headers
Thor Lancelot Simon | 1 Aug 2008 02:53

Re: carp and dead daemon

On Thu, Jul 31, 2008 at 04:32:13PM -0400, matthew sporleder wrote:
> On Thu, Jul 31, 2008 at 3:57 PM, Emmanuel Dreyfus <manu <at> netbsd.org> wrote:
> > matthew sporleder <msporleder <at> gmail.com> wrote:
> >
> >> Have you considered using something like pen (pkgsrc /net/pen) for
> >> your services instead?
> >
> > If the service is SMTP, then having the real sender IP (and not pen load
> > balancer's one) is important for spam filtering,
> 
> This is a common problem with load balancers and proxies

Only ones which suck.  Load balancers which don't suck can proxy connections
at the application layer _without_ changing the source IP address seen by
the ultimate server.

(Disclosure: I work for one of the several load balancer vendors whose
 products do not suck, at least not this way.)

Thor
Permalink | Reply | 0 comments |
headers
Carlos Linares | 1 Aug 2008 04:58

802.11n / Atheros

Hello all.  Are there drivers for any 802.11n radios in NetBSD-current?

Do any radios based on the Atheros 5008 or 9xxx chipsets (like the D-link
DWA-556 or the sweet AR9160-based Ubuquiti SR71-A) work yet?  Could they work (or be made to work) under the
ath driver as 802.11[a]/b/g radios?

Atheros
just released ISC-license open source drivers for their 9xxx/5xxx 802.11n chips with no proprietary HAL. 
How easy would it be to port this to NetBSD?  Would this be a good project for someone new to coding in NetBSD to
embark upon (with guidance, of course)?

Thanks!

Atheros Releases Free Linux Driver For Its 802.11n Devices
http://mobile.slashdot.org/article.pl?sid=08/07/26/2138228

<<carlos>>
<<qvidnvnc>>
Permalink | Reply | 0 comments |
headers
Scott Presnell | 1 Aug 2008 18:44
Favicon

ex0: transmit overrun: debugging? configurational changes?

Hi,
	I recently booted a NetBSD-4.0 MP kernel on my Dell 420, after having
run 3.0_STABLE for several years without problems. Mostly it works, except for 
the following ethernet errors:

Aug  1 01:40:17 dirt /netbsd: ex0: transmit underrun (9000)  <at> 20
Aug  1 01:40:18 dirt /netbsd: ex0: transmit underrun (9000)  <at> 20, new threshold is 40
...
Aug  1 01:50:00 dirt /netbsd: ex0: transmit underrun (9000)  <at> 40
Aug  1 01:50:05 dirt /netbsd: ex0: transmit underrun (9000)  <at> 40
Aug  1 01:50:05 dirt /netbsd: ex0: transmit underrun (9000)  <at> 40, new threshold is 60
...
Aug  1 08:24:35 dirt /netbsd: ex0: transmit underrun (9000)  <at> 60
Aug  1 08:24:35 dirt /netbsd: ex0: transmit underrun (9000)  <at> 60, new threshold is 80
Aug  1 08:43:18 dirt /netbsd: ex0: transmit underrun (9000)  <at> 80
Aug  1 08:53:09 dirt /netbsd: ex0: transmit underrun (9000)  <at> 80
Aug  1 08:58:40 dirt /netbsd: ex0: transmit underrun (9000)  <at> 80
Aug  1 09:22:31 dirt /netbsd: ex0: transmit underrun (9000)  <at> 80

and associated interface Oerrs.  On the switch side, a HP 2424m, I have also 
seen "Excessive CRC/alignment errors" on this port.  These errors are new since
booting 4.0

The interface is configured as follows at boot time:

ex0: flags=8863<UP,BROADCAST,NOTRAILERS,RUNNING,SIMPLEX,MULTICAST> mtu 1500
         capabilities=3f00
         enabled=0
         address: 00:b0:d0:ab:aa:c6
         media: Ethernet autoselect (100baseTX full-duplex)
(Continue reading)

Permalink | Reply | 0 comments |
headers
Patrick Welche | 1 Aug 2008 23:35
Picon
Picon
Favicon

Re: hostapd fun

.. and the answer is, proto should be WPA not RSN..

On Sat, Jul 05, 2008 at 09:23:10PM +0100, Patrick Welche wrote:
> On Fri, Jul 04, 2008 at 01:02:38PM -0500, David Young wrote:
> > On Fri, Jul 04, 2008 at 04:47:31PM +0100, Patrick Welche wrote:
> > > I'm having my first attempt at hostapd, but get
> > >
> > >   ioctl[SIOCS80211]: No such file or directory  == ENENT
> >
> > Does this stop it from working?
> 
> No, it probably doesn't work because of my configuration... I finally saw
> a "Authentication with ... timed out" scroll by..
> 
> Here's my attempted configuration. hostapd.conf:
> 
> interface=ath0
> logger_syslog=-1
> logger_syslog_level=2
> logger_stdout=-1
> logger_stdout_level=1
> debug=2
> dump_file=/tmp/hostapd.dump
> ctrl_interface=/var/run/hostapd
> ctrl_interface_group=wheel
> ssid=NetBSD
> macaddr_acl=0
> auth_algs=2     # i.e., SHARED
> wpa=2           # i.e., RSN
> wpa_psk=4bc6b2b83d40a40922bf739a42f5885ef3402e51f620cc34fa045b978379d9d1
(Continue reading)

Permalink | Reply | 0 comments |
headers
6bone | 4 Aug 2008 09:21
Picon

panic at _bus_dmamap_sync

hello,

I am testing NetBSD 4.99.72. (amd64) During the boot the kernel crashed. 
You can find the screenshot at http://suse.uni-leipzig.de/dump6.jpg

You can also download the kernel, map and gdb file

http://suse.uni-leipzig.de/netbsd.kernel
http://suse.uni-leipzig.de/netbsd.map
http://suse.uni-leipzig.de/netbsd.gdb

Any idea what the problem can be?
It looks like a network error, but I am not sure.

