (the technical term is "sucking-wind"), so as a
temporary workaround I have increased my "Maximum Retransmissions per
Packet" above the [5] I usually run. Of course I need to get my act
together and pursue this link problem with my ISP.


4) Use linkcontrol to see if you have any LINK level problems, this
should be run before and after a transfer to see any change.

:linkcontrol @;status=all
...

Transmit bytes             2290582198
Transmits                     1619411
Transmits no error            1619411
Transmits dropped                   0
Transmits deferred            1137287
Transmits 1 retry              343428
Transmits >1 retry               9043

All of the transmit numbers should be evaluated as a percentage of
"Transmit bytes".

Receive bytes                49276954
Recv CRC error                      0
Recv Maxsize error                  0
Recv dropped: buffer                0
Recv dropped: descr                 0
Recv dropped: other                 0
Recv watchdg timeout                0
Recv collisions                     0
Recv overruns                       0

Note: ":linkcontrol <linkname>;status=reset" can be used to clear the
statistics prior to running your test.

Your problem is likely a link problem "packet loss" which is possibly
seen as huge delays due to a large value configured for the TCP
"Retransmission Interval Lower Bounds".  I recommend a value of [1]
for the "Lower Bound" which may improve your performance over poor
quality significantly, but of course resolving the link problem (root
cause) is recommended.

I hope this helps.

Regards,

James Hofmeister
Hewlett Packard - Global Solutions Engineering (WTEC)
P.S. My Ideals are my own, not necessarily my employers.




________________________________________________________________
Sign Up for Juno Platinum Internet Access Today
Only $9.95 per month!
Visit www.juno.com

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

ATOM RSS1 RSS2