date:20080314

Hello everyone,

I found out i couldn't disable my dhcp daemon yesterday and i think i've
traced the problem to this in /etc/rc:

if [ X${dhcpd_flags} != XNO -a -f /etc/dhcpd.conf ]; then

Now i have no clue what  -a -f  does (anyone care to point me to the right
manual?), but just setting dhcpd_flags=NO in rc.conf didn't disable it.
I had to rename the file dhcpd.conf file, so i guess the line does an OR
comparison instead of an AND comparison, anyone care to enlighten me or did
i actually find a minor bug?

Kind regards,

Erwin van Maanen
ACMEWeb I.S.

Re: dhcpd rc bug?

On Fri, Mar 14, 2008 at 08:15:51AM +0100, Erwin van Maanen wrote:

 Hello everyone,
 
 I found out i couldn't disable my dhcp daemon yesterday and i think i've
 traced the problem to this in /etc/rc:
 
   if [ X${dhcpd_flags} != XNO -a -f /etc/dhcpd.conf ]; then
 
 Now i have no clue what  -a -f  does (anyone care to point me to the right
 manual?), but just setting dhcpd_flags=NO in rc.conf didn't disable it.
 I had to rename the file dhcpd.conf file, so i guess the line does an OR
 comparison instead of an AND comparison, anyone care to enlighten me or did
 i actually find a minor bug?
 
 Kind regards,
 
 Erwin van Maanen
 ACMEWeb I.S.

That expression is right (see test(1)).

Where did you set dhcpd_flags to NO? In /etc/rc.conf? Do you perhaps
have another definition in /etc/rc.conf.local or something like that? 

-Otto

Re: dhcpd rc bug?

2008-03-14 Thread Richard Toohey


On 14/03/2008, at 8:15 PM, Erwin van Maanen wrote:


Hello everyone,

I found out i couldn't disable my dhcp daemon yesterday and i think  
i've

traced the problem to this in /etc/rc:

if [ X${dhcpd_flags} != XNO -a -f /etc/dhcpd.conf ]; then

Now i have no clue what  -a -f  does (anyone care to point me to  
the right
manual?), but just setting dhcpd_flags=NO in rc.conf didn't disable  
it.
I had to rename the file dhcpd.conf file, so i guess the line does  
an OR
comparison instead of an AND comparison, anyone care to enlighten  
me or did

i actually find a minor bug?

Kind regards,

Erwin van Maanen
ACMEWeb I.S.


man test

 -f file   True if file exists and is a regular file.

 expression1 -a expression2
   True if both expression1 and expression2 are true.

So I would read the logic to be ...

If dhcp_flags not eq NO AND the named file exists then do whatever ...

Which means - I think - that you should not have seen the behaviour  
that you did.


What EXACTLY did you put in rc.conf, and what did you do after changing
it (i.e. to get it to notice your change?)

And nothing relevant in rc.conf.local?

HTH a bit

Re: dhcpd rc bug?

Oeps :p

Indeed I had (in rc.conf and I have a .local too), forgot all about that I
decided to use the .local since it made updating easier...
Thanks for the rtfm hint :)

Just I more question if you don't mind answering it... which procress starts
the rc and rc.local, is that the kernel itself or some kind of process that
is started by the kernel.
No need for a full explenation ofcourse, just another rtfm hint would be
great tho :)

Erwin

-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of
Otto Moerbeek
Sent: vrijdag 14 maart 2008 8:29
To: Erwin van Maanen
Cc: misc@openbsd.org
Subject: Re: dhcpd rc bug?

On Fri, Mar 14, 2008 at 08:15:51AM +0100, Erwin van Maanen wrote:

 Hello everyone,
 
 I found out i couldn't disable my dhcp daemon yesterday and i think 
 i've traced the problem to this in /etc/rc:
 
   if [ X${dhcpd_flags} != XNO -a -f /etc/dhcpd.conf ]; then
 
 Now i have no clue what  -a -f  does (anyone care to point me to the 
 right manual?), but just setting dhcpd_flags=NO in rc.conf didn't disable
it.
 I had to rename the file dhcpd.conf file, so i guess the line does an 
 OR comparison instead of an AND comparison, anyone care to enlighten 
 me or did i actually find a minor bug?
 
 Kind regards,
 
 Erwin van Maanen
 ACMEWeb I.S.

That expression is right (see test(1)).

Where did you set dhcpd_flags to NO? In /etc/rc.conf? Do you perhaps have
another definition in /etc/rc.conf.local or something like that? 

-Otto

Re: [ and test (was dhcpd rc bug?)

2008-03-14 Thread Richard Toohey


On 14/03/2008, at 8:31 PM, Richard Toohey wrote:

On 14/03/2008, at 8:15 PM, Erwin van Maanen wrote:

if [ X${dhcpd_flags} != XNO -a -f /etc/dhcpd.conf ]; then
Now i have no clue what  -a -f  does (anyone care to point me to  
the right

manual?)

[cut]

man test
   -f file   True if file exists and is a regular file.
 expression1 -a expression2
   True if both expression1 and expression2 are true.



Apologies to all who know this already - but I wondered about that  
[ file

for ages until I was enlightened.

So, if you are wondering why man test ... at the top of the man page:

SYNOPSIS
 test expression
 [ expression ]

(See the square brackets?)  And if you look in /bin:

$ ls -l /bin
total 13652
-r-xr-xr-x  2 root  bin   79136 Aug 29  2007 [
[cut]
-r-xr-xr-x  2 root  bin   79136 Aug 29  2007 test

man [ will take you to the test man page.

See, for example:

http://developer.apple.com/documentation/OpenSource/Conceptual/ 
ShellScripting/shell_scripts/chapter_2_section_10.html#//apple_ref/ 
doc/uid/TP40004268-CH237-SW4

Re: dhcpd rc bug?

Thanks for your help :)

-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of
Richard Toohey
Sent: vrijdag 14 maart 2008 8:31
To: Erwin van Maanen
Cc: misc@openbsd.org
Subject: Re: dhcpd rc bug?

On 14/03/2008, at 8:15 PM, Erwin van Maanen wrote:

 Hello everyone,

 I found out i couldn't disable my dhcp daemon yesterday and i think 
 i've traced the problem to this in /etc/rc:

   if [ X${dhcpd_flags} != XNO -a -f /etc/dhcpd.conf ]; then

 Now i have no clue what  -a -f  does (anyone care to point me to the 
 right manual?), but just setting dhcpd_flags=NO in rc.conf didn't 
 disable it.
 I had to rename the file dhcpd.conf file, so i guess the line does an 
 OR comparison instead of an AND comparison, anyone care to enlighten 
 me or did i actually find a minor bug?

