Re: i386/Omnibook 800ct: Can't see SCSI CDROM

[Duncan Patton a Campbell]

On Sun, 6 Jan 2008 20:52:29 -0600
Michael White [EMAIL PROTECTED] wrote:

 Hi all (there was no maintainer listed on my machine - sorry if I pulled in 
 too many folks),
 
 I'm having a problem with OpenBSD 4.2  my HP Omnibook 800CT.  Specifically, 
 the CDROM drive (SCSI ID 0) is not recognized.  It is a SCSI CDROM drive, 
 since the notebook has built in SCSI.  The SCSI chipset is recognized:
 

Mmm.  10+ years ago I got some of these for company I worked for.  
Huge brickish things that we ran BSDI on.  Very idiosyncratic
hardwarefirmware.  You don't mention whether the dmesg shows
a cd entry or which floppy set you are using to boot.

Dhu


 
 Jan  6 00:07:54 omnibook /bsd: siop0 at pci0 dev 5 function 0 Symbios Logic 
 53c810 rev 0x12: irq 10
 Jan  6 00:07:54 omnibook /bsd: scsibus0 at siop0: 8 targets
 
 
 During the install, the CDROM drive was not presented as a possible source 
 (the 800CT cannot boot from CDROM).
 
 If I try to mount /dev/cd0a (or cd0c, or cd1a, or cd1c), I see the following:
 
 --
 # cat /etc/fstab
 /dev/wd0a / ffs rw 1 1
 /dev/cd0a /mnt/cdrom cd9660 ro,noauto 0 0 
 # mount /mnt/cdrom
 mount_cd9660: /dev/cd0a on /mnt/cdrom: Device not configured
 ---
 
 I have two CDROM drives (but only one connected at a time), and neither are 
 recognized.  On boot, however, I hear noise from the CDROM drive as if it was 
 being accessed.  These drives work when I boot from an install floppy from 
 RedHat 8.0 (just re-tested).  I'd rather not go back to RedHat 8.0.
 
 System details:
 
 - As far as I know, I'm using the stable 4.2.  I simply booted from floppy 
 then did a network install off of openbsd.org a couple of days ago.
 
 # sysctl -n kern.version
 OpenBSD 4.2 (GENERIC) #375: Tue Aug 28 10:38:44 MDT 2007
 [EMAIL PROTECTED]:/usr/src/sys/arch/i386/compile/GENERIC
 
 Pentium MMX
 80 MB RAM
 4.3 GB IDE hard drive (boot)
 No other SCSI devices attached
 SCSI terminator dip switch is ON on the CDROM drive
 
 Any help would be appreciated.
 
 Thanks.
 -- 
 Michael White To protect people from the effects of folly is to
fill the world with fools. -Herbert Spencer
 
 

Re: i386/Omnibook 800ct: Can't see SCSI CDROM

[Duncan Patton a Campbell]

On Sun, 6 Jan 2008 22:36:03 -0600
Michael White [EMAIL PROTECTED] wrote:

 Dhu,
 
 No entries in the syslog for CDROM drives, or any output from dmesg|grep -i 
 cd.
 
 It now boots from the hard drive after the install was performed.  Sorry if 
 my 
 email was confusing on that point.  The only time I booted from floppy was 
 during the installation process.  I pulled the floppy disks from one of the 
 download sites.  I don't recall which one I booted from (floppy42/B42/C42), 
 but if it makes a difference, I'll run them through again to see which one it 
 was.
 

You may well need to build a specialized kernel if one of the distibuted
geneneric kernels won't support your CD.  There are differences in the 
kernels on the floppies that have different hardware sets.

