Hi,
I'm trying to debug some general network slowness with my 6.8 server
(i7-3930k) that seems to affect all protocols (e.g. Samba capping at
70MB/s, FTP at 45MB/s for upload). I've run some iperf3/tcpbench tests
and the results seems low even when running both client and server
on the same
Hi. I tried what you said. Both recvspace and sendspace were set to
16384. I set both of them to 131070 and tried iperf again. It wasn't
any faster.
Thanks,
Gabe
On Mon, May 31, 2010 at 10:44 PM, jean-philippe luiggi
j...@didconcept.com wrote:
Hello,
Please check the result of :
#sysctl
Using iperf, I get around 300 mbits/s. between my openbsd machine and
my windows xp machine and also to my imac with os x. I tried the
kernel tweaks mentioned here:
https://calomel.org/network_performance.html, but they did not help a
whole lot (before tweaks I was getting around 220 mbits/s).
Hello everybody,
I'm experiencing a very bad network performance, when I try to connect
to a remote server.
The point-to-point connection is a E3 line, with 34MBit/s, with a cisco 2800
router on each side, terminating the point-to-point connection.
These cisco routers have two gigabit
On Fri, Oct 24, 2008 at 12:58:27PM +0200, Sebastian Reitenbach wrote:
Hello everybody,
I'm experiencing a very bad network performance, when I try to connect
to a remote server.
The point-to-point connection is a E3 line, with 34MBit/s, with a cisco 2800
router on each side, terminating
-Urspr|ngliche Nachricht-
Von: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Im Auftrag von Otto Moerbeek
Gesendet: Freitag, 24. Oktober 2008 13:11
An: Sebastian Reitenbach
Cc: misc@openbsd.org
Betreff: Re: slow network performance behind cisco
On Fri, Oct 24, 2008 at 12:58:27PM
network performance behind cisco
On Fri, Oct 24, 2008 at 12:58:27PM +0200, Sebastian Reitenbach wrote:
Hello everybody,
I'm experiencing a very bad network performance, when I try
to connect
to a remote server. The point-to-point connection is a E3
line, with
34MBit/s
Moerbeek
Gesendet: Freitag, 24. Oktober 2008 13:11
An: Sebastian Reitenbach
Cc: misc@openbsd.org
Betreff: Re: slow network performance behind cisco
On Fri, Oct 24, 2008 at 12:58:27PM +0200, Sebastian Reitenbach wrote:
Hello everybody,
I'm experiencing a very bad network
Hi,
-Urspr|ngliche Nachricht-
Von: Otto Moerbeek [EMAIL PROTECTED]
Gesendet: 24.10.08 13:11:39
An: Sebastian Reitenbach [EMAIL PROTECTED]
CC: misc@openbsd.org
Betreff: Re: slow network performance behind cisco
On Fri, Oct 24, 2008 at 12:58:27PM +0200, Sebastian Reitenbach wrote
Sebastian Reitenbach wrote:
Von: Otto Moerbeek [EMAIL PROTECTED]
OpenBSD uses a pretty low default send and receive buffer size for
sockets. Try increasing net.inet.tcp.recvspace and
net.inet.tcp.sendspace, after reading a bit about bandwidth * delay
products.
... after a bit of reading,
-Urspr|ngliche Nachricht-
Von: Lars Noodin [EMAIL PROTECTED]
Gesendet: 24.10.08 17:34:03
An: Sebastian Reitenbach [EMAIL PROTECTED]
CC: misc@openbsd.org
Betreff: Re: slow network performance behind cisco
Sebastian Reitenbach wrote:
Von: Otto Moerbeek [EMAIL PROTECTED]
OpenBSD
Thanks Otto and Sebastian.
Otto Moerbeek [EMAIL PROTECTED] wrote:
OpenBSD uses a pretty low default send and receive buffer size for
sockets. Try increasing net.inet.tcp.recvspace and
net.inet.tcp.sendspace, after reading a bit about bandwidth * delay
products.