 Kind regards,

 Erwin van Maanen
 ACMEWeb I.S.

man test

  -f file   True if file exists and is a regular file.

  expression1 -a expression2
True if both expression1 and expression2 are true.

So I would read the logic to be ...

If dhcp_flags not eq NO AND the named file exists then do whatever ...

Which means - I think - that you should not have seen the behaviour that you
did.

What EXACTLY did you put in rc.conf, and what did you do after changing it
(i.e. to get it to notice your change?)

And nothing relevant in rc.conf.local?

HTH a bit

Re: dhcpd rc bug?

On Fri, Mar 14, 2008 at 08:59:07AM +0100, Erwin van Maanen wrote:

 Oeps :p
 
 Indeed I had (in rc.conf and I have a .local too), forgot all about that I
 decided to use the .local since it made updating easier...
 Thanks for the rtfm hint :)
 
 Just I more question if you don't mind answering it... which procress starts
 the rc and rc.local, is that the kernel itself or some kind of process that
 is started by the kernel.
 No need for a full explenation ofcourse, just another rtfm hint would be
 great tho :)

See init(8).

-Otto

 
 Erwin
 
 -Original Message-
 From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of
 Otto Moerbeek
 Sent: vrijdag 14 maart 2008 8:29
 To: Erwin van Maanen
 Cc: misc@openbsd.org
 Subject: Re: dhcpd rc bug?
 
 On Fri, Mar 14, 2008 at 08:15:51AM +0100, Erwin van Maanen wrote:
 
  Hello everyone,
  
  I found out i couldn't disable my dhcp daemon yesterday and i think 
  i've traced the problem to this in /etc/rc:
  
  if [ X${dhcpd_flags} != XNO -a -f /etc/dhcpd.conf ]; then
  
  Now i have no clue what  -a -f  does (anyone care to point me to the 
  right manual?), but just setting dhcpd_flags=NO in rc.conf didn't disable
 it.
  I had to rename the file dhcpd.conf file, so i guess the line does an 
  OR comparison instead of an AND comparison, anyone care to enlighten 
  me or did i actually find a minor bug?
  
  Kind regards,
  
  Erwin van Maanen
  ACMEWeb I.S.
 
 That expression is right (see test(1)).
 
 Where did you set dhcpd_flags to NO? In /etc/rc.conf? Do you perhaps have
 another definition in /etc/rc.conf.local or something like that? 
 
   -Otto

Re: dhcpd rc bug?

2008-03-14 Thread Richard Toohey


On 14/03/2008, at 8:59 PM, Erwin van Maanen wrote:
Just I more question if you don't mind answering it... which  
procress starts
the rc and rc.local, is that the kernel itself or some kind of  
process that

is started by the kernel.



Ummm,

man rc

DESCRIPTION
 rc is the command script that is invoked by init(8) during an  
automatic
 reboot and after single user mode is exited; it performs system  
house-
 keeping chores and starts up system daemons.  Additionally, rc  
is intri-
 cately tied to the netstart(8) script, which runs commands and  
daemons

 pertaining to the network.

And then man init, and so on.

Etc.

OpenBSD is renowned for its documentation, for a very good reason.

Re: [ and test (was dhcpd rc bug?)

Ah lol :)
That's why I couldn't figure it out, there is a binary called [

Guess those 2 filenames point to the same inode :)

-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of
Richard Toohey
Sent: vrijdag 14 maart 2008 9:02
To: Erwin van Maanen
Cc: OpenBSD-Misc Misc
Subject: Re: [ and test (was dhcpd rc bug?)

On 14/03/2008, at 8:31 PM, Richard Toohey wrote:
 On 14/03/2008, at 8:15 PM, Erwin van Maanen wrote:
  if [ X${dhcpd_flags} != XNO -a -f /etc/dhcpd.conf ]; then Now i 
 have no clue what  -a -f  does (anyone care to point me to the 
 right
 manual?)
[cut]
 man test
-f file   True if file exists and is a regular file.
  expression1 -a expression2
True if both expression1 and expression2 are true.


Apologies to all who know this already - but I wondered about that [ file
for ages until I was enlightened.

So, if you are wondering why man test ... at the top of the man page:

SYNOPSIS
  test expression
  [ expression ]

(See the square brackets?)  And if you look in /bin:

$ ls -l /bin
total 13652
-r-xr-xr-x  2 root  bin   79136 Aug 29  2007 [
[cut]
-r-xr-xr-x  2 root  bin   79136 Aug 29  2007 test

man [ will take you to the test man page.

See, for example:

http://developer.apple.com/documentation/OpenSource/Conceptual/
ShellScripting/shell_scripts/chapter_2_section_10.html#//apple_ref/
doc/uid/TP40004268-CH237-SW4

Re: dhcpd rc bug?

Thanks :) 
I'll try and readup on the misc a bit more as well, maybe I can provide some
assistance somewhere in return :)

-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of
Otto Moerbeek
Sent: vrijdag 14 maart 2008 9:09
To: Erwin van Maanen
Cc: misc@openbsd.org
Subject: Re: dhcpd rc bug?

On Fri, Mar 14, 2008 at 08:59:07AM +0100, Erwin van Maanen wrote:

 Oeps :p

 Indeed I had (in rc.conf and I have a .local too), forgot all about 
 that I decided to use the .local since it made updating easier...
 Thanks for the rtfm hint :)

 Just I more question if you don't mind answering it... which procress 
 starts the rc and rc.local, is that the kernel itself or some kind of 
 process that is started by the kernel.
 No need for a full explenation ofcourse, just another rtfm hint would 
 be great tho :)

See init(8).

-Otto

 Erwin

 -Original Message-
 From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf 
 Of Otto Moerbeek
 Sent: vrijdag 14 maart 2008 8:29
 To: Erwin van Maanen
 Cc: misc@openbsd.org
 Subject: Re: dhcpd rc bug?

 On Fri, Mar 14, 2008 at 08:15:51AM +0100, Erwin van Maanen wrote:

  Hello everyone,

  I found out i couldn't disable my dhcp daemon yesterday and i think 
  i've traced the problem to this in /etc/rc:

  if [ X${dhcpd_flags} != XNO -a -f /etc/dhcpd.conf ]; then

  Now i have no clue what  -a -f  does (anyone care to point me to 
  the right manual?), but just setting dhcpd_flags=NO in rc.conf 
  didn't disable
 it.
  I had to rename the file dhcpd.conf file, so i guess the line does 
  an OR comparison instead of an AND comparison, anyone care to 
  enlighten me or did i actually find a minor bug?

  Kind regards,

  Erwin van Maanen
  ACMEWeb I.S.

 That expression is right (see test(1)).

