linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* 2.6.10 TCP troubles
@ 2005-01-05  8:13 Hubert Tonneau
  2005-01-05 12:07 ` Francois Romieu
  2005-01-05 23:33 ` Stephen Hemminger
  0 siblings, 2 replies; 10+ messages in thread
From: Hubert Tonneau @ 2005-01-05  8:13 UTC (permalink / raw)
  To: linux-kernel

Here is the senario:
the Linux machine is writting through libsmbclient
to an OSX machine running Samba

Switching the Linux machine from 2.6.8 to 2.6.10 made the network speed
drop drastically: 20 seconds with 2.6.8, 800 seconds with 2.6.10


^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: 2.6.10 TCP troubles
  2005-01-05  8:13 2.6.10 TCP troubles Hubert Tonneau
@ 2005-01-05 12:07 ` Francois Romieu
  2005-01-05 23:33 ` Stephen Hemminger
  1 sibling, 0 replies; 10+ messages in thread
From: Francois Romieu @ 2005-01-05 12:07 UTC (permalink / raw)
  To: Hubert Tonneau; +Cc: linux-kernel

Hubert Tonneau <hubert.tonneau@fullpliant.org> :
> Here is the senario:
> the Linux machine is writting through libsmbclient
> to an OSX machine running Samba
> 
> Switching the Linux machine from 2.6.8 to 2.6.10 made the network speed
> drop drastically: 20 seconds with 2.6.8, 800 seconds with 2.6.10

Are there any differences in:
- dmesg output
- /proc/interrupts 
- disk traffic
- tcpdump output (of course there will)

--
Ueimor

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: 2.6.10 TCP troubles
  2005-01-05  8:13 2.6.10 TCP troubles Hubert Tonneau
  2005-01-05 12:07 ` Francois Romieu
@ 2005-01-05 23:33 ` Stephen Hemminger
  2005-01-05 23:51   ` Francois Romieu
  1 sibling, 1 reply; 10+ messages in thread
From: Stephen Hemminger @ 2005-01-05 23:33 UTC (permalink / raw)
  To: linux-kernel

On Wed, 05 Jan 2005 08:13:17 GMT
Hubert Tonneau <hubert.tonneau@fullpliant.org> wrote:

> Here is the senario:
> the Linux machine is writting through libsmbclient
> to an OSX machine running Samba
> 
> Switching the Linux machine from 2.6.8 to 2.6.10 made the network speed
> drop drastically: 20 seconds with 2.6.8, 800 seconds with 2.6.10
> 
> -
> 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/

Some possiblities:
	2.6.8 still had the broken TCP segmentation offload that didn't obey
	congestion/slow start.  Are you using hardware that supports TSO?
	Does 2.6.8 behaviour change if you turn TSO off with ethtool?

	Is there window scaling or other issues?  Does 2.6.10 get faster if
	you turn of window scaling sys.net.ipv4.tcp_window_scaling=0?
	Is there a window scale corrupting firewall (like OpenBSD pf)
	in the way?

	Is there more packet loss on the router or the Mac?

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: 2.6.10 TCP troubles
  2005-01-05 23:33 ` Stephen Hemminger
@ 2005-01-05 23:51   ` Francois Romieu
  0 siblings, 0 replies; 10+ messages in thread
From: Francois Romieu @ 2005-01-05 23:51 UTC (permalink / raw)
  To: Stephen Hemminger; +Cc: linux-kernel, hubert.tonneau

Stephen Hemminger <shemminger@osdl.org> :
[...]
> 	Is there window scaling or other issues?  Does 2.6.10 get faster if

Bingo !

tcpdump exhibits a 2^5 factor differences in the advertised window when
the network speed is low.

--
Ueimor

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: 2.6.10 TCP troubles
  2005-01-05 12:50 Hubert Tonneau
  2005-01-05 14:04 ` Alan Cox
  2005-01-05 15:41 ` Barry K. Nathan
@ 2005-01-05 18:44 ` Francois Romieu
  2 siblings, 0 replies; 10+ messages in thread
From: Francois Romieu @ 2005-01-05 18:44 UTC (permalink / raw)
  To: Hubert Tonneau; +Cc: linux-kernel

Hubert Tonneau <hubert.tonneau@fullpliant.org> :
[...]
> The problem seems to me to be related to the way the TCP layer is handling small
> troubles (probably lost packets on the Mac side because the Linux machine is
> gigabit connected to the switch, with flow control enabled, and the Mac is
> 100 Mbps connected, full duplex, but without flow control).

tcpdump should enlighten it.

--
Ueimor

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: 2.6.10 TCP troubles
  2005-01-05 12:50 Hubert Tonneau
  2005-01-05 14:04 ` Alan Cox
@ 2005-01-05 15:41 ` Barry K. Nathan
  2005-01-05 18:44 ` Francois Romieu
  2 siblings, 0 replies; 10+ messages in thread
From: Barry K. Nathan @ 2005-01-05 15:41 UTC (permalink / raw)
  To: Hubert Tonneau; +Cc: Francois Romieu, linux-kernel

On Wed, Jan 05, 2005 at 12:50:57PM +0000, Hubert Tonneau wrote:
[quote reformatted to fit within 80 columns]
> The problem seems to me to be related to the way the TCP layer is
> handling small troubles (probably lost packets on the Mac side because
> the Linux machine is gigabit connected to the switch, with flow control
> enabled, and the Mac is 100 Mbps connected, full duplex, but without
> flow control).