Dhu

 Thanks.
 -- 
 Michael White To protect people from the effects of folly is to
fill the world with fools. -Herbert Spencer
 
 On Sunday 06 January 2008 22:17, Duncan Patton a Campbell wrote:
  On Sun, 6 Jan 2008 20:52:29 -0600
 
  Michael White [EMAIL PROTECTED] wrote:
   Hi all (there was no maintainer listed on my machine - sorry if I
   pulled in too many folks),
  
   I'm having a problem with OpenBSD 4.2  my HP Omnibook 800CT. 
   Specifically, the CDROM drive (SCSI ID 0) is not recognized.  It is a
   SCSI CDROM drive, since the notebook has built in SCSI.  The SCSI chipset
   is recognized:
 
  Mmm.  10+ years ago I got some of these for company I worked for.
  Huge brickish things that we ran BSDI on.  Very idiosyncratic
  hardwarefirmware.  You don't mention whether the dmesg shows
  a cd entry or which floppy set you are using to boot.
 
  Dhu
 
   
   Jan  6 00:07:54 omnibook /bsd: siop0 at pci0 dev 5 function 0 Symbios
   Logic 53c810 rev 0x12: irq 10
   Jan  6 00:07:54 omnibook /bsd: scsibus0 at siop0: 8 targets
   
  
   During the install, the CDROM drive was not presented as a possible
   source (the 800CT cannot boot from CDROM).
  
   If I try to mount /dev/cd0a (or cd0c, or cd1a, or cd1c), I see the
   following:
  
   --
   # cat /etc/fstab
   /dev/wd0a / ffs rw 1 1
   /dev/cd0a /mnt/cdrom cd9660 ro,noauto 0 0
   # mount /mnt/cdrom
   mount_cd9660: /dev/cd0a on /mnt/cdrom: Device not configured
   ---
  
   I have two CDROM drives (but only one connected at a time), and neither
   are recognized.  On boot, however, I hear noise from the CDROM drive as
   if it was being accessed.  These drives work when I boot from an install
   floppy from RedHat 8.0 (just re-tested).  I'd rather not go back to
   RedHat 8.0.
  
   System details:
  
   - As far as I know, I'm using the stable 4.2.  I simply booted from
   floppy then did a network install off of openbsd.org a couple of days
   ago.
  
   # sysctl -n kern.version
   OpenBSD 4.2 (GENERIC) #375: Tue Aug 28 10:38:44 MDT 2007
   [EMAIL PROTECTED]:/usr/src/sys/arch/i386/compile/GENERIC
  
   Pentium MMX
   80 MB RAM
   4.3 GB IDE hard drive (boot)
   No other SCSI devices attached
   SCSI terminator dip switch is ON on the CDROM drive
  
   Any help would be appreciated.
  
   Thanks.
   --
   Michael White To protect people from the effects of folly is to
  fill the world with fools. -Herbert Spencer
 

Re: UPDATE:graphics/swfdec, NEW:www/swfdec-mozilla

[Duncan Patton a Campbell]

On Tue, 13 May 2008 19:47:40 -0400
James Turner [EMAIL PROTECTED] wrote:

 On Tue, May 13, 2008 at 01:36:09PM +0300, Valery Masiutsin wrote:
  Hello, list !
  
  Here are updates for swfdec - opensource flash rendering library,
  and flash mozilla plugin that using this library.
  Please replace swfdec directory with one from archive.
  I will generate diff against the in-tree copy at some point later,
  this are mainly for testing purposes.
  Works for me on i386, i am able to watch youtube movies and flash
  cartoons from mult.ru, as well as many others.
  Please mind that swfdec exposes some bugs in Xenocara,and if you
  see black box instead of movie, before blaming swfdec please try to 
  reproduce
  issue using vesa driver. Crash reports are welcome as well.
  
  
  Regards Valery
 
 I tried this on amd64, compiles and installs fine.  Playing a youtube
 video, kind of works.  Video is very slow and sometimes doesn't show up.
 Sound also seems quite.  I don't follow swfdec development so these
 could be known amd64 issues for all I know, just wan't to give me
 feedback.
 
 -- 
 James Turner
 BSD Group Consulting
 http://www.bsdgroup.org
 

I tried to goto your url to have a look and
it can't be found, and afterwords browsers
can't find anything else (can't find google
until a restart!) most peculiar.



Dhu


 

Re: opera 10 beta with unite

[Duncan Patton a Campbell]

It would be nice to have Opera without the Linux baggage.

Who would I nag at Opera for this to be accomplished?

