[lkml]   [1998]   [May]   [24]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: Poor TCP throughput in 2.1.103
Content-Type: text/plain; charset=US-ASCII

>>>>> "Shaw" == Shaw Carruthers <> writes:

Shaw> The problem is suspected to lie with my ISP's Ascend servers,
Shaw> using larger ( > 12k ) windows causing them to stall the TCP
Shaw> pipeline. So many other people could have the same problem as
Shaw> Ascends are very popular.

This rings a bell with me. When I used to connect to an Ascend server,
I found that when my PPP link was running at capacity the connection
would "stall" after about 10 seconds. It used to drive me nuts... at
the time I assumed it was a buffer problem with the ascend. Ie, when
the ascend's buffer was exceeded, packets were dropped and I stopped
ACKing. I then assumed the stall was the pause waiting for the sending
host to retransmit. This explanation is probably wrong, but I didn't
know enough about the mechanics of TCP/IP to formulate a proper
explanation and haven't thought about it since switching ISP. :)

I found this problem with every 2.1.x kernel I tried, from about
2.1.4x (just before the dud kernel release that trashed Linus' HD?)
until the early 2.1.9x kernels (at which point I went back to 2.0).

- Andrew Snare
#!/usr/bin/env python
print(lambda s:s+"("+`s`+")")\
('#!/usr/bin/env python\012print(lambda s:s+"("+`s`+")")\\\012')
print(lambda x:x%`x`)('print(lambda x:x%%`x`)(%s)')
[unhandled content-type:application/pgp-signature]
 \ /
  Last update: 2005-03-22 13:42    [W:0.054 / U:8.040 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site