from:"Polytropon"

On Sun, 20 May 2012 20:59:50 +1000, andrew clarke wrote:
 In FreeBSD I use /bin/ls:
 
 setenv LSCOLORS ExGxFxdxCxDxDxhbadExEx
 alias ls 'ls -D %Y-%m-%d %H:%M:%S'
 
 The -D stuff is to display ISO 8601 style timestamps like GNU ls's
 --time-style=long-iso format, eg:
 
 -r-xr-xr-x  1 root  wheel  12612347 2011-09-28 19:13:57 /boot/GENERIC/kernel
 
 I don't know if this helps the OP. :-)

At least it helps me, many thanks for the inspiration!
I now have (and intend to keep using):

setenv LSCOLORS ExGxdxdxCxDxDxBxBxegeg
alias ls 'ls -FG -D %Y-%m-%d %H:%M:%S'
alias ll 'ls -laFG -D %Y-%m-%d %H:%M:%S'

This is nice because I haven't found a feature of gls yet
that I needed, but which system's ls couldn't provide.


-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: eliminate character with sed

On Sun, 20 May 2012 12:08:04 -0400, Tim Dunphy wrote:
 Hello list,
 
  I have a few php config files that have the windows delimiter
 character in them  ('^M') that I would like to get rid of. I'm trying
 to use sed to do it, and for some reason I am not having any luck.
 
 Here's the line that I'm trying to use:
 
  #sed -i '.bak' 's/^M//g' config.php
 
 However when I have a look at the backup file that's been created with
 this command, it looks like there was no effect:
 
 ?php ^M/*   Global Variables   */^Mif(!defined('DS'))^M
 define('DS',DIRECTORY_SEPARATOR);^M^M
 if(!defined(_MAINSITEPATH_))^M
 define(_MAINSITEPATH_,dirname(__FILE__).DS);^M
 
 I was wondering is someone had a tip on how to run this command
 effectively in this situation.

It seems that you need proper quoting. The character
sequence ^M != ASCII code of CR. Note that ^M is
just a _visual representation_ of Ctrl-M, which does
output the same ASCII code as the CR key would.
The corresponding escape sequence is \r. Together
with \n, the DOS-based line end \r\n is generated.
Remove the \r part, and you have the default (normal)
line end - as you correctly intended.

Note that different viewers or editors might have a
different representation than ^M!

As you just want to delete the ^M (the CR), why not
use tr instead?

% tr -d '\r'  config.php  config.php.new

Put a bit of scripting around it, and it will do the
same. See man tr for details.

Regarding the use of sed: I'm not sure if it's possible
to do something like

% sed -i '.bak' 's/\r//g' config.php

because I assume (not tested!) that it's not possible to
put in escape sequences like that. But try for yourself
and surprise me. :-)



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: eliminate character with sed

On Sun, 20 May 2012 21:10:51 +0200, Michael Ross wrote:
 Maybe you can use /usr/ports/converters/dosunix.
 
   Usage:
 dosunix INPUTFILE OUTPUTFILE
 
   DosUnix converts files from DOS text format to Unix text format
   by replacing each carriage return  newline pair with a single
   newline character.
 
 does not do backups, though.

If more complex conversions (including character sets) is
needed (which often is the case for non-US Windows files),
the recode port is very useful:

% recode cp437..iso8859-1 file

It also doesn't do backups, which would be required to do
using a surrounding script.