Thanks,

Dhu

On Wed, 24 Jun 2009 12:12:48 +0200
frantisek holop min...@obiit.org wrote:

 hi there,
 
 the following is a (less then stellar) port of opera's
 10b with unite.
 
 this port is _not_ an update for 9.64.
 it makes a separate package called opera10 (not opera-*)
 and installs everything mozilla style into /usr/local/opera10.
 it is possible to install it alongside 9.64.  it does not install
 the system wide configs, the desktop file, the icons and the
 man page.
 
 it also has a modified startup script and automatically uses
 .opera10 as profile directory (as opposed to .opera) and
 leaves your other opera alone.
 
 if you want to move your prefs from 9.64, just copy them over.
 but some filenames/directories have changed, so beware (still,
 i have no problems running it with my 9.64 profile directory,
 but back it up, because downgrading is not so bump free)
 
 the point of this port is to let people play with unite and with
 the beta -- it is blindingly fast (esp with javascript heavy sites
 like gmail and facebook) -- i am not sending this so it gets
 committed.
 
 i have not tested the flash plugin and have no interest in it
 really.
 
 the only static version at the moment for linux is gcc4 compiled.
 if this is a step forward, remains to be seen.  i hope there will
 be more choices for the final builds.
 
 start it with
 $ /usr/local/opera10/bin/opera
 
 have fun.
 
 -f
 -- 
 synonym: a word you use when you can't spell the other.
 

Re: Repost: new: textproc/filepp

[Duncan Patton a Campbell]

On Fri, 24 Jul 2009 12:21:43 -0400
William Yodlowsky b...@openbsd.rutgers.edu wrote:

 On 10 February 2009 at 21:16, b...@openbsd.rutgers.eduwrote:
 
  If you've ever wanted to use cpp for files other than C programs, this
  is the tool you were after.
  
  Port:   http://openbsd.rutgers.edu/filepp-1.8.0.tar
  
  Thanks
 
 Comments / ok please?  This is a very useful thing for e.g.
 close-but-not-the-same configuration files.  #ifdef to the rescue...
 or something.
 
 Thanks.
 
 

Well, other than the fact that I don't know what this does
(some kinda language tool, I guess) I still didn't get very
far before things went sideways:

# pwd  
/usr/ports/mystuff/filepp
# make 
Fatal: one category in textproc devel should match PKGPATH=filepp (in filepp)
Fatal: bogus PKGPATH=filepp (no subdirectory) (in filepp)
*** Error code 1

Stop.

Dhu

Re: Repost: new: textproc/filepp

[Duncan Patton a Campbell]

On Fri, 24 Jul 2009 22:39:34 +0200
Robert rob...@openbsd.pap.st wrote:

 On Fri, 24 Jul 2009 14:25:38 -0600
 Duncan Patton a Campbell campb...@neotext.ca wrote:
 
  Well, other than the fact that I don't know what this does
  (some kinda language tool, I guess) I still didn't get very
  far before things went sideways:
  
  # pwd  
  /usr/ports/mystuff/filepp
 
 ^ You are missing the category subdirectory.
 
 It should be: /usr/ports/mystuff/textproc/filepp
 

Yes, that appears to be the case but I was initially misled by this message:

[r...@gate:/tmp/filepp] # make
Can't find /tmp/filepp under PORTSDIR_PATH=/usr/ports:/usr/ports/mystuff
Fatal: can't figure out PKGPATH (in /tmp/filepp)
Fatal: one category in textproc devel should match PKGPATH=/tmp/filepp (in 
/tmp/filepp)
*** Error code 1

Stop.

Dhu

  # make 
  Fatal: one category in textproc devel should match PKGPATH=filepp (in
  filepp) Fatal: bogus PKGPATH=filepp (no subdirectory) (in filepp)
  *** Error code 1
  
  Stop.
  
  Dhu
 
 - Robert
 
 

Re: Why does apache2 need the db package? Anyway to avoid that?

[Duncan Patton a Campbell]

On Sat, 26 Dec 2009 23:09:54 -0600
Andres Salazar ndrsslz...@gmail.com wrote:

 Hello,
 
 When compiling apache2.2.14 from ports I noticed that it needs the
 db-4.6.21p0 Berkeley DB package, revision 4. Is this actually
 needed by the apache core? or is it a result of enabling a certain
 module i can actually disable??
 