Sebastian Reitenbach wrote:
On 16:46:18 Oct 24, Pierre Riteau wrote:
Because the delay between the two machines is very low.
It appears you need to read about bandwidth-delay product as well.
Actually it is quite a deep concept.
How TCP guesses the bandwidth is quite a challenge.
And often TCP gets it wrong.
It is
On Wed, 2006-11-08 at 17:35 +1300, Josh wrote:
Can anyone help out here please?
IIRC, a high amount of interrupts from a NIC during a transfer seems to
imply a poor NIC...
Now, I've never seen this kind of behavior out of an Intel NIC -- they
tend to be of at least decent quality -- this is
Hello. Ive got a few openbsd firewalls set up with carp and pfsync
running on them.
Anyway, basically any transfer that goes thru the firewalls is slow, as
in 1megabyte per second.
I notice that while the transfer is happening, in top, I see the % of
interrupt go to around 80%, and any other
Can anyone help out here please?
On Wed, 2006-11-08 at 15:17 +1300, Josh wrote:
Hello. Ive got a few openbsd firewalls set up with carp and pfsync
running on them.
Anyway, basically any transfer that goes thru the firewalls is slow, as
in 1megabyte per second.
I notice that while the
Can anyone help out here please?
On Wed, 2006-11-08 at 15:17 +1300, Josh wrote:
Hello. Ive got a few openbsd firewalls set up with carp and pfsync
running on them.
Anyway, basically any transfer that goes thru the firewalls is slow, as
in 1megabyte per second.
I notice that while the
...
wd0(pciide0:0:0): timeout
type: ata
c_bcount: 512
c_skip: 0
wd0e: DMA error writing fsbn 2651200 of 2651200-2651211 (wd0 bn 3819472; cn
3789 tn 2 sn 34), retrying
wd0: soft error (corrected)
wd0(pciide0:0:0): timeout
type: ata
c_bcount: 0
Hi!
Thank you for the tips!
Sorry, for the long mail, but I wanna show you some server statistics.
To answer a few further questions:
1. sysctl -w net.inet.ip.mtudisc=0
doesn't have any effect
2. no important messages in /var/log/messages during up-/download
3. Samba downloads show similar
On 2/15/06, Joachim Mathes [EMAIL PROTECTED] wrote:
So I think the problem is my hdd.
What do you think?
I think the same, get a new disk in there. Until then you can run a
dd if=/rwd0c of=/dev/null bs=1m to read every sector of the hdd and
maybe allocate some of the spare sectors to some of
I recently tried to use netperf, but it seemed more to test my CPU than
the network and thus reporting low througput. benchmarks/netstrain is
much less demanding on the CPU. Of course, one may use ftp to download
large files since the OpenBSD one reports speed as well.
/Sigfred
Sebastian
Joachim Mathes wrote:
- switched network cards from Realtek 8139 (100Mbit) to Realtek 8029 (10Mbit;
works fine for pppoe) (I know they are of poor quality!)
Throw out the Realtek cards and buy a good nic. I had a server (clone
P3) with network performance issues, saving files to the
[EMAIL PROTECTED] wrote:
I recently tried to use netperf, but it seemed more to test
my CPU than
the network and thus reporting low througput. benchmarks/netstrain is
much less demanding on the CPU. Of course, one may use ftp to download
large files since the OpenBSD one reports speed as
Hi OpenBSD community!
I built a small ethernet network with Windows and Linux boxes and an
OpenBSD router which is connected to the www via dsl.
The problem is that the upload speed from an intranet box to the BSD
server (over scp for example) is quite ok (about 13 MBit/s) but the
download
Hi,
if I understand you correctly your testing consists of up and downloads
on your openbsd router. This is not the proper way to test network
performance on a router. I recommend using tools that dont involve any
i/o operations such as netperf etc.
According to your dmesg I wouldn't be
25 matches
Mail list logo