-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: NanoBSD Build Failure

On Sun, 20 May 2012 19:49:49 -0700, Thomas D. Dean wrote:
 I am attempting to rebuild nanoBSD on an AMD64 system, using the same 
 config file I used a couple years ago on a 32-bit system.
 [...]
 /usr/home/tomdean/nanoBSD/MYKERNEL: unknown option I486_CPU

It seems that the amd64 sources do not understand cpu I486_CPU.
Compare to /sys/i386/conf/GENERIC and /sys/i386/conf/NOTES
to the respective /sys/amd64/conf/GENERIC and /sys/amd64/conf/NOTES
for details, especially that amd64 uses cpu HAMMER instead
of cpu I{4/5/6}86_CPU.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: ATI Radeon HD5500 driver question

2012-05-19 Thread Polytropon

On Sat, 19 May 2012 11:59:03 -0600, Gary Aitken wrote:
 1.  Since the driver notes that Acceleration was disabled, but there are 
 no other errors, shouldn't the driver work in dumb frame buffer mode? 
 (Hoping to get something hobbling along...)  Or is this the result of 
 the need for KMS and I'm SOL?

Do you have drm/dri (direct renering) installed, port and
kernel module? I've been using that with a ATI Radeon 9200
(I think, RV250, no HD) with excellent 2D and 3D results
both with XFree86 and X.org - tested with excessive gaming. :-)



 2.  Since the server didn't exit, is it actually pretending to run? 

Check using ps or top.



 Shouldn't I be seeing the standard X grey hatched background? 

No. The default new background is plain black. Nothing to see.
No grey pattern, no twm, nothing. And in case HAL and DBUS _or_
xorg.conf settings don't really match, you don't even see the
X-shaped mouse cursor.



 3.  The Xorg man page notes that ctrlaltbksp should cause it to 
 exit.  However, it doesn't, and I had to use kill -TERM.  Any hints on 
 why ctrlaltbksp doesn't cause it to exit?

This is also a new default to _not_ work anymore. You have more
than two (if I remember correctly) options in making it work.
You'll find them in the Handbook:

http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/x-config.html

One possibility when X has been compiled _without_ HAL support
(and no hald running), placing 

Option DontZap false

into the ServerLayout section should work. Additionally, I see
that I have

Option XkbOptions terminate:ctrl_alt_bksp

in the InputDevice section of Keyboard0. It just works. :-)





-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: stay up to date with ports and packages, problem

2012-05-19 Thread Polytropon

On Sat, 19 May 2012 19:43:09 +0100, RW wrote:
 On Sat, 19 May 2012 11:08:19 -0700 (PDT)
 Beastie-Boy wrote:
 
  Ok, many thanks for your replies.
  I forgot to tell that i recently upgraded from 8.1 to 9.0-RELEASE.
  That excplains maybe why i had obsolete/old packages/ports on my disk.
 
 
 When you cross a major OS release boundary, you need to force a rebuild
 of all installed package, or reinstall from package files.

It's often easy to do this using a port management tool.
See man portmaster containing an example of exactly
this procedure (EXAMPLES section).


-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: stay up to date with ports and packages, problem

2012-05-19 Thread Polytropon

On Sat, 19 May 2012 11:08:19 -0700 (PDT), Beastie-Boy wrote:
 Ok, many thanks for your replies.
 I forgot to tell that i recently upgraded from 8.1 to 9.0-RELEASE.
 That excplains maybe why i had obsolete/old packages/ports on my disk.

When you do such an update (major version number), you should
always reinstall (update) your applications. You can avoid it
by installing the compat-Nx-i386 or compat-Nx-amd64 ports (where
N is the previously used major version number).

You've found many advices on how to do that already from the
list.



 The problem i had was that gdm, gnome didnt start after the upgrade.

That was to be expected.



 So i tried to build the gnome and gdm thing again via pkg_add(didnt work)
 and make install clean in ports(either).

You should make sure _all_ dependencies get recompiled. Using
a port management tool for this task often is more comfortable
than dealing with the bare ports (but it basically is not
wrong).



 Right now i deleted all ports in /usr, deleted packages in /var and
 portsnaped me the all stuff again.

Depending on how you deleted, it _might_ be required to reconstruct
the directory subtree /usr/local from the respective mtree file
in /etc/mtree. If you _really_ intend to delete everything, make
sure you have backups of config files, data files or your own
modifications to something located in the local/ subtree (for
example /usr/local/etc).



