rters wherever possible. Sometimes
it is not possible.
--eli
- --
Eli Dart Office: (510) 486-5629
ESnet Network Engineering Group Fax:(510) 486-6712
Lawrence Berkeley National Laboratory
PGP Key fingerprint = C970 F8D3 C
it is removed for the default
config... In other words, something like:
net.link.mtu_limits_received_pktsize = 0|1
Then, default it to 0 to preserve 4.x behavior.
Thanks,
--eli
--
Eli Dart Office: (510) 486-5629
ESnet Network Engineering
see below...
Julian Elischer wrote:
Eli Dart wrote:
Stephen Clark wrote:
So was any decision reached on this issue - will FreeBSD changed
to accept a packet on an interface that is larger than the mtu on
that interface?
If possible, I'd like to see the ability to enforce interfac
In reply to [EMAIL PROTECTED] :
>
> Hi Nickolay,
>
> > "Nickolay" == Nickolay A Kritsky <[EMAIL PROTECTED]> writes:
>
> Nickolay> Hello sirola, Try to switch one of the NICs to
> Nickolay> half-duplex mode. This should do the trick.
>
> thanks for the tip, it works now. What could
In reply to "Marc G. Fournier" <[EMAIL PROTECTED]> :
>
> I have 5 servers sitting on a Linksys 10/100 switch ... 4 of the 5 are
> running fxp0 ethernet, while the 5th is running em ... and the 5th
> performs atrociously:
>
> neptune# netstat -ni | head
> NameMtu Network Address
In reply to Gregory Kuhn <[EMAIL PROTECTED]> :
> Hello,
>
>
> I have discovered an anomaly (I am sure it was discovered long
> ago) when trying to traceroute to my FBSD 4.8 and 4.9 servers. Here are
> the details.
Have you set sysctl net.inet.udp.blackhole=1 ?
If so, the FreeBSD b
In reply to Darcy Buskermolen <[EMAIL PROTECTED]> :
> On August 25, 2004 06:44 am, Marko Zec wrote:
> > On Wednesday 25 August 2004 00:15, Julian Elischer wrote:
> > > You do know don't you, that if you continue to do these things, you will
> > > be punnished by
> > > getting a CVS commit bit..?
In reply to Omer Faruk Sen <[EMAIL PROTECTED]> :
> According to the very old article stated
> http://www.totse.com/en/technology/computer_technology/162444.html there is
> no way to tune time_wait timeout in FreeBSD. But since it is very old
> article my question is this:
>
> Is there a way t
In reply to Mike Silbersack <[EMAIL PROTECTED]> :
>
> On Mon, 11 Oct 2004, Eli Dart wrote:
>
> >
> >> Is there a way to change the time_wait timeout value in FreeBSD?
> >
> > sysctl -w net.inet.tcp.msl=
> >
> > The default is 3 (30
> [ ... ]
> > I then tested with the whole school going through the firewall: very bad.
> > packets were being droped and ping times were around 600ms. Internet was
> > pretty much unuseable.
>
> This report sounds consistent, although you could also have a bad cable or
> switch port, too. It
In reply to "Andrew Seguin" <[EMAIL PROTECTED]> :
> Thank you for your reply.
>
> > Note that if your interface is set to full duplex and the switch is
> > set to half, you'll see lousy performance and very likely see no
> > errors on your side.
>
> I have to question though... could this expla
In reply to Julian Elischer <[EMAIL PROTECTED]> :
> I have no control over or access to the link.. all I have is a promise
> that they will deliver
> 14Mb/Sec. with approc 300mSec. RTT to me but there is no promise about
> packet order.
My guess is that they are doing round-robin load balanci
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi all,
We are having trouble getting a polling-enabled kernel to boot. It
gets as far as listing the timecounter values, and then resets twa0
every couple of minutes forever.