 Where did you set dhcpd_flags to NO? In /etc/rc.conf? Do you perhaps 
 have another definition in /etc/rc.conf.local or something like that?

   -Otto

Re: [ and test (was dhcpd rc bug?)

On Fri, Mar 14, 2008 at 09:02:19PM +1300, Richard Toohey wrote:

On 14/03/2008, at 8:31 PM, Richard Toohey wrote:
On 14/03/2008, at 8:15 PM, Erwin van Maanen wrote:
if [ X${dhcpd_flags} != XNO -a -f /etc/dhcpd.conf ]; then
Now i have no clue what -a -f does (anyone care to point me to the
right
manual?)
[cut]
man test
-f file True if file exists and is a regular file.
expression1 -a expression2
True if both expression1 and expression2 are true.

Apologies to all who know this already - but I wondered about that [ file
for ages until I was enlightened.

So, if you are wondering why man test ... at the top of the man page:

SYNOPSIS
test expression
[ expression ]

(See the square brackets?) And if you look in /bin:

$ ls -l /bin
total 13652
-r-xr-xr-x 2 root bin 79136 Aug 29 2007 [
[cut]
-r-xr-xr-x 2 root bin 79136 Aug 29 2007 test

man [ will take you to the test man page.

See, for example:

http://developer.apple.com/documentation/OpenSource/Conceptual/ShellScripting/shell_scripts/chapter_2_section_10.html#//apple_ref/doc/uid/TP40004268-CH237-SW4

Some background...

Originally, the shell had vitually no builtin commands, all processing
was done by external commands. Even loops were implemented by forking
a child that interpreted the condition and moved the seek position in
the shell script being executed. That could be done since fd's were
shared.

Of course that was all dog slow, and that's why more and more commands
are now executed as builtins. External version of the commands still
remain, though.

-Otto

Re: dhcrelay on carp interface (above vlan)

2008-03-14 Thread Falk Brockerhoff - smartTERRA GmbH


Am 14.03.2008 um 08:13 schrieb Marc Balmer:


Falk Brockerhoff - smartTERRA GmbH wrote:

I think a good solutions is to look if the given interface is a  
carp interface and to figure out the carpdev interface. Then this  
can be used to listen on. But my programming skills are really  
poor, else I would provide a patch...


you can provide the interface name on the command line using -i:

e.g. carp0 carpdev vr0


Yes, I know. But I have to provide a numbered interface. In this case  
the carp interface. This results in have dhcprelay listening on this  
carp interface, too. But it have to listen on the vlan (in your  
example the physical interface vr0) interface to catch the dhcp  
request. That's my problem :-)


Regards,

Falk Brockerhoff

Problem with libiconv-1.9.2p3 on 4.2

2008-03-14 Thread Eric Pancer

There seems to be a problem with the libiconv-1.9.2p3 package; I took it
from the main FTP server, as well as several mirrors and had this problem.


$ date
Fri Mar 14 03:47:34 CDT 2008

$ uname -a
OpenBSD foo.example.org 4.2 GENERIC#375 i386

$ sudo pkg_add gettext-0.14.6p0.tgz
Can't install gettext-0.14.6p0: lib not found expat.8.0
Dependencies for gettext-0.14.6p0 resolve to: libiconv-1.9.2p3
Full dependency tree is libiconv-1.9.2p3

$ md5 libiconv-1.9.2p3.tgz
MD5 (libiconv-1.9.2p3.tgz) = e0c719123bc569b450898b20c910cd46

$ pkg_info -L libiconv-1.9.2p3.tgz
Information for file:./libiconv-1.9.2p3.tgz

Files:
/usr/local/lib/libcharset.so.1.0
/usr/local/lib/libiconv.so.4.0
/usr/local/bin/iconv
/usr/local/include/iconv.h
/usr/local/include/libcharset.h
/usr/local/include/localcharset.h
/usr/local/lib/charset.alias
/usr/local/lib/libcharset.a
/usr/local/lib/libcharset.la
/usr/local/lib/libiconv.a
/usr/local/lib/libiconv.la
/usr/local/man/man1/iconv.1
/usr/local/man/man3/iconv.3
/usr/local/man/man3/iconv_close.3
/usr/local/man/man3/iconv_open.3
/usr/local/share/doc/libiconv/iconv.1.html
/usr/local/share/doc/libiconv/iconv.3.html
/usr/local/share/doc/libiconv/iconv_close.3.html
/usr/local/share/doc/libiconv/iconv_open.3.html

Re: Problem with libiconv-1.9.2p3 on 4.2

2008-03-14 Thread Nicolas Letellier


Hi.

On 4.2, you must install xbase set (which contains lib expat).
Or waint for 4.3, this problem is fixed.

- Nicolas.


Eric Pancer a icrit :

There seems to be a problem with the libiconv-1.9.2p3 package; I took it
from the main FTP server, as well as several mirrors and had this problem.


$ date
Fri Mar 14 03:47:34 CDT 2008

$ uname -a
OpenBSD foo.example.org 4.2 GENERIC#375 i386

$ sudo pkg_add gettext-0.14.6p0.tgz
Can't install gettext-0.14.6p0: lib not found expat.8.0
Dependencies for gettext-0.14.6p0 resolve to: libiconv-1.9.2p3
Full dependency tree is libiconv-1.9.2p3

$ md5 libiconv-1.9.2p3.tgz
MD5 (libiconv-1.9.2p3.tgz) = e0c719123bc569b450898b20c910cd46

$ pkg_info -L libiconv-1.9.2p3.tgz
Information for file:./libiconv-1.9.2p3.tgz

Files:
/usr/local/lib/libcharset.so.1.0
/usr/local/lib/libiconv.so.4.0
/usr/local/bin/iconv
/usr/local/include/iconv.h
/usr/local/include/libcharset.h
/usr/local/include/localcharset.h
/usr/local/lib/charset.alias
/usr/local/lib/libcharset.a
/usr/local/lib/libcharset.la
/usr/local/lib/libiconv.a
/usr/local/lib/libiconv.la
/usr/local/man/man1/iconv.1
/usr/local/man/man3/iconv.3
/usr/local/man/man3/iconv_close.3
/usr/local/man/man3/iconv_open.3
/usr/local/share/doc/libiconv/iconv.1.html
/usr/local/share/doc/libiconv/iconv.3.html
/usr/local/share/doc/libiconv/iconv_close.3.html
/usr/local/share/doc/libiconv/iconv_open.3.html

Re: [ and test (was dhcpd rc bug?)