Thank you for your efforts
Uwe
Permalink | Reply | 0 comments |
headers
Iain Hibbert | 5 Aug 2008 00:11

ifconfig -a error

Hi,

with the new ifconfig (on i386) I notice an error

% ifconfig -a
eon0: flags=3<UP,BROADCAST> mtu 1500
ifconfig: link_status: ioctl: Invalid argument

% ifconfig eon0
eon0: flags=3<UP,BROADCAST> mtu 1500
ifconfig: link_status: ioctl: Invalid argument

% ifconfig ppp0
ppp0: flags=8051<UP,POINTOPOINT,RUNNING,MULTICAST> mtu 1500
        inet 10.34.129.29 -> 10.64.64.64 netmask 0xff000000

It seems to be related to the SIOCGLIFADDR in af_link.c but I don't know
what that does. I have "options EON,ISO,TPIP" in my kernel config file
(though I don't use the eon interface) left over from when I was working
in the netiso code so perhaps its an incompatibility with the eon driver?

iain
Permalink | Reply | 4 comments |
headers
Ignatios Souvatzis | 5 Aug 2008 07:45
Picon

Re: ifconfig -a error


On Mon, Aug 04, 2008 at 11:11:00PM +0100, Iain Hibbert wrote:

> It seems to be related to the SIOCGLIFADDR in af_link.c but I don't know
> what that does. I have "options EON,ISO,TPIP" in my kernel config file
> (though I don't use the eon interface) left over from when I was working
> in the netiso code so perhaps its an incompatibility with the eon driver?

That's what I suspect - I was about to do some tests today.

	-is
Permalink | Reply | 2 comments |
headers
Jason Thorpe | 5 Aug 2008 18:05

Re: Socket options KPI


On Jun 26, 2008, at 2:40 PM, Elad Efrat wrote:

> On Thu, Jun 26, 2008 at 10:02 PM, Thor Lancelot Simon <tls <at> rek.tjls.com 
> > wrote:
>> On Fri, Jan 18, 2008 at 07:08:37PM +0200, Elad Efrat wrote:
>>> Hi,
>>>
>>> I have a patch to make a large portion of the network code use a new
>>> "struct sockopt" (inspired by similar work done in FreeBSD, with  
>>> input
>>> from dyoung <at>  and yamt <at> ) instead of passing socket options in mbufs.
>>
>> What happened to this?  It looked like you got mostly positive  
>> comments...
>
> See http://mail-index.netbsd.org/tech-net/2008/05/02/msg000501.html.

Let's get this checked in!

>
>
> -e.

-- thorpej
Permalink | Reply | 3 comments |
headers
David Young | 5 Aug 2008 19:11
Picon
Favicon

Re: ifconfig -a error

On Mon, Aug 04, 2008 at 11:11:00PM +0100, Iain Hibbert wrote:
> Hi,
> 
> with the new ifconfig (on i386) I notice an error
> 
> % ifconfig -a
> eon0: flags=3<UP,BROADCAST> mtu 1500
> ifconfig: link_status: ioctl: Invalid argument
> 
> % ifconfig eon0
> eon0: flags=3<UP,BROADCAST> mtu 1500
> ifconfig: link_status: ioctl: Invalid argument
> 
> % ifconfig ppp0
> ppp0: flags=8051<UP,POINTOPOINT,RUNNING,MULTICAST> mtu 1500
>         inet 10.34.129.29 -> 10.64.64.64 netmask 0xff000000
> 
> It seems to be related to the SIOCGLIFADDR in af_link.c but I don't know
> what that does. I have "options EON,ISO,TPIP" in my kernel config file
> (though I don't use the eon interface) left over from when I was working
> in the netiso code so perhaps its an incompatibility with the eon driver?

There is a bug in eonioctl().  In eonioctl(), change the return value
in the 'default:' case from EINVAL to ENOTTY.  This is just a stopgap fix.

Dave

--

-- 
David Young             OJC Technologies
dyoung <at> ojctech.com      Urbana, IL * (217) 278-3933 ext 24
(Continue reading)

Permalink | Reply | 0 comments |
headers
Iain Hibbert | 6 Aug 2008 00:41

Re: Socket options KPI

On Tue, 5 Aug 2008, Jason Thorpe wrote:

>
> On Jun 26, 2008, at 2:40 PM, Elad Efrat wrote:
>
> > On Thu, Jun 26, 2008 at 10:02 PM, Thor Lancelot Simon <tls <at> rek.tjls.com>
> > wrote:
> > > On Fri, Jan 18, 2008 at 07:08:37PM +0200, Elad Efrat wrote:
> > > > Hi,
> > > >
> > > > I have a patch to make a large portion of the network code use a new
> > > > "struct sockopt" (inspired by similar work done in FreeBSD, with input
> > > > from dyoung <at>  and yamt <at> ) instead of passing socket options in mbufs.
> > >
> > > What happened to this?  It looked like you got mostly positive comments...
> >
> > See http://mail-index.netbsd.org/tech-net/2008/05/02/msg000501.html.
>
> Let's get this checked in!

Sorry. I didn't want to check it in while nobody was around to fix any
problems (I was away for a month+, and Elad was busy)

In the meantime, we have moved to feature freeze and this is a large
change so I was holding off..  if we want it in 5.0 I can check it in
though?

iain
(Continue reading)

Permalink | Reply | 2 comments |

Gmane