What OS is the Mac running? If it's Mac OS, then is it Mac OS X or is it
an earlier version?

-Barry K. Nathan <barryn@pobox.com>


^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: 2.6.10 TCP troubles
@ 2005-01-05 15:30 Hubert Tonneau
  0 siblings, 0 replies; 10+ messages in thread
From: Hubert Tonneau @ 2005-01-05 15:30 UTC (permalink / raw)
  To: Barry K. Nathan; +Cc: Francois Romieu, linux-kernel

Barry K. Nathan wrote:
>
> Mail agent comments:
>   Sender configuration is suspicious.
> 
> On Wed, Jan 05, 2005 at 12:50:57PM +0000, Hubert Tonneau wrote:
> [quote reformatted to fit within 80 columns]
> > The problem seems to me to be related to the way the TCP layer is
> > handling small troubles (probably lost packets on the Mac side because
> > the Linux machine is gigabit connected to the switch, with flow control
> > enabled, and the Mac is 100 Mbps connected, full duplex, but without
> > flow control).
> 
> What OS is the Mac running? If it's Mac OS, then is it Mac OS X or is it
> an earlier version?

The Mac are all running OSX.
 


^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: 2.6.10 TCP troubles
@ 2005-01-05 15:14 Hubert Tonneau
  0 siblings, 0 replies; 10+ messages in thread
From: Hubert Tonneau @ 2005-01-05 15:14 UTC (permalink / raw)
  To: Alan Cox; +Cc: Francois Romieu, Linux Kernel Mailing List

Alan Cox wrote:
>
> Mail agent comments:
>   Sending server is suspicious.
> 
> On Mer, 2005-01-05 at 12:50, Hubert Tonneau wrote:
> > troubles (probably lost packets on the Mac side because the Linux machine is
> > gigabit connected to the switch, with flow control enabled, and the Mac is
> > 100 Mbps connected, full duplex, but without flow control).
> 
> Through a firewall ?

No:

Mac <-> 100 Mbps switch <-> gigabit switch <-> Linux

One possible explaination, even if unlikely, might be that Linux 2.6.10 is
faster than 2.6.8, so the Mac start missing packets.

If you want me to make tests, I can switch back to 2.6.10 at night, perform
tests, and switch back to 2.6.8 before production resumes in the morning.


^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: 2.6.10 TCP troubles
  2005-01-05 12:50 Hubert Tonneau
@ 2005-01-05 14:04 ` Alan Cox
  2005-01-05 15:41 ` Barry K. Nathan
  2005-01-05 18:44 ` Francois Romieu
  2 siblings, 0 replies; 10+ messages in thread
From: Alan Cox @ 2005-01-05 14:04 UTC (permalink / raw)
  To: Hubert Tonneau; +Cc: Francois Romieu, Linux Kernel Mailing List

On Mer, 2005-01-05 at 12:50, Hubert Tonneau wrote:
> troubles (probably lost packets on the Mac side because the Linux machine is
> gigabit connected to the switch, with flow control enabled, and the Mac is
> 100 Mbps connected, full duplex, but without flow control).

Through a firewall ?


^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: 2.6.10 TCP troubles
@ 2005-01-05 12:50 Hubert Tonneau
  2005-01-05 14:04 ` Alan Cox
                   ` (2 more replies)
  0 siblings, 3 replies; 10+ messages in thread
From: Hubert Tonneau @ 2005-01-05 12:50 UTC (permalink / raw)
  To: Francois Romieu; +Cc: linux-kernel

Francois Romieu wrote:
>
> Hubert Tonneau <hubert.tonneau@fullpliant.org> :
> > Here is the senario:
> > the Linux machine is writting through libsmbclient
> > to an OSX machine running Samba
> > 
> > Switching the Linux machine from 2.6.8 to 2.6.10 made the network speed
> > drop drastically: 20 seconds with 2.6.8, 800 seconds with 2.6.10
> 
> Are there any differences in:
> - dmesg output

No.

> - /proc/interrupts 
> - disk traffic
> - tcpdump output (of course there will)

I cannot see anymore since it's our main production server, so I switched
back at once. Sorry about that. Anyway, both network traffic and disk traffic
was very low.

The problem is not related to the Linux machine beeing slow because the network
exchange was very fast with other gigabit with flow control connected machines.
The problem seems to me to be related to the way the TCP layer is handling small
troubles (probably lost packets on the Mac side because the Linux machine is
gigabit connected to the switch, with flow control enabled, and the Mac is
100 Mbps connected, full duplex, but without flow control).

Please notice that the Linux machine is the client, and is pushing files to
the Mac, which is quite unusual. If the Mac was the client pulling files from
the PC, I bet things might be very different.


^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2005-01-05 23:52 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2005-01-05  8:13 2.6.10 TCP troubles Hubert Tonneau
2005-01-05 12:07 ` Francois Romieu
2005-01-05 23:33 ` Stephen Hemminger
2005-01-05 23:51   ` Francois Romieu
2005-01-05 12:50 Hubert Tonneau
2005-01-05 14:04 ` Alan Cox
2005-01-05 15:41 ` Barry K. Nathan
2005-01-05 18:44 ` Francois Romieu
2005-01-05 15:14 Hubert Tonneau
2005-01-05 15:30 Hubert Tonneau

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).