2008-03-14 Thread Alexander Hall


Otto Moerbeek wrote:

On Fri, Mar 14, 2008 at 09:02:19PM +1300, Richard Toohey wrote:


On 14/03/2008, at 8:31 PM, Richard Toohey wrote:

On 14/03/2008, at 8:15 PM, Erwin van Maanen wrote:

if [ X${dhcpd_flags} != XNO -a -f /etc/dhcpd.conf ]; then
Now i have no clue what  -a -f  does (anyone care to point me to the 
right

manual?)

[cut]

man test
   -f file   True if file exists and is a regular file.
 expression1 -a expression2
   True if both expression1 and expression2 are true.


Apologies to all who know this already - but I wondered about that [ file
for ages until I was enlightened.

So, if you are wondering why man test ... at the top of the man page:

SYNOPSIS
 test expression
 [ expression ]

(See the square brackets?)  And if you look in /bin:

$ ls -l /bin
total 13652
-r-xr-xr-x  2 root  bin   79136 Aug 29  2007 [
[cut]
-r-xr-xr-x  2 root  bin   79136 Aug 29  2007 test

man [ will take you to the test man page.

See, for example:

http://developer.apple.com/documentation/OpenSource/Conceptual/ShellScripting/shell_scripts/chapter_2_section_10.html#//apple_ref/doc/uid/TP40004268-CH237-SW4


Some background...

Originally, the shell had vitually no builtin commands, all processing
was done by external commands. Even loops were implemented by forking
a child that interpreted the condition and moved the seek position in
the shell script being executed. That could be done since fd's were
shared. 


Of course that was all dog slow, and that's why more and more commands
are now executed as builtins. External version of the commands still
remain, though.


... which means that people should actually have pointed the OP top 
ksh(1) rather than test(1) or [(1)... :-)


/Alexander

Why Sendmail?

2008-03-14 Thread Gustavo Polillo

Why is sendmail a default smtp server and not is postfix or qmail?

Re: Why Sendmail?

On Fri, Mar 14, 2008 at 10:07:49AM -0300, Gustavo Polillo wrote:

 Why is sendmail a default smtp server and not is postfix or qmail?

What are people not reading the FAQ?

-Otto

Re: Why Sendmail?

2008-03-14 Thread Martin Schröder

2008/3/14, Gustavo Polillo [EMAIL PROTECTED]:
 Why is sendmail a default smtp server and not is postfix or qmail?

http://www.openbsd.org/faq/faq1.html#HowAbout

Re: Why Sendmail?

2008-03-14 Thread [EMAIL PROTECTED]

On Fri, Mar 14, 2008 at 9:07 AM, Gustavo Polillo [EMAIL PROTECTED] wrote:
 Why is sendmail a default smtp server and not is postfix or qmail?



For several very good reasons including tradition, licensing, sendmail
being well suited for the job, and the people in charge liking
sendmail.

Re: Why Sendmail?

2008-03-14 Thread Dave Wilson


Gustavo Polillo wrote:

Why is sendmail a default smtp server and not is postfix or qmail?

  
Why is it that this message got attatched to an existing thread in my 
mail client?


Ah. Because it has the exact same subject line, and covers the exact 
same subject. Funny that.


To make this email have at least *some* value, I direct the OP to 
http://marc.info/?l=openbsd-miscm=116428904604885w=2



The thing that I always wonder at, is that the same people that are 
perfectly capable of doing enough research to find out which list to 
post to, and how, are the people who completely fail to read the advice 
on where to look before posting. Oh well, back to work.


Si1entDave

Re: Why Sendmail?

2008-03-14 Thread Gustavo Polillo

oh! sorry.. thanks... :|

2008/3/14, Martin Schrvder [EMAIL PROTECTED]:
 2008/3/14, Gustavo Polillo [EMAIL PROTECTED]:

  Why is sendmail a default smtp server and not is postfix or qmail?


 http://www.openbsd.org/faq/faq1.html#HowAbout

Re: Why Sendmail?

2008-03-14 Thread Alexander Hall


Otto Moerbeek wrote:

Why is sendmail a default smtp server and not is postfix or qmail?


What are people not reading the FAQ?


Because then they would have no excuse for not searching the mailing 
list archives before posting?

Re: Trying to get usbnet working to gumstix

2008-03-14 Thread Unix Fan

Sean Kennedy wrote:

 For what it's worth, I'm interested too in a tech@ tutorial on (How to add

 Unknown, or Semi-Known USB) devices.

 

 I have had success with adding commonly defined things (Keyboards Mice, and

 the occasional USB wireless/wired Network adapter)

 

 But for something that has little or no description, (and no blobs) and

 FreeBSD or NetBSD support (L*nux documentation too, but avoiding GPL)

 It would be nice to see if I could add in USB support.

 

 There really is no RT*M reference I could find at the best of times, to direct

 where to begin.

 

 -sean



Dmitri Alenitchev wrote a few articles once upon a time, I keep them 
bookmarked.. not sure where he's at these days though. ;)



http://allroot.blogspot.com/2006/10/hacking-usb-device-drivers-part-1.html

http://allroot.blogspot.com/2006/10/hacking-usb-device-drivers-part-2.html







-Nix Fan.

OpenBSD + python + cron

2008-03-14 Thread Stuart VanZee

Hello everyone,

I have a python script that I have written that uses
the GnuPGInterface module to encrypt and sign some
files.  It works great when I run it from a command
prompt but when I set it to run via cron it errors
out.  Here is a copy of the traceback:

Traceback (most recent call last):
  File /usr/local/sbin/svfil.py, line 269, in module
main()
  File /usr/local/sbin/svfil.py, line 234, in main
encrypt( CONST_workspace + workfname, CONST_key)
  File /usr/local/sbin/svfil.py, line 30, in encrypt
p1.wait()
  File /usr/local/lib/python2.5/site-packages/GnuPGInterface.py, line 639,
in wait
raise IOError, GnuPG exited non-zero, with code %d % (e  8)
IOError: GnuPG exited non-zero, with code 35584

I suspect that when it is run via cron it is not being
able to access the keyring like it is when I run this
via the command prompt, but I don't know why that would
be.  When I run it via the command prompt I su to root
and when it is run via cron it is run by a job in root's
crontab.

Stuart van Zee
[EMAIL PROTECTED]

Re: OpenBSD + python + cron

2008-03-14 Thread Mike Erdely

On Fri, Mar 14, 2008 at 10:43:31AM -0400, Stuart VanZee wrote:
 I have a python script that I have written that uses
 the GnuPGInterface module to encrypt and sign some
 files.  It works great when I run it from a command
 prompt but when I set it to run via cron it errors
 out.  Here is a copy of the traceback:

snip

Stuart,

Try putting this in your crontab:

x y * * * env PATH=${PATH}:/usr/local/sbin:/usr/local/bin 
/usr/local/sbin/your.py
The default crontab PATH does not include the local s?bin directories.