 I have compiled apache.2.2.14 from source directly from apache.org in
 a test machine and i dont recall having to satisfy this one.
 
 Please advise.

Howdy Andres?  

I wondered about this too, but just installed db.  
If you find out please cc me, thanks and happy
New Year (or the local equivalent),

Dhu

 
 Thanks
 
 Andres
 

Re: Inclusion of software with subscription fee in the ports tree

[Duncan Patton a Campbell]

On Wed, 26 Nov 2008 02:22:54 +
Stuart Henderson [EMAIL PROTECTED] wrote:

 On 2008/11/25 20:40, Alban Deniz wrote:
  Hello,
  
  I am developing an application, and would appreciate guidance on whether
  it would be appropriate for inclusion in the ports tree.  I've looked at
  http://www.openbsd.org, and also the mailing list archives, but wasn't
  able to find the guidance I'm looking for.
  
  The application (SNFMilter) is an anti-spam email milter, and is
  licensed under the Artistic license.  However--The application needs a
  data file (periodically updated) in order to run, and the data file and
  periodic updates are available only for a fee.
  
  Would it be appropriate to include SNFMilter in the ports tree?
 
 I don't see why not if it works ok and is in keeping with ports
 standards and hier(7) (binaries tucked away in /var/spool/snfilter
 with .exe extensions so that it can rewrite it's file name to
 produce it's log file and to read it's rule file might need a
 bit of reworking though ;-)
 
 

Is the data file a proprietary or otherwise constrained format?

If so, I would be skeptical.  

Dhu

Gprolog Port

[Duncan Patton a Campbell]

Howdy all?  

Does anyone know if the Gprolog port is being maintained
anywhere?  

The 1.2.16 version in ports was broken, but this has
been fixed since 1.3.0 and we now have 1.3.1.  

Thanks,

Dhu

Cups/ghostscript Problems

[Duncan Patton a Campbell]

Howdy?

I'm trying to upgrade (reinstall) a system that works with my Brother 7020 DCP.

Presently the ghostview doesn't run: 

 cat input.ps | gs q dNOPAUSE sDEVICE=hl1250 sOutputfile=- - | lpr
GPL Ghostscript GPL Ghostscript 8.638.63: : Unrecoverable error, exit code 1
Unrecoverable error, exit code 1
lpr: Unsupported format 'text/plain'!

where input.ps is small ps derived via pdf2ps from a pdf.  

I've had various problems with cups and this printer here and in 
4.7 and am now wondering if this is a deprecated mechanism.

Thanks,

Dhu

Re: print-preview broken in firefox?

[Duncan Patton a Campbell]

I'd been tearing hair out over this and the general
inability of the Apple cups to cope with my admittedly
elderly Brother DCP 7020.  I ran across a comment in 
one of the linux forums saying that the Brother/Cups
(and some other printers) combination was fouled up
because of mishandling of the bidir usb.  So I pulled 
it all out and replaced it with a local build of cups-1.2.12
and now it all works again, including ctrl-P.  

Now if only samba would still work... but I my have to
go retro there, too.

Dhu


On Tue, 28 Sep 2010 15:08:26 +0200 (CEST)
Antoine Jacoutot ajacou...@bsdfrog.org wrote:

 On Tue, 28 Sep 2010, Edd Barrett wrote:
 
  Hi,
  
  I can not seem to print preview in either firefox35 or 36. It just 
  disappears
  (nothing printed on the terminal). I am using an amd64 machine with cups 
  and I
  *do* have the gtk2 cups dialog installed (as suggested in README.OpenBSD).
  
  Making a new profile changes nothing.
  
  Disabling cups (/usr/local/sbin/cupsd-disable) makes no difference.
 
 This issue has been known for years. Sometimes the preview disappears, 
 or printing crashes...
 No one knows why, but it only happens when gtk2-cups is used for 
 printing in firefox. Our last hint was a conflict between openssl, 
 gnutls and nss but really, no one knows.
 Talk to landry I think he has some bugzilla entries links that talk 
 about some Firefox+cups related crashes.
 