 After that i pkg_add -r gnome2 again and now it looks better.

Erm... when you're installing binary packages, you don't have to
deal with ports at all.



 Before i had problems that package-1.2.3 is needed to build an only
 package-1.2.2 is installed.

Correct, this happens when packages have lower version numbers
(not totally up to date) than the respective port would have.
That's why it's often a good idea to use _either_ ports _or_
packages (even though technically there is no problem mixing
them).

Again, allow me to mention port management tools. Using for
example portmaster, many tasks are easier to perform than
dealing with bare ports. Even the use of precompiled
packages (if desired) is possible. See man portmaster
and its EXAMPLES section for inspiration.



 Sorry i cant paste logs, bsd is running on another machine.

You can use SSH to log into the BSD machine and cut text from
the session. :-)




-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: ls-F tcsh built-in command

2012-05-17 Thread Polytropon

On Thu, 17 May 2012 20:24:02 +0900, fake fake wrote:
 Thank you for replying.
 But I am telling 'ls-F' (tcsh built-in command), not 'ls -F'.

Please see man csh:

ls-F acts like `ls -CF', unless listflags contains
an `x', in which case it acts like `ls -xF'.  ls-F  passes  its
arguments  to  ls(1)  if it is given any switches, so `alias ls
ls-F' generally does the right thing.

So if you use ls-F -l, the C shell will _not_ use ls-F, but
call /bin/ls instead. So what you've been observing seems to
be the intended behaviour.


-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: ls-F tcsh built-in command

2012-05-17 Thread Polytropon

On Thu, 17 May 2012 21:59:52 +0900, fake fake wrote:
 Oh, dear. I didn't notice it.
 So, is there no way to color directory in ISO 6429 codes with using tcsh?

Judging from man csh:

File  names  can also be colorized based on filename extension.
This is specified in the LS_COLORS variable  using  the  syntax
*ext=string.  For example, using ISO 6429 codes, to color all
C-language source files blue you would specify *.c=34.   This
would color all files ending in .c in blue (34) color.

[...]

If your terminal does use ISO 6429 color codes, you can compose

[...]

Not all commands will work on all systems or display devices.

It seems that the terminal emulator in use also plays an
important role.



Search for LS_COLORS in the environment variables section
of man csh. However, I've always been satisfied with using
$LSCOLORS as ExGxdxdxCxDxDxBxBxegeg. :-)



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Ethernet driver HP ProLiant BL460c G7

On Wed, 16 May 2012 15:29:05 +0400, hasanhasanli Hasan wrote:
 I bought server HP ProLiant BL460c G7. I couldn't find Ethernet
 driver for the FreeBSD. 
 
 Does anyone know where can I get driver for my OpenBSD(or freeBSD) server ?

You shouldn't need to get a driver because it's the
operating system's job to provide the drivers. FreeBSD
comes with lots of drivers for Ethernet cards.

After booting the machine with a FreeBSD system (or even
after installing it), try

# ifconfig -a

and look for which Ethernet interfaces have been success-
fully created. Missing something? Compare to the hardware
specification which lists the kinds of installed network
hardware (should be in your documentation provided with
the server).

Check those lists:

http://www.freebsd.org/releases/8.0R/hardware.html#ETHERNET

http://www.freebsd.org/relnotes/CURRENT/hardware/support.html#ETHERNET

You can always use man driver to find out more about
drivers on an installed system, or you can use the hyperlinks
in that lists to access the online documentation of the drivers.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: disable console messages

On Wed, 16 May 2012 00:27:07 -0700, mahdieh salamat wrote:
 hi all. how I can disbale console messages and clear screen on boot?
 Thanks

The console messages can be suppressed by commenting out the
line

*.err;kern.warning;auth.notice;mail.crit/dev/console

in /etc/syslog.conf; if you want the messages redirected to
a file (instead of the screen), just provide a different
target, such as

*.err;kern.*;auth.notice;mail.crit;console.info/var/log/console.log

Make sure /var/log/console.log does exist: run

# touch /var/log/console.log

There are more useful examples in that file. See man 5 syslog.conf
for details.



