Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 26 Feb 2001 13:39:37 -0800
From:      Mark Peek <mark@whistle.com>
To:        Jonathan Lemon <jlemon@flugsvamp.com>, Paul Herman <pherman@frenchfries.net>
Cc:        Jonathan Lemon <jlemon@flugsvamp.com>, Garrett Wollman <wollman@khavrinen.lcs.mit.edu>, net@FreeBSD.ORG
Subject:   Re: I have delayed ACK problems
Message-ID:  <p0510010ab6c07a3f770d@[10.1.10.113]>
In-Reply-To: <20010225085626.V5714@prism.flugsvamp.com>
References:  <20010224142742.T5714@prism.flugsvamp.com> <Pine.BSF.4.32.0102250221250.25989-100000@husten.security.at12.de> <20010225085626.V5714@prism.flugsvamp.com>

next in thread | previous in thread | raw e-mail | index | archive | help
At 8:56 AM -0600 2/25/01, Jonathan Lemon wrote:
>On Sun, Feb 25, 2001 at 11:10:54AM +0100, Paul Herman wrote:
>>  On Sat, 24 Feb 2001, Jonathan Lemon wrote:
>>
>>  > On Sat, Feb 24, 2001 at 11:19:02AM -0800, Mark Peek wrote:
>>  > > Was there ever a final resolution to this problem?
>>  >
>>  > The patches are still sitting in my tree, as I've been unable
>>  > to come up with a test case that actually makes a difference.
>>  >
>>  > The "tar cf host:..." example is bogus, as the problem here is
>>
>>  Jonathan is right, the patch doesn't solve the general "tar cf host:"
>>  problem, but it was similar enough to what we were seeing in
>>  production -- changing the MTU on lo0 to 1500 will make the
>>  "tar cf host:" problem/solution more apparent, when host == localhost.
>
>Hmm, yes, by doing that, I'm able to reproduce the condition locally,
>thanks for pointing out a good test case.


Thanks for the update.

Jonathan, does it make sense to commit the current patch since it 
solves Paul's problem (and presumably doesn't break existing 
networking) or should it wait until the "tar cf host:" problem can be 
solved? I'm basically neutral on which way to go but it sounded like 
a big network performance issue for Paul when he initially reported 
the bug. (It might be affecting more people but they probably haven't 
done the analysis to better understand the problem.)

Mark

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-net" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?p0510010ab6c07a3f770d>