The box is 2x3.2GHz Xeons, 4GB RAM, SuperMicro X6DH8-XG2 motherboard
with
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi List,
We've just deployed a new syslog server infrastructure consisting of
two hosts -- one NFS server that has a large disk array attached via
fibre channel and one NFS client that receives syslog messages,
mounts the server's disk, and writes to
ient NFS layer fetching the data from the server and then
immediately expiring it?
Thanks,
--eli
>
>noac Suppress data and attribute caching.
>
> :-)
>
- --
- ---
Eli Dart
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Olivier Nicole wrote:
> Hi,
>
> I am facing the following problem: I have a web server with an
> application that calls a MySQL server.
>
> For class and test run, I may have 100 users accessing the same web
> page to login to the same database.
>
In reply to "Bruce A. Mah" <[EMAIL PROTECTED]> :
> If memory serves me right, Eli Dart wrote:
>
> > In reply to "Crist J . Clark" <[EMAIL PROTECTED]> :
> >
> > > On Tue, Jan 22, 2002 at 05:37:07PM -0800, Eli Dart wrote:
> > >
The syskonnect 1000baseSX and 1000baseTX cards work quite well as of
4.2-RELEASE (man sk). I can only assume that they continue to be
supported (they are as of 4.4-R.).
If you have existing tigon nics, this won't help you. If you're
looking for stuff to buy, syskonnect cards work fine.
less /usr/include/netinet/ip_icmp.h
The stuff that I think you want starts around line 135.
Looking at a copy of Stevens TCP/IP Illustrated vol. 1 might also
help, but I always just read the header file to find out what's what.
--eli
In reply to "Peter Brezny" <[EMAIL PROTECT
It's actually fairly easy to take the output of netstat -inb and put
it into rrdtool. If you grep Link out of netstat -inb you get in and
out packets and bytes, and you get error counters as well.
My $0.02
--eli
In reply to Larry Sica <[EMAIL PROTECTED]> :
> This is a multi
e anything in sysctl that looked obvious, but I'm
perfectly willing to believe I missed it.
So, can this be turned off? Also, what is the timeout on this data
in the kernel?
Thanks!
--eli
----
g something?
--eli
In reply to Eli Dart <[EMAIL PROTECTED]> :
>
> --==_Exmh_1318550241P
> Content-Type: text/plain; charset=us-ascii
>
> Hi all,
>
> I'm seeing something strange here... I have a freebsd box running
> iperf (4.6-RELEASE-p1,
In reply to "Jim McGrath" <[EMAIL PROTECTED]> :
>
> > Where could I get the errata sheet?
>
> Product Specification Updates i.e. errata, and the Product Specification
> itself are available from Intel under a Non Disclosure Agreement. Unless
> you work for a company that is doing business with
irewall filter?
--eli
> $ sysctl -a | grep mtu
> net.inet.tcp.path_mtu_discovery: 1
>
> Now if I change the mtu of the gre to 1412 everything works.
>
> Any insight would be appreciated.
>
> Thanks,
> Steve
- --
Eli Dart
We maintain our own patches here as well to get around this problem.
IMHO, it is far better to have some applications "waste" a meg or two
of buffer space than to hamstring any high-performance bpf app that
runs on a FreeBSD box.
This is most likely a trivial code fix -- how hard would it be to
In reply to Rostislav Krasny <[EMAIL PROTECTED]> :
> Thank you for your trying to help me. Your version of ppp.conf is very
> similar to mine. I don't have LAN here, but only one box with FreeBSD
> connected to the outside world through my ADSL modem. So ' set nat' and
> ' set proxy' options are
r post
that this works). You could also run with a slightly smaller MTU and
declare victory :)
--eli
In reply to Rostislav Krasny <[EMAIL PROTECTED]> :
>
> --0-2081754345-1041105297=:86278
> Content-Type: text/plain; charset=us-ascii
> Content-Disposition: in
In reply to Rostislav Krasny <[EMAIL PROTECTED]> :
>
> --0-1140876309-1041159137=:12973
> Content-Type: text/plain; charset=us-ascii
> Content-Disposition: inline
>
> I have some sniffer in Win98SE but don't know how to save its dump in
> the text format to make it easy to read. So I maked a sc
In reply to Luigi Rizzo <[EMAIL PROTECTED]> :
> is this with ipfw1 or ipfw2 or both ?
>
> cheers
> luigi
>
> On Fri, Jan 24, 2003 at 03:56:54AM -0800, Josh Brooks wrote:
> >
> > I have inserted this ipfw rule, based on guidance from the archives:
> >
> > count icmp from any to any
Look at the man page for ifconfig, and look at the alias option.
--eli
In reply to David Myer <[EMAIL PROTECTED]> :
> Hi,
> Can anyone tell me how to set multiple IP addresses
> for one interface in UNIX/LINUX ?
> Thanks
> Dave
>
> __
In reply to "Eicke" <[EMAIL PROTECTED]> :
> Hi folks,
>
> I have two servers in the same network segment and with the same =
> hardware.
> The First work as a router and second is a DNS, a HTTP, and Postifix =
> server.
>
> I realized a test, I tried to copy a 1Gb file from servers to my =
> ma
In reply to "Peter J. Blok" <[EMAIL PROTECTED]> :
[snip]
> My understanding is that the Giga-bit definition includes large frame support
It doesn't.
> and if you claim to have a Giga-bit switch you should support large frames,
> unless specifically excluded.
There is no IEEE standard for jum
In reply to Mark Allman <[EMAIL PROTECTED]> :
>
> Hi folks!
>
> Are there any plans to incorporate SACK in FreeBSD? It'd sure be
> handy for me to have (I prefer the *BSDs, and, alas, they are the
> only remaining SACK-less systems worth mentioning). I think there
> are research implementati
In reply to Alex (ander Sendzimir) <[EMAIL PROTECTED]> :
> First, I know very little about networking, especially performance
> turning. I would really like to learn more but don't know where/how to
> start effectively.
Take a look at the tools ttcp, netperf and iperf. They build
straight ou
In reply to David Gilbert <[EMAIL PROTECTED]> :
> In our experience, switch to fxp ethernet cards, test several
> motherboards and enable polling.
>
> fxp and em cards appear to have the best performance ... outrunning
> other cards by a fair margin.
Hmmmwe've been using SysKonnect (older o
In reply to Andriy Korud <[EMAIL PROTECTED]> :
>
> Hi,
> At last I've managed to build stable NAT on FreeBSD box for 34Mbit link and
> ~2000 clients (cable modem network).
> At full speed (34Mbit) CPU usage is 0% and system load is 0.0 :-)
>
> The solution was to find proper ip_nat.h file and p
In reply to Don Lewis <[EMAIL PROTECTED]> :
> On 28 Jan, Garrett Wollman wrote:
> > < >
> >> Can different MTUs be mixed on the same wire
> >
> > No.
>
> It's ugly, but I wonder if adding host routes with the -lock -mtu
> options might work ...
I wouldn't even mess with it. If you have an MT
In reply to Mike Hunter <[EMAIL PROTECTED]> :
> On Jan 30, "To [EMAIL PROTECTED]" wrote:
>
> > Hi,
>
> [snip]
>
> I switched the two pieces of hardware, and the photons still prefer going
> uphill, so maybe there's a problem with the fiber after all. I'd still
> appreciate any hints on what t
In reply to Brandon Erhart <[EMAIL PROTECTED]> :
> Hello everyone,
> However, I have run into a new problem. I am getting a good amount of
> blocks stuck in FIN_WAIT_1, FIN_WAIT_2 or LAST_ACK that stick around for a
> long while.
Could you define "long" in this case? Are we talking about 60
In reply to Brandon Erhart <[EMAIL PROTECTED]> :
> Well, I responded to the group that I had taken one of the fellows advice
> posting here, and modified the tcp_usrclosed in netinet/tcp_usrreq.c.
I understand that -- I was trying to discover if you'd come across
something that needed a more
40 matches
Mail list logo