To also silence the kernel, you could add

boot_mute=YES

or (and?)

consolse=nullconsole

to /boot/loader.conf. See man 8 loader and man 5 loader.conf
and less /boot/defaults/loader.conf for details.



Finally:

Add the command clear or /usr/bin/tput clear at the end of
the /etc/rc script (not very clean, but works).



A quick web search brings up some inspiration:

http://lists.freebsd.org/pipermail/freebsd-questions/2003-October/022911.html
http://forums.freebsd.org/showthread.php?t=10341
http://forums.freebsd.org/showthread.php?p=58256#post58256

You'll find a suggestion there on how to avoid fiddling with
the /etc/rc boot script (which should stay untouched due to
many reasons).



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Building FreeBSD to install or update in two DESTDIRs

On Wed, 16 May 2012 03:13:10 -0400, Thomas Mueller wrote:
   Better to make buildkernel and make installkernel as two
   separate steps, rather than make kernel?
 
  Yes. You only need to make buildkernel once, then make installkernel
  for both $DESTDIRs.
 
 The idea was to make buildkernel once and make buildworld once
 and install to two different DESTDIRs.

I'm not sure I understand: The two install* targets (make installkernel
and make installworld) are only able to install to _one_ location,
which is the _default_ location *or* the location pointed to by
DESTDIR. It is not possible to perform _one_ install step which
will cause results in _two_ locations (without any means of
hidden duplication, e. g. by using a mirroring technique).



   After rebooting single-user, do mergemaster -p,
   then mergemaster -p -D /mnt, and then make installworld and
   immediately following that, make installworld DESTDIR=/mnt ?
 
  Refer to the commend header in /usr/src/Makefile for the
  correct procedure. Without having it tested, the following
  commands in SUM (after you have successfully installed the
  new kernels) should work as intended:
 
  # merpemaster -p
  # make installworld
  # make delete-old
  # mergemaster
 
  # merpemaster -p -D /mnt
  # make installworld DESTDIR=/mnt
  # make delete-old DESTDIR=/mnt
  # mergemaster -D /mnt
 
  # reboot
 
  Also see the comment regarding make delete-old-libs to be
  applied after reboot correspondingly.
 
 I assume your merpemaster is a typo for mergemaster?

Ah yes, the well-known P next to G typing error, procreated
by copy  paste. :-)



 I would have done each step for main installation and then for
 USB stick (DESTDIR=/mnt) before going to the next step, or
 maybe that doesn't really matter?

The order of targets does not matter. However, the order of
steps _per_ target does matter (e. g. make installkernel
first, _then_ installworld - it matters as soon as you boot).



 I think the second mergemaster was supposed to be done
 before make delete-old, or maybe that doesn't really matter
 either?

I'm refering to the instructions presented in /usr/src/Makefile:

# For individuals wanting to upgrade their sources (even if only a
# delta of a few days):
#
#  1.  `cd /usr/src'   (or to the directory containing your source tree).
#  2.  `make buildworld'
#  3.  `make buildkernel KERNCONF=YOUR_KERNEL_HERE' (default is GENERIC).
#  4.  `make installkernel KERNCONF=YOUR_KERNEL_HERE'   (default is GENERIC).
#   [steps 3.  4. can be combined by using the kernel target]
#  5.  `reboot'(in single user mode: boot -s from the loader prompt).
#  6.  `mergemaster -p'
#  7.  `make installworld'
#  8.  `make delete-old'
#  9.  `mergemaster'(you may wish to use -i, along with -U or -F).
# 10.  `reboot'
# 11.  `make delete-old-libs' (in case no 3rd party program uses them anymore)

That means: mergemaster -p before installing world, mergemaster
(maybe with additional options) after installing world. I think
the delete-old and delete-old-libs steps can also be performed at
the end of the whole process. But doing what the instructions say
has always been the most comfortable way of avoiding trouble. :-)



   I installed to USB stick only after fully upgrading on main
   installation, finally copied /boot/kernel directory, and that
   USB stick is now bootable.  So now I know how to make a USB
   stick bootable with GPT.
 