-ME

Re: ip(4) still says IP_MAX_MEMBERSHIPS is 20

2008-03-14 Thread Jason McIntyre

On Thu, Mar 13, 2008 at 09:24:15AM -0700, Matthew Dempsky wrote:
 ip(4) says that IP_MAX_MEMBERSHIPS is 20, but it was bumped to 4095 in
 rev 1.73 of netinet/in.h.
 

fixed, thanks (the change happened in -r1.74 though).
jmc

 Index: ip.4
 ===
 RCS file: /cvs/src/share/man/man4/ip.4,v
 retrieving revision 1.26
 diff -p -u -r1.26 ip.4
 --- ip.431 May 2007 19:19:50 -  1.26
 +++ ip.413 Mar 2008 16:12:10 -
 @@ -310,7 +310,7 @@ programs running on multihomed hosts may
  join the same group on more than one interface.
  Up to
  .Dv IP_MAX_MEMBERSHIPS
 -(currently 20) memberships may be added on a
 +(currently 4095) memberships may be added on a
  single socket.
  .Pp
  To drop a membership, use:

Re: ip(4) still says IP_MAX_MEMBERSHIPS is 20

2008-03-14 Thread Matthew Dempsky

On 3/14/08, Jason McIntyre [EMAIL PROTECTED] wrote:
 fixed, thanks (the change happened in -r1.74 though).

D'oh, you're right. :-)

gettimeofday() dramatical slowdown from 4.1-4.2

Upon trying to locate an unexplained, massive performance reduction
when switching host for a number of applications from obsd 4.1 to 4.2,
I found that it seems gettimeofday() has taken a nosedive in
performance as of openbsd 4.2.

A very blunt test confirmed it; however, I'm not sure wherein the
process of gettimeofday() this happens. I can only imagine the
performance issues this has lead to in environments that do frequent
time-polling (heavily burdened webservers come to mind).

http://pastebin.com/m311250a6

Re: gettimeofday() dramatical slowdown from 4.1-4.2

2008-03-14 Thread Paul de Weerd

On Fri, Mar 14, 2008 at 06:08:42PM +0100, stolendata.net wrote:
| Upon trying to locate an unexplained, massive performance reduction
| when switching host for a number of applications from obsd 4.1 to 4.2,
| I found that it seems gettimeofday() has taken a nosedive in
| performance as of openbsd 4.2.
| 
| A very blunt test confirmed it; however, I'm not sure wherein the
| process of gettimeofday() this happens. I can only imagine the
| performance issues this has lead to in environments that do frequent
| time-polling (heavily burdened webservers come to mind).
| 
| http://pastebin.com/m311250a6

Have you tried 4.3 / -current yet ? 

My 3GHz amd64 machine running 4.3-ish : 1.675s
My 440MHz sparc64 machine running 4.2 : 3.512s

So it doesn't seem to affect all platforms (although I have nothing to
compare it against for my sparc64).

Cheers,

Paul 'WEiRD' de Weerd

-- 
[++-]+++.+++[---].+++[+
+++-].++[-]+.--.[-]
 http://www.weirdnet.nl/

kernel trap in 4.3

2008-03-14 Thread Jose Fragoso

Hi,

Following a suggestion from a misc member after I complained
about slow IO on a IBM xSeries 336 (see 'write cache on scsi'),
I tried to install a snapshot.

Except for the very slow filesystems creation, the install
process went through ok.

But when I tried to boot the newly installed machine, I got
a kernel trap and was sent to ddb

bios: IBM eServer xSeries 336 -[883721U]-
acpi0 at bios0: rev 0
acpi0: tables DSDT FACP APIC MCFG
acpi0: wakeup device PCI0(S5)
acpitimer0 at acpi0: 3579545 Hz, 24bits
acpiprt0 at acpi0uvm_fault(0xd07ca0c0, 0xd1977000, 0, 3) - 3
kernel: paga fault trap code=0
Stopped at bcopy+0x1a: repe movsl(%esi),%es:(%edi)
bcopy(d1972684,73,d1973910,ac) at bcopy+0x1a
aml_parseop(d1972684,d1973910,74) at aml_parseop+0xe6
aml_parseterm(d1972684,d092c8d0,390,d07a41d0,d1972684) at aml_parseterm+0x2c
aml_callmethod(d1972684,d092c8d0,d198295f,d1965984,d077fb42) at
aml_callmethod+0x26
aml_evalmethod(0,d1965984,0,0,d092c8d0) at aml_evalmethod+0x41
aml_evalnode(d1960e00,d1965984,0,0,d092c8d0,d092c8d0,d092c8e8,d0673662) at
aml_evalnode+0xc7
acpiprit_getpcipus(d19726c0,d1965604,d092c958,d0673662,d1965384) at
acpiprt_getpcibus+0x30
acpiprit_getpcipus(d19726c0,d1965484,64,d1955ef0,d092ca6c) at
acpiprt_getpcibus+0xcf
acpiprit_getpcipus(d19726c0,d1971d04,d092ca58,d1955f00,0) at
acpiprt_getpcibus+0xcf
acpiprit_attach(d1960e00,d19726c0,d092cae0,d1960e00,d1960e00) at
acpiprt_attach+0x22

I did try to save theses messages in the dmesg and copy and paste
them afterwords but I did not manage, I this was all written down
and then typed in here.

The output of ps looks like this:

ddb ps

  PID   PPID   PGRP   UIDS  FLAGS   WAITCOMMAND
*  0 -1 0  0 7  0x80200  swappper

The output of trace looks like this:

