[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[lwip-users] Re: decreasing throughput ends in lwip-collapse
From: |
Andre Puschmann |
Subject: |
[lwip-users] Re: decreasing throughput ends in lwip-collapse |
Date: |
Mon, 30 Oct 2006 19:09:34 +0100 |
User-agent: |
Thunderbird 1.5.0.4 (X11/20060619) |
Kieran Mansley wrote:
> On Sat, 2006-10-28 at 13:01 +0200, Andre Puschmann wrote:
>> Hello,
>>
>> sorry for the delay once again.
>> now I can give some more detailed information about what's going on. I
>> attached a ethereal capture file and a log file.
>
> I'd also check to see if there are large numbers of packets being
> discarded due to bad checksums (at both ends). Although ethereal often
> mis-reports bad checksums (e.g. if it has captured only part of a
> packet) there are a large number being flagged as bad in that trace.
>
> Kieran
Hi Kieran,
I will post a full captured connection tomorrow, since I don't have more
files on this box.
I also noticed that ethereal often (always?) mis-reports bad checksums.
If I remember correctly all ACKs of the windows-box are flagged as
"incorrect".
The curious of all those problems is that at the beginning everything
works well.
And than at a particular point .. the whole system is producing very odd
things.
Sometimes I can transfer up to 1,5gb (with 25mbit/s) with one
connection. So it doesn't seem to be a buffer or memory problem, true?
The fact that I can provoke this point e.g. starting a new connection or
producing some load backs this up. But this doesn't work regularly.
Regards
André