  Maybe kernel modules for GPT have been missing? Check /etc/src.conf
  for any strange settings, see man 3 src.conf for details. You
  can use this file to customize and tweak your builds.
 
 I think I must have all GPT modules there; I have no trouble
 accessing hard-disk partitions, and USB stick when partitioned GPT.
 
 I might want to prevent building ulpt module because of hplip
 and HP 1212nf MFP printer idiosyncrasies, though that may or
 may not make any difference.
 
 I could also prevent building other modules that would not be used.

There are several means you can use for this: a custom kernel
configuration, settings in /etc/src.conf, settings in /boot/loader.conf.
It should not _require_ you to deal with a custom kernel if you want
to avoid that.




-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Building FreeBSD to install or update in two DESTDIRs

On Wed, 16 May 2012 18:46:28 -0400, Thomas Mueller wrote:
   The idea was to make buildkernel once and make buildworld once
   and install to two different DESTDIRs.
 
  I'm not sure I understand: The two install* targets (make installkernel
  and make installworld) are only able to install to _one_ location,
  which is the _default_ location *or* the location pointed to by
  DESTDIR. It is not possible to perform _one_ install step which
  will cause results in _two_ locations (without any means of
  hidden duplication, e. g. by using a mirroring technique).
 
 The idea was to make buildkernel once and make buildworld once,
 but make installkernel and make installworld would each have
 to be done once for each DESTDIR, meaning twice each.  Building
 takes much more computer resources than installing, so I try to
 avoid building the same thing twice.

Yes, _that_ is the correct approach.

In case you need to do more than one additional installation,
you should consider creating a tar archive of the fully installed
system and then use tar --unlink to the mounted target. If you
need to create many bootable systems from scratch, a script
performing the disklabel, newfs, mount and tar steps should
be easy to write.



  I'm refering to the instructions presented in /usr/src/Makefile:
 
  # For individuals wanting to upgrade their sources (even if only a
  # delta of a few days):
  #
  #  1.  `cd /usr/src'   (or to the directory containing your source 
  tree).
  #  2.  `make buildworld'
  #  3.  `make buildkernel KERNCONF=YOUR_KERNEL_HERE' (default is 
  GENERIC).
  #  4.  `make installkernel KERNCONF=YOUR_KERNEL_HERE'   (default is 
  GENERIC).
  #   [steps 3.  4. can be combined by using the kernel target]
  #  5.  `reboot'(in single user mode: boot -s from the loader 
  prompt).
  #  6.  `mergemaster -p'
  #  7.  `make installworld'
  #  8.  `make delete-old'
  #  9.  `mergemaster'(you may wish to use -i, along with -U or 
  -F).
  # 10.  `reboot'
  # 11.  `make delete-old-libs' (in case no 3rd party program uses them 
  anymore)
 
 I checked /usr/src/UPDATING, and the sequence was
 
 ...
 reboot in single user mode
 mergemaster -p 
 make installworld
 mergemaster -i  (one may wish also -U)
 make delete-old
 reboot
 
 I checked /usr/src/Makefile , and found you quoted correctly,
 see 'make delete-old' and the second 'mergemaster' are transposed
 relative to what I saw in /usr/src/UPDATING 
 
 Sort of confusing; make either way works?

It should not be much difference if you consider what the
steps in permutation do: mergemaster modifies files,
make delete-old removes stuff that isn't needed anymore.
Both steps don't seem to affect each other. a + b = b + a. :-)



 Subsequently I would also want to build for i386, but this
 would be after the amd64 build and installation/update.

In case you're creating different TARGET= architectures,
the fun doubles. :-)



 It would be nice if bsdinstall had an option for update as
 well as fresh install.

This step can easily be performed manually using freebsd-update
right after installation.

If you understand update == overwrite, just don't format the
partitions that are already present on the target media.



 For i386, I would follow advice on http://wiki.freebsd.org/Wine
 but would want a full installation capable of running
 independently of amd64, would need the kernel, would install
 on 16 GB USB stick which could be mounted on /compat/i386. 
 I would want to use hard-drive PORTSDIR, would need to so
 adjust /etc/make.conf .  I don't really want to think of
 building big ports on a USB stick, especially on the older
 computer with 256 MB RAM and USB 1.1 on motherboard.

Also note that USB sticks are not real R/W media, they
suffer more from using than hard disks do, and they're
slower of course. :-)





-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Building FreeBSD to install or update in two DESTDIRs

2012-05-15 Thread Polytropon

On Mon, 14 May 2012 20:45:51 -0400, Thomas Mueller wrote:
 I guess after the first installkernel, to default location,
 I should immediately make installkernel again, this time with
 DESTDIR=/mnt?

That should be possible, you only have to make sure that both
install targets are fine with the kernel you just built (e. g.
both i386 _or_ amd64).



 Better to make buildkernel and make installkernel as two
 separate steps, rather than make kernel?

Yes. You only need to make buildkernel once, then make installkernel
for both $DESTDIRs.



 After rebooting single-user, do mergemaster -p,
 then mergemaster -p -D /mnt, and then make installworld and
 immediately following that, make installworld DESTDIR=/mnt ?

Refer to the commend header in /usr/src/Makefile for the
correct procedure. Without having it tested, the following
commands in SUM (after you have successfully installed the
new kernels) should work as intended:

# merpemaster -p
# make installworld
# make delete-old
# mergemaster

# merpemaster -p -D /mnt
# make installworld DESTDIR=/mnt
# make delete-old DESTDIR=/mnt
# mergemaster -D /mnt

# reboot

Also see the comment regarding make delete-old-libs to be
applied after reboot correspondingly.



 After that, I would do mergemaster -i followed by
 mergemaster -i -D /mnt?  And then make delete-old followed
 by DESTDIR=/mnt make delete-old? 

It should be possible to pass DESTDIR= to make instead of prefixing
make with it. The parameter seems to be applied for _any_ of the
targets (as long as it would affect that target). You can add
additional parameters to the mergemaster examples above (such
as -i).



 Would I need to do make distribution?

I don't think so, unless you want to create a distribution media.



 First time, make installkernel DESTDIR=/mnt only installed part. 

What parts (of the kernel set) have been installed? To observe
differences, it might be helpful to save a `ls` or `ls -lR`
output before and after the installation and compare them.



 I installed to USB stick only after fully upgrading on main
 installation, finally copied /boot/kernel directory, and that
 USB stick is now bootable.  So now I know how to make a USB
 stick bootable with GPT.

Maybe kernel modules for GPT have been missing? Check /etc/src.conf
for any strange settings, see man 3 src.conf for details. You
can use this file to customize and tweak your builds.


 
 Maybe some of the files were cleaned out?

I'm not sure in how far the install* targets to remove files.
I suppose they will overwrite files if required...



 It is surely useful to have a rescue backup, considering the
 possibility of an update going awry on the main installation.

That's right. :-)




-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: This does look strange

2012-05-15 Thread Polytropon

On Tue, 15 May 2012 11:11:44 +0200, Bernt Hansson wrote:
 Hello list
 
 After a reinstall of winxp, yes I know but the games.

You like playing with obsoleted OS imitations, that's okay. :-)



 I have a fat32 slice/partition/postcard whatever it's called.

It should be called a slice, because slice refers to a DOS
primary partition, and those are covered with a FAT or NTFS
file system directly (unlike BSD which puts partitions into
a slice to carry more than one file system).