ddb trace
bcopy(d1972684,73,d1973910,ac) at bcopy+0x1a
aml_parseop(d1972684,d1973910,74) at aml_parseop+0xe6
aml_parseterm(d1972684,d092c8d0,390,d07a41d0,d1972684) at aml_parseterm+0x2c
aml_callmethod(d1972684,d092c8d0,d198295f,d1965984,d077fb42) at
aml_callmethod+0x26
aml_evalmethod(0,d1965984,0,0,d092c8d0) at aml_evalmethod+0x41
aml_evalnode(d1960e00,d1965984,0,0,d092c8d0,d092c8d0,d092c8e8,d0673662) at
aml_evalnode+0xc7
acpiprit_getpcipus(d19726c0,d1965604,d092c958,d0673662,d1965384) at
acpiprt_getpcibus+0x30
acpiprit_getpcipus(d19726c0,d1965484,64,d1955ef0,d092ca6c) at
acpiprt_getpcibus+0xcf
acpiprit_getpcipus(d19726c0,d1971d04,d092ca58,d1955f00,0) at
acpiprt_getpcibus+0xcf
acpiprit_attach(d1960e00,d19726c0,d092cae0,d1960e00,d1960e00) at
acpiprt_attach+0x22
config_attach(d1960e00,d07852c8,d092cae0,d06713c4) at config_attach+0xf0
aci_foundprt(d1971d04,d1960e00,d0670bc8,d1960e00,0) at acpi_foundprt+0x95
aml_find_node(d1965384,d077e3b7,d0670bc8,d1960e00) at aml_find_node+0x6e
aml_find_node(d1965504,d077e3b7,d0670bc8,d1960e00) at aml_find_node+0x5f
aml_find_node(d195f5c4,d077e3b7,d0670bc8,d1960e00) at aml_find_node+0x5f
aml_find_node(d195fc84,d077e3b7,d0670bc8,d1960e00) at aml_find_node+0x5f
aml_find_node(d1965384,d077e3b7,d0670bc8,d1960e00,d195fec4,d077e3b2,d0670b40,
d1960e00) at aml_find_node+0x5f
acpi_attach(d195ff80,d1960e00,d092cd50,d195ff80,0) at acpi_attach+0x431
config_attach(d195ff80,d0785184,d092cd50,d0603378) at config_attach+0xfd
biosattach(d195ffc0,d195ff80,d092ce80,d195ffc0,d0202251) at biosattach+0x353
config_attach(d195ffc0,d07843e0,d092ce80,d04a4d80,d06d26f8) at
config_attach+0xfd
mainbus_attach(0,d195ffc0,0,de701000,d092b334) at mainbus_attach+0x3d
config_attach(0,d0781d34,0,0,0) at config_attach+0xfd
config_rootfound(d06d0f6b,0,d092cf38,d0478826) at config_rootfound+0x27
cpu_configure(d0898ca0,1,3,0,2) at cpu_configure+0x29
main(0,0,0,0,0) at main+0x38a

The only way I managed to boot the machine was to disable acpi
at the ukc prompt.

But then again, I did not solve my original problem. A simple
command like 'mv src.tar.gz ..' takes more than 10 seconds
to execute.

Here is the output of top | cat while the mv is going on.

load averages:  0.35,  0.18,  0.1215:06:28
22 processes:  21 idle, 1 on processor
CPU states:  0.0% user,  0.0% nice,  0.0% system,  0.0% interrupt,  100% idle
Memory: Real: 8416K/144M act/tot  Free: 856M  Swap: 0K/1024M used/tot

  PID USERNAME PRI NICE  SIZE   RES STATEWAIT  TIMECPU COMMAND
 8033 root  -50  436K  148K sleepgetblk0:00  0.05% mv
18074 root   20 1056K 1848K sleepselect0:01  0.00% sendmail
12787 root   20  692K  840K idle select0:00  0.00% cron
20289 ell20 3372K 1900K sleepselect0:00  0.00% sshd
16301 _syslogd   20  620K  776K sleeppoll  0:00  0.00% syslogd
 4170 root   20 3336K 2428K idle netio 0:00  0.00% sshd
10977 root   20 3368K 2372K idle netio 0:00  0.00% sshd
12611 root   30  664K  512K idle ttyin 0:00  0.00% ksh
25086 root  180  528K  512K sleeppause 0:00  0.00% ksh
21668 ell20 3304K 1932K sleepselect0:00  0.00%

Re: gettimeofday() dramatical slowdown from 4.1-4.2

4.3-snapshot of today: 8.0sec   on same 1.83ghz C2D

-SD

On Fri, Mar 14, 2008 at 7:01 PM, Paul de Weerd [EMAIL PROTECTED] wrote:

 On Fri, Mar 14, 2008 at 06:08:42PM +0100, stolendata.net wrote:
  | Upon trying to locate an unexplained, massive performance reduction
  | when switching host for a number of applications from obsd 4.1 to 4.2,
  | I found that it seems gettimeofday() has taken a nosedive in
  | performance as of openbsd 4.2.
  |
  | A very blunt test confirmed it; however, I'm not sure wherein the
  | process of gettimeofday() this happens. I can only imagine the
  | performance issues this has lead to in environments that do frequent
  | time-polling (heavily burdened webservers come to mind).
  |
  | http://pastebin.com/m311250a6

  Have you tried 4.3 / -current yet ?

  My 3GHz amd64 machine running 4.3-ish : 1.675s
  My 440MHz sparc64 machine running 4.2 : 3.512s

  So it doesn't seem to affect all platforms (although I have nothing to
  compare it against for my sparc64).

  Cheers,

  Paul 'WEiRD' de Weerd

  --
  [++-]+++.+++[---].+++[+
  +++-].++[-]+.--.[-]
  http://www.weirdnet.nl/

Re: kernel trap in 4.3

2008-03-14 Thread Chris Cappuccio

In cases where ACPI doesn't work, on some machines, the acpi information is
faulty and openbsd has to work around it (like on some Intel branded 
motherboards.)  On other machines, the acpi tables are valid but expose
bugs in the openbsd parser.  The first step towards resolution would be
to run 'acpidump' and submit it to OpenBSD via sendbug along with dmesg
and this back trace.

Jose Fragoso [EMAIL PROTECTED] wrote:
 Hi,
 
 Following a suggestion from a misc member after I complained
 about slow IO on a IBM xSeries 336 (see 'write cache on scsi'),
 I tried to install a snapshot.
 
 Except for the very slow filesystems creation, the install
 process went through ok.
 
 But when I tried to boot the newly installed machine, I got
 a kernel trap and was sent to ddb
 
 bios: IBM eServer xSeries 336 -[883721U]-
 acpi0 at bios0: rev 0
 acpi0: tables DSDT FACP APIC MCFG
 acpi0: wakeup device PCI0(S5)
 acpitimer0 at acpi0: 3579545 Hz, 24bits
 acpiprt0 at acpi0uvm_fault(0xd07ca0c0, 0xd1977000, 0, 3) - 3
 kernel: paga fault trap code=0
 Stopped at bcopy+0x1a: repe movsl(%esi),%es:(%edi)
 bcopy(d1972684,73,d1973910,ac) at bcopy+0x1a
 aml_parseop(d1972684,d1973910,74) at aml_parseop+0xe6
 aml_parseterm(d1972684,d092c8d0,390,d07a41d0,d1972684) at aml_parseterm+0x2c
 aml_callmethod(d1972684,d092c8d0,d198295f,d1965984,d077fb42) at
 aml_callmethod+0x26
 aml_evalmethod(0,d1965984,0,0,d092c8d0) at aml_evalmethod+0x41
 aml_evalnode(d1960e00,d1965984,0,0,d092c8d0,d092c8d0,d092c8e8,d0673662) at
 aml_evalnode+0xc7
 acpiprit_getpcipus(d19726c0,d1965604,d092c958,d0673662,d1965384) at
 acpiprt_getpcibus+0x30
 acpiprit_getpcipus(d19726c0,d1965484,64,d1955ef0,d092ca6c) at
 acpiprt_getpcibus+0xcf
 acpiprit_getpcipus(d19726c0,d1971d04,d092ca58,d1955f00,0) at
 acpiprt_getpcibus+0xcf
 acpiprit_attach(d1960e00,d19726c0,d092cae0,d1960e00,d1960e00) at
 acpiprt_attach+0x22
 
 I did try to save theses messages in the dmesg and copy and paste
 them afterwords but I did not manage, I this was all written down
 and then typed in here.
 