 -- 
 Antoine
 

Re: print-preview broken in firefox?

[Duncan Patton a Campbell]

On Tue, 28 Sep 2010 15:46:46 -0600
Duncan Patton a Campbell campb...@neotext.ca wrote:

 
 I'd been tearing hair out over this and the general
 inability of the Apple cups to cope with my admittedly
 elderly Brother DCP 7020.  I ran across a comment in 
 one of the linux forums saying that the Brother/Cups
 (and some other printers) combination was fouled up
 because of mishandling of the bidir usb.  So I pulled 
 it all out and replaced it with a local build of cups-1.2.12
 and now it all works again, including ctrl-P.  
 

Oh, and I had to pkg_deletepkg_add foomatic-filters


 Now if only samba would still work... but I my have to
 go retro there, too.
 
 Dhu
 
 
 On Tue, 28 Sep 2010 15:08:26 +0200 (CEST)
 Antoine Jacoutot ajacou...@bsdfrog.org wrote:
 
  On Tue, 28 Sep 2010, Edd Barrett wrote:
  
   Hi,
   
   I can not seem to print preview in either firefox35 or 36. It just 
   disappears
   (nothing printed on the terminal). I am using an amd64 machine with cups 
   and I
   *do* have the gtk2 cups dialog installed (as suggested in README.OpenBSD).
   
   Making a new profile changes nothing.
   
   Disabling cups (/usr/local/sbin/cupsd-disable) makes no difference.
  
  This issue has been known for years. Sometimes the preview disappears, 
  or printing crashes...
  No one knows why, but it only happens when gtk2-cups is used for 
  printing in firefox. Our last hint was a conflict between openssl, 
  gnutls and nss but really, no one knows.
  Talk to landry I think he has some bugzilla entries links that talk 
  about some Firefox+cups related crashes.
  
  -- 
  Antoine
  

Re: print-preview broken in firefox?

[Duncan Patton a Campbell]

On Tue, 28 Sep 2010 23:10:18 +0100
Stuart Henderson s...@spacehopper.org wrote:

 On 2010/09/28 15:46, Duncan Patton a Campbell wrote:
  
  Now if only samba would still work... but I my have to
  go retro there, too.
 
 More details wouldn't go amiss...
 

Only speculation at this point.  I tried to use the same
config  location stuff that the OBSD port of cups does,
but it might be easier to use the generic source... dunno
yet.

Dhu

Re: openvpn user wanted

[Duncan Patton a Campbell]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Howdy?  

I've just been doing a bunch of stuff using openvpn
on 5.5r,5.7r and 5.8snap (i386,amd64,sparc64).  
These are just straight package installs onto 
fairly idiosyncratic systems.

Does this have to be built in the ports tree?  

Dhu

On Sun, 13 Sep 2015 19:34:25 +0100
Stuart Henderson  wrote:

> Could someone try OpenVPN with this please? It should be a noop
> currently but makes it easier to handle a forthcoming change.
> 
> 
> Index: patches/patch-src_openvpn_tun_c
> ===
> RCS file: /cvs/ports/net/openvpn/patches/patch-src_openvpn_tun_c,v
> retrieving revision 1.4
> diff -u -p -r1.4 patch-src_openvpn_tun_c
> --- patches/patch-src_openvpn_tun_c   25 Jun 2015 08:46:49
> - 1.4 +++ patches/patch-src_openvpn_tun_c 13 Sep
> 2015 18:33:41 - @@ -1,41 +1,164 @@
>  $OpenBSD: patch-src_openvpn_tun_c,v 1.4 2015/06/25 08:46:49 sthen
> Exp $
>  src/openvpn/tun.c.orig   Mon Jun  8 07:16:35 2015
> -+++ src/openvpn/tun.cThu Jun 25 09:38:54 2015
> -@@ -944,7 +944,19 @@ do_ifconfig (struct tuntap *tt,
> +--- src/openvpn/tun.c.orig   Fri Jul 17 06:43:38 2015
>  src/openvpn/tun.cSun Sep 13 19:33:19 2015
> +@@ -912,8 +912,17 @@ do_ifconfig (struct tuntap *tt,
> + 
> +   tt->did_ifconfig = true;
> + 
> +-#elif defined(TARGET_OPENBSD)
> ++#elif defined(TARGET_NETBSD) || defined(TARGET_OPENBSD)
> + 
> ++/*
> ++ * OpenBSD can do IPv6. NetBSD's IPv6 support over tun can be
> detected by ++ * availability of the TUNSIFHEAD ioctl() - see next
> TARGET_NETBSD block ++ * for more details.
> ++ */
> ++#if defined(TUNSIFHEAD)||defined(TARGET_OPENBSD)
> ++# define MULTI_AF
> ++#endif
> ++
> +   /*
> +* On OpenBSD, tun interfaces are persistent if created with
> +* "ifconfig tunX create", and auto-destroyed if created by
> +@@ -923,78 +932,26 @@ do_ifconfig (struct tuntap *tt,
> +   /* example: ifconfig tun2 10.2.0.2 10.2.0.1 mtu 1450 netmask
> 255.255.255.255 up */
> +   if (tun)
> + argv_printf (,
> ++#if defined(TARGET_OPENBSD)
> +   "%s %s %s %s mtu %d netmask
> 255.255.255.255 up -link0", ++#else
> ++  "%s %s %s %s mtu %d netmask
> 255.255.255.255 up", ++#endif
> +   IFCONFIG_PATH,
> +   actual,
> +   ifconfig_local,
> +   ifconfig_remote_netmask,
> +-  tun_mtu
> +-  );
> +-  else
> +-if ( tt->topology == TOP_SUBNET )
> +-{
> +-argv_printf (,
> +-  "%s %s %s %s mtu %d netmask %s up -link0",
> +-  IFCONFIG_PATH,
> +-  actual,
> +-  ifconfig_local,
> +-  ifconfig_local,
> +   tun_mtu,
> +-  ifconfig_remote_netmask
> );
> - }
> +-}
> else
>  -argv_printf (,
> -+  {
> -+if (tt->topology == TOP_SUBNET)
> -+  argv_printf (,
> -+  "%s %s %s %s netmask %s mtu %d up",
> -+  IFCONFIG_PATH,
> -+  actual,
> -+  ifconfig_local,
> -+  ifconfig_local,
> -+  ifconfig_remote_netmask,
> -+  tun_mtu
> -+  );
> -+else
> -+  argv_printf (,
> -   "%s %s %s netmask %s mtu %d broadcast %s
> link0", +-  "%s %s %s netmask %s mtu %d
> broadcast %s link0", +- IFCONFIG_PATH,
> +-  actual,
> +-  ifconfig_local,
> +-  ifconfig_remote_netmask,
> +-  tun_mtu,
> +-  ifconfig_broadcast
> +-  );
> +-  argv_msg (M_INFO, );
> +-  openvpn_execve_check (, es, S_FATAL, "OpenBSD ifconfig
> failed"); +-  if ( do_ipv6 )
> +-{
> +-  argv_printf (,
> +-  "%s %s inet6 %s/%d",
> +-  IFCONFIG_PATH,
> +-  actual,
> +-  ifconfig_ipv6_local,
> +-  tt->netbits_ipv6
> +-  );
> +-  argv_msg (M_INFO, );
> +-  openvpn_execve_check (, es, S_FATAL, "OpenBSD
> ifconfig inet6 failed"); +-
> +-  /* and, hooray, we explicitely need to add a route... */
> +-  add_route_connected_v6_net(tt, es);
> +-}
> +-  tt->did_ifconfig = true;
> +-
> +-#elif defined(TARGET_NETBSD)
> +-
> +-/* whether or not NetBSD can do IPv6 can be seen by the
> availability of +- * the TUNSIFHEAD ioctl() - see next TARGET_NETBSD
> block for more details +- */
> +-#ifdef TUNSIFHEAD
> +-# define NETBSD_MULTI_AF
> +-#endif
> +-
> +-  if (tun)
> +-argv_printf (,
>