But postcard is also okay. :-)



 Mocking me with:
 
 testbox# fsck -y -t msdosfs /dev/ad4
 ** /dev/ad4
 Invalid signature in fsinfo block
 Fix? yes
 fsck: /dev/ad4: Floating point exception: 8
 testbox#

Very strage. You're not supposed to fsck /dev/ad4 I think,
but you should name the _slice_ where Windows XP is installed
on. That should be something like /dev/ad4s1 (if it's the
1st primary partition on that disk).

Furthermore, -t msdosfs looks strange. As far as I know, the
newer versions of Windows come with NTFS as the primary
file losing system, so -t ntfs should be worth a try.

The command should be something like that:

# fsck -y -t ntfs /dev/ad4s1

or

# fsck -y -t msdosfs /dev/ad4s1

if you have _not_ formatted the postcard using NTFS, but FAT
(which corresponds to msdosfs).



 Anyone know what to do, is there a msdosfs fsck?

Yes, it's a native tool called CHKDSK.EXE. :-)

Really: You should first use the native tools provided by
Windows to fix a problem that seems to be a Windows
problem. If everything fails, you can always relapse to
forensic tools running on FreeBSD, or simply load your
backup sets.

There's also emulators/mtools in the ports collection
which might contain tools useful in this situation.

If you've just accidentally tried to fsck the wrong device
file, just forget everything I mentioned and use the correct
one. However, I'm not fully sure if FreeBSD's fsck can be
used to _really_ perform file system checks on FAT or NTFS
partitions, erm postcards.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Article Inquiry

2012-05-15 Thread Polytropon

On Tue, 15 May 2012 12:33:15 -0700, Lauren Scott wrote:
 Hello,
 
 I was wondering if you would be able to help me locate a copy of the below
 article:
 
 General Commands Manual for RDIST
 June 3, 1993
 FreeBSD
 
 I appreciate any help you are able to provide.

Are you sure about the date? If I read /usr/share/misc/bsd-family-tree
correctly, FreeBSD 1.0 is of 1993-11-01 (newer than 1993-06-03, the
date you provided).

The rdist program isn't part of the FreeBSD OS, it is provided
as a port. However, the web manpage collection contains the
following manual:

FreeBSD General Commands Manual for RDIST
March 17, 1994
4.3 Berkeley Distribution

Source:

http://www.freebsd.org/cgi/man.cgi?query=44bsd-rdistsektion=1apropos=0manpath=FreeBSD+9.0-RELEASE+and+Ports

Does this help?




-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: CSH prompt

2012-05-14 Thread Polytropon

On Sun, 13 May 2012 08:16:51 -0600, Reed Loefgren wrote:
 On 05/13/12 07:25, Polytropon wrote:
  On Sun, 13 May 2012 14:56:31 +0200, Jos Chrispijn wrote:
  In cshell I use this prompt: set prompt = %B[%@]%b %m[%/]  
  The problem I face now is that if I use this prompt with symbolic links,
  the presented location is displaying the symbolic link rather than the
  real directory name.
  Is there a way of preventin this?
  Yes, a very ugly way which I just found out:
 
  alias precmd 'set WD=`pwd`; set prompt = %B[%@]%b %m[$WD]  '
 
  Example:
 
  [3:21pm] r56[/]  cd /sys
  [3:21pm] r56[/usr/src/sys]  _
 
  It redefines the whole prompt at any command that could
  affect the current working directory (not only cd can
  do that). This is needed as any call to `pwd` stored into
  a variable will only affect $prompt once - this is when
  it's set, only at this time $WD would be evaluated. So
  that's why this strange command. :-)
 
  Oh, and I just improved it. How about this?
 
  alias precmd 'set prompt = %B[%@]%b %m[`pwd`]  '
 
  Much better. :-)
 
 I've butchered it further, but thanks for doing the *real* work:
 
 user:
 alias precmd 'set prompt = \n%{\033[32m%}%m [%h] [%@]%b%{\033[0m%} 
 [`pwd`]$ '
 
 root:
 alias precmd 'set prompt = \n%{\033[31m%}%m [%h] [%@]%b%{\033[0m%} 
 [`pwd`]$ '

Allow me a final note:

It's normal to denote non-root access with % (for csh) or $ (for
sh, bash and many others), and root access with #. You can easily
configure that to be automatically instead of $ if you like.