 The output of ps looks like this:
 
 ddb ps
 
   PID   PPID   PGRP   UIDS  FLAGS   WAITCOMMAND
 *  0 -1 0  0 7  0x80200  swappper
 
 The output of trace looks like this:
 
 ddb trace
 bcopy(d1972684,73,d1973910,ac) at bcopy+0x1a
 aml_parseop(d1972684,d1973910,74) at aml_parseop+0xe6
 aml_parseterm(d1972684,d092c8d0,390,d07a41d0,d1972684) at aml_parseterm+0x2c
 aml_callmethod(d1972684,d092c8d0,d198295f,d1965984,d077fb42) at
 aml_callmethod+0x26
 aml_evalmethod(0,d1965984,0,0,d092c8d0) at aml_evalmethod+0x41
 aml_evalnode(d1960e00,d1965984,0,0,d092c8d0,d092c8d0,d092c8e8,d0673662) at
 aml_evalnode+0xc7
 acpiprit_getpcipus(d19726c0,d1965604,d092c958,d0673662,d1965384) at
 acpiprt_getpcibus+0x30
 acpiprit_getpcipus(d19726c0,d1965484,64,d1955ef0,d092ca6c) at
 acpiprt_getpcibus+0xcf
 acpiprit_getpcipus(d19726c0,d1971d04,d092ca58,d1955f00,0) at
 acpiprt_getpcibus+0xcf
 acpiprit_attach(d1960e00,d19726c0,d092cae0,d1960e00,d1960e00) at
 acpiprt_attach+0x22
 config_attach(d1960e00,d07852c8,d092cae0,d06713c4) at config_attach+0xf0
 aci_foundprt(d1971d04,d1960e00,d0670bc8,d1960e00,0) at acpi_foundprt+0x95
 aml_find_node(d1965384,d077e3b7,d0670bc8,d1960e00) at aml_find_node+0x6e
 aml_find_node(d1965504,d077e3b7,d0670bc8,d1960e00) at aml_find_node+0x5f
 aml_find_node(d195f5c4,d077e3b7,d0670bc8,d1960e00) at aml_find_node+0x5f
 aml_find_node(d195fc84,d077e3b7,d0670bc8,d1960e00) at aml_find_node+0x5f
 aml_find_node(d1965384,d077e3b7,d0670bc8,d1960e00,d195fec4,d077e3b2,d0670b40,
 d1960e00) at aml_find_node+0x5f
 acpi_attach(d195ff80,d1960e00,d092cd50,d195ff80,0) at acpi_attach+0x431
 config_attach(d195ff80,d0785184,d092cd50,d0603378) at config_attach+0xfd
 biosattach(d195ffc0,d195ff80,d092ce80,d195ffc0,d0202251) at biosattach+0x353
 config_attach(d195ffc0,d07843e0,d092ce80,d04a4d80,d06d26f8) at
 config_attach+0xfd
 mainbus_attach(0,d195ffc0,0,de701000,d092b334) at mainbus_attach+0x3d
 config_attach(0,d0781d34,0,0,0) at config_attach+0xfd
 config_rootfound(d06d0f6b,0,d092cf38,d0478826) at config_rootfound+0x27
 cpu_configure(d0898ca0,1,3,0,2) at cpu_configure+0x29
 main(0,0,0,0,0) at main+0x38a
 
 The only way I managed to boot the machine was to disable acpi
 at the ukc prompt.
 
 But then again, I did not solve my original problem. A simple
 command like 'mv src.tar.gz ..' takes more than 10 seconds
 to execute.
 
 Here is the output of top | cat while the mv is going on.
 
 load averages:  0.35,  0.18,  0.1215:06:28
 22 processes:  21 idle, 1 on processor
 CPU states:  0.0% user,  0.0% nice,  0.0% system,  0.0% interrupt,  100% idle
 Memory: Real: 8416K/144M act/tot  Free: 856M  Swap: 0K/1024M used/tot
 
   PID USERNAME PRI NICE  SIZE   RES STATEWAIT  TIMECPU COMMAND
  8033 root  -50  436K  148K sleepgetblk0:00  0.05% mv
 18074 root   20 1056K 1848K sleepselect0:01  0.00% sendmail
 12787 root   20  692K  840K idle

Re: gettimeofday() dramatical slowdown from 4.1-4.2

2008-03-14 Thread Unix Fan

$ ./time



100 calls to gettimeofday() ... 4.503s



$ uname -srp

OpenBSD 4.2 AMD Athlon(TM) XP 2600+ (AuthenticAMD 686-class, 512KB L2 cache)

$



Seems fine here, looks like the error is on your end.. ;)



Have you tested on 4.3/snapshots.. perhaps enabling/disabling acpi..  etc?







-Nix Fan.

Re: gettimeofday() dramatical slowdown from 4.1-4.2

On Fri, Mar 14, 2008 at 06:08:42PM +0100, stolendata.net wrote:

 Upon trying to locate an unexplained, massive performance reduction
 when switching host for a number of applications from obsd 4.1 to 4.2,
 I found that it seems gettimeofday() has taken a nosedive in
 performance as of openbsd 4.2.
 
 A very blunt test confirmed it; however, I'm not sure wherein the
 process of gettimeofday() this happens. I can only imagine the
 performance issues this has lead to in environments that do frequent
 time-polling (heavily burdened webservers come to mind).
 
 http://pastebin.com/m311250a6

This is likely the effect of the new timecounter code. iirc the old
code just reads a memory location while the new code actually reads a
timer and does some processing. The main reason for the switch to
timecounters is to reduce the differences between archs. Timecounters
are also much better to avoid clock problems on SMP systems, and they
can switch to different clock sourcdes depending on the hardware
available. 

-Otto

Re: gettimeofday() dramatical slowdown from 4.1-4.2

The question is, how long would that take on the same hardware but on
4.1? :-) My guess is approx. 16 times less time.

