public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Kasper Sandberg <lkml@metanurb.dk>
To: Maciej Soltysiak <solt@dns.toxicfilms.tv>
Cc: LKML Mailinglist <linux-kernel@vger.kernel.org>
Subject: Re: tcp_window_scaling degrades performance
Date: Thu, 15 Jul 2004 17:21:51 +0200	[thread overview]
Message-ID: <1089904911.15291.4.camel@localhost> (raw)
In-Reply-To: <1956423367.20040715170937@dns.toxicfilms.tv>

hey, are you having some kind of hardware router?
bert hubert, helped me, and we found out my router was causing me the
problems, however, my problems were more severe, as my limit was 50kb/s

On Thu, 2004-07-15 at 17:09 +0200, Maciej Soltysiak wrote:
> Hi
> 
> I have been experiencied weird problems with network throughput
> lately and I after experimenting with /proc/sys/net/ipv4 knobs
> I found that when I have tcp_window_scaling 0 I can
> get throughput from a distant server of about 600kB/s (well, 200kB/s
> is fast enough)
> 
>    # links ftp://ftp.task.gda.pl/ls-lR
> 
> When I turn tcp_window_scaling to 1 the transfer starts with 600kB,
> then drops nicely to 20kB/s then starts growing again and reaches
> about 400kB/s
> 
> Anyway It works, but there are some hosts, with which I can't talk to
> faster than 800B/s !
> 
> This is a Debian SID with vanilla-2.6.7-bk20, but I am having these
> problems for a few weeks now.
> 
> It might be a TCP kernel issue, but I know that between me and the
> other hosts is a checkpoint fw-1 fp3, it may be the cause of problems,
> because, well, checkpoint is crazy.
> 
> Anyway, I can provide tcpdump, and tcptrace analysis of it and the
> state of sysctls if anyone would be willing to help me figure out
> what's going on.
> 
> Regards,
> Maciej
> 
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
> 


  reply	other threads:[~2004-07-15 15:21 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-07-15 15:09 tcp_window_scaling degrades performance Maciej Soltysiak
2004-07-15 15:21 ` Kasper Sandberg [this message]
2004-07-15 15:25   ` Re[2]: " Maciej Soltysiak
2004-07-15 16:20     ` Kasper Sandberg
     [not found] <2igbK-82L-13@gated-at.bofh.it>
2004-07-15 18:00 ` Andi Kleen

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1089904911.15291.4.camel@localhost \
    --to=lkml@metanurb.dk \
    --cc=linux-kernel@vger.kernel.org \
    --cc=solt@dns.toxicfilms.tv \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox