bge and checksums

2005-03-24 Thread Boris Kovalenko
Hello!
	I try to use DSNiff with my FreeBSD 5.4-PRE and bge NIC. Unfortunatelly 
it does no work. My supposition is that the root of problem is bad tcp 
checksums (as shown by tcpdump). And DSNiff (and underlaying libnids) 
are checking for checksums. As I undrestand, bge has txcsum flag, so tcp 
stack does not computes checksum itself. Am I right? And may I turn off 
txcsum flag without modifying bge driver?

--
With respect,
Boris
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "[EMAIL PROTECTED]"


Re: bge and checksums

2005-03-24 Thread Boris Kovalenko
Boris Kovalenko wrote:
Hello!
Forgot to add - when useing fxp with 5.4-PRE all is working fine.
--
With respect,
Boris
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "[EMAIL PROTECTED]"


Re: PPP Lan Bridge

2005-03-24 Thread Brian Somers
I would have thought that

server:
add 192.168.1.234/30 HISADDR

in ppp.linkup should work...

On Wed, 23 Mar 2005 13:00:03 -0600, "Chris Tusa at Linisys, LLC" <[EMAIL 
PROTECTED]> wrote:
> OK,
> 
> I put the solution in place and had great results. The ONLY issue that
> I had, is with some routing. In order for the rest of the network to
> communicate with the hosts at the maintainence shed, I had to set a
> 'static route'. I did so using DHCP, no problems there.
> 
> However, I have a problem on the PPP server in its routing table:
> 
> *  I cannot add the static route to 'rc.conf':
> 
> static_routes="theshed"
> route_theshed="192.168.1.234 192.168.1.233"
> 
> because it creates the table entry pointing to the wrong interface -
> fxp0 instead of tun0
> 
> * I tried the same thing in the 'ppp.conf' file. I tried several
> variations with no success:
> 
> server:
>add 192.168.1.234 192.168.1.233
>add 192.168.1.234/30 192.168.1.233
> 
> * I tried placing a shell script to be run in 'ppp.linkup' on the server
> server:
>   bg /etc/ppp/setroute.sh
> 
> and then setroute.sh:
> 
> #!/bin/sh
> route delete 192.168.1.234
> route add 192.168.1.234 192.168.1.233
> 
> STILL FAILS!
> 
> But once the PPP link is established, if I add the route manually or
> run the shell script, it works fine thereafter. So the problem is that
> the routing table gets updated with the WRONG interface.  The FreeBSD
> man page for 'route' is unclear about how to specify the interface on
> the command line. Any additional advice?
> 
> 
> 
> On Tue, 22 Mar 2005 16:17:53 -0600, Chris Tusa at Linisys, LLC
> <[EMAIL PROTECTED]> wrote:
> > Thanks Again,
> > 
> > Just a quick note on your last reply, don't know whether this will
> > make an impact:
> > 
> > > I guess the missing bit is that the timeclock machine needs a special
> > > routing table with 192.168.1.234/30 on it's crossed-over ethernet
> > > interface and 192.168.1.233 as the default route, allowing data to
> > > get back to the rest of the /24 subnet.
> > 
> > The timeclock itself is a proprietary STANDALONE device, not an actual
> > machine with an OS. It has its own configuration interface via its LCD
> > screen and buttons. It only allows me to add the IP, Netmask &
> > Gateway. So I probably can't perform routing table changes to that
> > device.
> > 
> > ( http://www.timeclockplus.com/products/hardware/markIII/mark3.aspx )
> > 
> > 
> > > When you get it all working it'll start to make sense (if it doesn't
> > > already).
> > 
> > I will try the configuration tommorrow and see what happens. Thanks so much.
> > 
> > --
> > Chris Tusa
> > [EMAIL PROTECTED]
> > http://people.linisys.com/ctusa
> > 
> > Buy books from my Half.com inventory:
> > http://half.ebay.com/shops/shops.jsp?seller_id=1691584
> > 
> 
> 
> -- 
> Chris Tusa
> [EMAIL PROTECTED]
> http://people.linisys.com/ctusa
> 
> Buy books from my Half.com inventory:
> http://half.ebay.com/shops/shops.jsp?seller_id=1691584
> 


-- 
Brian Somers  <[EMAIL PROTECTED]>
Don't _EVER_ lose your sense of humour !   <[EMAIL PROTECTED]>
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "[EMAIL PROTECTED]"


Re: Resolving MAC address

2005-03-24 Thread c0ldbyte
On Wed, 23 Mar 2005, Charlie Schluting wrote:
Olivier Nicole wrote:
? (10.0.0.1) at 00:e0:29:ad:5a:aa on em0 [ethernet]
will do the trick, but it is a bit too heavy for the purpose, I'd
prefer a solution that only send an ARP request.
If you just want to avoid the DNS lookup, you can use arp -an.
Its much faster :)
-Charlie
Also check out arping @ "/usr/ports/net/arping"
Best if luck,
--c0ldbyte
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "[EMAIL PROTECTED]"


Re: bge and checksums

2005-03-24 Thread Kelly Yancey
On Thu, 24 Mar 2005, Boris Kovalenko wrote:

> Hello!
>
>   I try to use DSNiff with my FreeBSD 5.4-PRE and bge NIC. Unfortunatelly
> it does no work. My supposition is that the root of problem is bad tcp
> checksums (as shown by tcpdump). And DSNiff (and underlaying libnids)
> are checking for checksums. As I undrestand, bge has txcsum flag, so tcp
> stack does not computes checksum itself. Am I right? And may I turn off
> txcsum flag without modifying bge driver?
>

  Have you tried the -txcsum option described in ifconfig(8)?

  Kelly

--
Kelly Yancey  -  [EMAIL PROTECTED],FreeBSD.org}  -  [EMAIL PROTECTED]
FreeBSD, The Power To Serve: http://www.freebsd.org/
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "[EMAIL PROTECTED]"


Re: Resolving MAC address

2005-03-24 Thread Peter Heerboth
One thing you can try if you want to avoid needing super user 
privileges is to send a small UDP packet to the given IP, which will 
force the kernel to ARP for the address if it is not already present.  
Then you can check the routing table through sysctl() by passing in 
{CTL_NET, PF_ROUTE, 0, AF_INET,NET_RT_FLAGS,RTF_LLINFO} for the name 
parameter and searching through the route message list.  All sending 
the UDP packet will do is force the kernel to arp, it doesn't actually 
matter if the other side is filtering or not.

On Mar 23, 2005, at 7:58 PM, Olivier Nicole wrote:
? (10.0.0.1) at 00:e0:29:ad:5a:aa on em0 [ethernet]
will do the trick, but it is a bit too heavy for the purpose, I'd
prefer a solution that only send an ARP request.
If you just want to avoid the DNS lookup, you can use arp -an.
Its much faster :)
Thanks. Off course I use arp -n to avoid DNS resolution.
But what i really want to avoid is sending any IP/ICMP packet when the
ARP resolution is all I need. (And some people even filter out the
ICMP echo request packets (Windows XP firewall), so I have to wait for
the time out).
Olivier
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "[EMAIL PROTECTED]"


Problems with High Point Tech RocketRaid 1640

2005-03-24 Thread John Barbieri
Good evening.
Recently I setup a free bsd 5.3-amd64 box with dual opterons and 1GB of ram.
I constructed a RAID 5 array using the HPT rocketraid 1640, conisiting 
of 4 disks, each with 200GB capacity.

Everything worked well, until I installed and ran Samba from the ports 
collection.

Now, apparently whenever I transfer with samba, the computer freezes for 
a few moments, then an array comes back, stating:

Mar 24 13:14:12 tie kernel: hpt374: Device removed: controller 1 channel 3
Mar 24 13:14:12 tie kernel: hpt374: Device removed: controller 1 channel 1

The weird thing is, I transferred over 15GB of data over FTP, but the 
second I used Samba (which is usually faster, which is why i switched to 
it), the controller stated the two drives disconnected.

How come I can transfer 15GB one way, and nothing the other?
Now my RAID array is broken, and I lost all the data on it (which is ok, 
since I have it backed up). But this is my 2nd time doing this today, 
thinking maybe the first time was a fluke.

info:
FreeBSD tie.jbarbieri.net 5.3-RELEASE-p5 FreeBSD 5.3-RELEASE-p5 #1: Wed 
Mar 23 21:01:23 EST 2005 
[EMAIL PROTECTED]:/usr/obj/usr/src/sys/TIE  amd64

pkg_info:
autoconf-2.13.000227_5 Automatically configure source code on many Un*x 
platforms
automake-1.4.6_1GNU Standards-compliant Makefile generator (legacy 
version
cvsup-without-gui-16.1h General network file distribution system 
optimized for CVS
gettext-0.14.1  GNU gettext package
gmake-3.80_2GNU version of 'make' utility
libiconv-1.9.2_1A character set conversion library
libtool-1.3.5_2 Generic shared library support script (version 1.3)
libtool-1.5.10_1Generic shared library support script (version 1.5)
linux_base-8-8.0_6  Base set of packages needed in Linux mode (only for 
i386)
m4-1.4.1GNU m4
perl-5.8.5  Practical Extraction and Report Language
popt-1.7A getopt(3) like library with a number of 
enhancements, fro
rpm-3.0.6_9 The Red Hat Package Manager
samba-2.2.12A free SMB and CIFS client and server for UNIX

as you can see, not much installed. I just rebuilt this box last night 
as well

If you have any ideas, please let me know. If you need more info, let me 
know.

Thank you in advanced
John Barbieri
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "[EMAIL PROTECTED]"


Re: Problems with High Point Tech RocketRaid 1640

2005-03-24 Thread Thomas M. Skeren III
Hey John...
Had many problems with operations and the amd 64 kernel.  I switched to 
the i386 kernel and have had no problems.

John Barbieri wrote:
Good evening.
Recently I setup a free bsd 5.3-amd64 box with dual opterons and 1GB 
of ram.

I constructed a RAID 5 array using the HPT rocketraid 1640, conisiting 
of 4 disks, each with 200GB capacity.

Everything worked well, until I installed and ran Samba from the ports 
collection.

Now, apparently whenever I transfer with samba, the computer freezes 
for a few moments, then an array comes back, stating:

Mar 24 13:14:12 tie kernel: hpt374: Device removed: controller 1 
channel 3
Mar 24 13:14:12 tie kernel: hpt374: Device removed: controller 1 
channel 1


The weird thing is, I transferred over 15GB of data over FTP, but the 
second I used Samba (which is usually faster, which is why i switched 
to it), the controller stated the two drives disconnected.

How come I can transfer 15GB one way, and nothing the other?
Now my RAID array is broken, and I lost all the data on it (which is 
ok, since I have it backed up). But this is my 2nd time doing this 
today, thinking maybe the first time was a fluke.

info:
FreeBSD tie.jbarbieri.net 5.3-RELEASE-p5 FreeBSD 5.3-RELEASE-p5 #1: 
Wed Mar 23 21:01:23 EST 2005 
[EMAIL PROTECTED]:/usr/obj/usr/src/sys/TIE  amd64

pkg_info:
autoconf-2.13.000227_5 Automatically configure source code on many 
Un*x platforms
automake-1.4.6_1GNU Standards-compliant Makefile generator (legacy 
version
cvsup-without-gui-16.1h General network file distribution system 
optimized for CVS
gettext-0.14.1  GNU gettext package
gmake-3.80_2GNU version of 'make' utility
libiconv-1.9.2_1A character set conversion library
libtool-1.3.5_2 Generic shared library support script (version 1.3)
libtool-1.5.10_1Generic shared library support script (version 1.5)
linux_base-8-8.0_6  Base set of packages needed in Linux mode (only 
for i386)
m4-1.4.1GNU m4
perl-5.8.5  Practical Extraction and Report Language
popt-1.7A getopt(3) like library with a number of 
enhancements, fro
rpm-3.0.6_9 The Red Hat Package Manager
samba-2.2.12A free SMB and CIFS client and server for UNIX

as you can see, not much installed. I just rebuilt this box last night 
as well

If you have any ideas, please let me know. If you need more info, let 
me know.

Thank you in advanced
John Barbieri
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "[EMAIL PROTECTED]"


Re: bge and checksums

2005-03-24 Thread Boris Kovalenko
Kelly Yancey wrote:
Hello!
	Arghh... my fault. Thinked that should be documented with bge manual. 
This really fixes my problem, thanks!

  Have you tried the -txcsum option described in ifconfig(8)?
  Kelly
--
Kelly Yancey  -  [EMAIL PROTECTED],FreeBSD.org}  -  [EMAIL PROTECTED]
FreeBSD, The Power To Serve: http://www.freebsd.org/

--
With respect,
Boris
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "[EMAIL PROTECTED]"