I have now tested this on a third machine, a 1.9ghz Sempron LE-1100,
on both 4.1, 4.2 and 4.3 (all i386 dist), and the result is the same;
approx. 16 times slower gettimeofday() on 4.2 and 4.3.

Unix Fan's explanation is surely the reason, but it still leaves the
problem as a fact - a quite performance impairing fact.

Maybe someone will take a look at it and find a way to improve it. The
scenarios that are affected are numerous.

-SD

On 14 Mar 2008 12:53:06 -0700, Unix Fan [EMAIL PROTECTED] wrote:
 $ ./time



  100 calls to gettimeofday() ... 4.503s



  $ uname -srp

  OpenBSD 4.2 AMD Athlon(TM) XP 2600+ (AuthenticAMD 686-class, 512KB L2 
 cache)

  $



  Seems fine here, looks like the error is on your end.. ;)



  Have you tested on 4.3/snapshots.. perhaps enabling/disabling acpi..  etc?







  -Nix Fan.

Re: gettimeofday() dramatical slowdown from 4.1-4.2

2008-03-14 Thread Stuart Henderson

Whichever timer hardware your system is using (you can see with
'sysctl kern.timecounter') seems a bit on the slow side, my 1200MHz
X40 runs your test program in 2.9s.

$ sysctl kern.timecounter
kern.timecounter.tick=1
kern.timecounter.timestepwarnings=0
kern.timecounter.hardware=ICHPM
kern.timecounter.choice=i8254(0) ICHPM(1000) dummy(-100)

Have you compared bsd with bsd.mp? We don't even know what code you
run, or what hardware you have, there's no dmesg...

As an aside, tc_init(9) is a good starting point if you want to
learn about the timecounter code.

Re: gettimeofday() dramatical slowdown from 4.1-4.2

my 4.1 and 4.2 machines are -stable, and all are running i386, clearly.

My surprise and question was all in the fact that this changed from
4.1 to 4.2, and WHY it changed from 4.1 to 4.2. Otto Moerbeek has
already explained that there was a change in the timecounter code, and
your addition puts the rest in clear light and verifies that there is
no change possible for getting the performance back.

-SD

On Fri, Mar 14, 2008 at 10:03 PM, Stuart Henderson [EMAIL PROTECTED] wrote:
 Whichever timer hardware your system is using (you can see with
  'sysctl kern.timecounter') seems a bit on the slow side, my 1200MHz
  X40 runs your test program in 2.9s.

  $ sysctl kern.timecounter
  kern.timecounter.tick=1
  kern.timecounter.timestepwarnings=0
  kern.timecounter.hardware=ICHPM
  kern.timecounter.choice=i8254(0) ICHPM(1000) dummy(-100)

  Have you compared bsd with bsd.mp? We don't even know what code you
  run, or what hardware you have, there's no dmesg...

  As an aside, tc_init(9) is a good starting point if you want to
  learn about the timecounter code.

Re: gettimeofday() dramatical slowdown from 4.1-4.2

2008-03-14 Thread Henning Brauer

* stolendata.net [EMAIL PROTECTED] [2008-03-14 22:34]:
 my 4.1 and 4.2 machines are -stable, and all are running i386, clearly.
 
 My surprise and question was all in the fact that this changed from
 4.1 to 4.2, and WHY it changed from 4.1 to 4.2. Otto Moerbeek has
 already explained that there was a change in the timecounter code, and
 your addition puts the rest in clear light and verifies that there is
 no change possible for getting the performance back.

well, it might be possible to speed up the code path gettimeofday() 
uses in the timecounter code. of course, that requires reading that code, 
and maybe profiling.

-- 
Henning Brauer, [EMAIL PROTECTED], [EMAIL PROTECTED]
BS Web Services, http://bsws.de
Full-Service ISP - Secure Hosting, Mail and DNS Services
Dedicated Servers, Rootservers, Application Hosting - Hamburg  Amsterdam

Re: Installation freeze....

2008-03-14 Thread Massimiliano Giorgi

Dave Cottle wrote:
 On 10/03/2008, Massimiliano Giorgi [EMAIL PROTECTED] wrote:
 It is working! (two days)
 I have modified the bios settings to reserve the irq 9 and now all the
 ethernet irqs are dispatched to irq 11.
 I don't know why the irq 9 is bad for the Intel PRO/1000MT Dual Port
 Server Adapter (but with Linux 2.6.18 it works).
 Thanks to all for the suggestions...

 -Massimiliano
 
 
 this goes back to the dawn of the IBM PC...
 
 on the XT, there were only 8 IRQ originally - 0= timer, 1=keybd, etc.
 
 then out came the AT with a new 8259A PIC supporting cascading.
 
 IRQ 2 triggered automatically IRQ 9 to allow access to the other additional
 7 IRQs.

I know this...

 
 IRQ 2  9 became known as troublesome because the drivers didn't handle
 well sharing their interrupts, though there's nothing forbidding it IIRC.
 
  in the immortal words of Nick Holland -
 http://monkey.org/openbsd/archive/misc/0011/msg00927.html
 
 perhaps that will lead others to comment further on my hazy memory.

But not this... thanks...

 
 wd0a: aborted command, interface CRC error reading fsbn 403040 of
 403040-403071 (wd0 bn 79041215; cn 4920 tn 22 sn 29), retrying
 wd0: soft error (corrected)
 wd0: transfer error, downgrading to Ultra-DMA mode 4
 wd0(pciide0:0:0): using PIO mode 4, Ultra-DMA mode 4
 wd0a: aborted command, interface CRC error reading fsbn 138080 of
 138080-138111 (wd0 bn 78776255; cn 4903 tn 151 sn 47), retrying
 wd0: soft error (corrected)


  the next thing you need to do is to replace wd0 - its on its way out.

Well.. I have used an old 40 wires cable now I have replaced this
with a 80 wires IDE cable... no more problems... the tail of dmesg is

pctr: user-level cycle counter enabled
dkcsum: wd0 matches BIOS drive 0x80
root on wd0a swap on wd0b dump on wd0b

the system is up and is running without problems (~6 days)... I think
that I can send a message to dmesg[AT]openbsd.org :-)

-Massimiliano

 
 A+
 Dave

OpenBSD Strage Problem

2008-03-14 Thread Peter_APIIT

Hello all expert network administrator, i truly new to openbsd.

I have some dhcp problem. 

http://forums.bsdnexus.com/viewtopic.php?id=1858
-- 
View this message in context: 
http://www.nabble.com/OpenBSD-Strage-Problem-tp16062121p16062121.html
Sent from the openbsd user - misc mailing list archive at Nabble.com.

Re: OpenBSD Strage Problem