Example:

set promptchars = %#
set prompt = %n@%m:%~%# 

For root, # will appear at the end, and % for non-root. Of course,
you can easily apply this to your setting if you like, and you can
define other characters if needed (e. g. $# or #).

Even though in your prompt shown above, root is shown by red color,
that important attribute _might_ be missing when using a non-color
terminal or a misconfigured emulator, so the user might not be
aware of the immense power currently active (as no user name is
shown in the prompt). Maybe this inspiration is useful to you.




-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: CSH prompt

2012-05-13 Thread Polytropon

On Sun, 13 May 2012 14:56:31 +0200, Jos Chrispijn wrote:
 In cshell I use this prompt: set prompt = %B[%@]%b %m[%/] 
 The problem I face now is that if I use this prompt with symbolic links, 
 the presented location is displaying the symbolic link rather than the 
 real directory name.
 Is there a way of preventin this?

Yes, a very ugly way which I just found out:

alias precmd 'set WD=`pwd`; set prompt = %B[%@]%b %m[$WD] '

Example:

[3:21pm] r56[/] cd /sys
[3:21pm] r56[/usr/src/sys] _

It redefines the whole prompt at any command that could
affect the current working directory (not only cd can
do that). This is needed as any call to `pwd` stored into
a variable will only affect $prompt once - this is when
it's set, only at this time $WD would be evaluated. So
that's why this strange command. :-)

Oh, and I just improved it. How about this?

alias precmd 'set prompt = %B[%@]%b %m[`pwd`] '

Much better. :-)



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: file permission template

2012-05-12 Thread Polytropon

On Sat, 12 May 2012 23:37:00 +0900, fake fake wrote:
 I need a sort of file permission template.
 Under some particular directory (like ~/secret), I need all those
 files (including newly creating one) mode 700.
 Is there any template-trick? Or chmod -R 700 every time?

Depending on your shell, there is a umask command that
can be used as a template. For example, if you're using
the default dialog shell csh, put the required umask value
into ~/.cshrc. Note that this will cause _all_ file creations
by that user to have that predefined value.

See man csh for details. (In case you're using bash or a
different shell, consult the respective documentation.)


-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: file permission template

2012-05-12 Thread Polytropon

On Sun, 13 May 2012 00:15:54 +0900, fake fake wrote:
 Thanks. But I need specific directory only.
 umask way seems to set mode not only under ~/secret  but other
 directories like ~/public.

You're sure you want to have something _public_ in your
home directory?


 Is there any elegant way?

Depends on how the files are created. A possibility is to
set umask prior to creating files, and resetting it to its
previous value when being done. If files are created
automatically, this could be done by a shell script. Such
a script could also be used to copy to secure directory,
performing the cp and the chmod step.

However, is there any problem _for your particular case_
that setting secret/ to rwx/-/- only, and leaving the
files inside with the default umask rw/r/r?

Maybe there really is a more elegant way.

-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: djvu viewers from ports or add capability to view to xpdf, gv or other

2012-05-10 Thread Polytropon

On Thu, 10 May 2012 09:04:16 -0500, Antonio Olivares wrote:
 I had a gut feeling that viewers like gv or xpdf would have the
 capability to view djvu files, since evince could do this, but I guess
 I was wrong.

As far as I know, the ability to deal with this file formate
requires the corresponding library to be used. Evince seems
to be able to, but xpdf and gv are just PS/PDF viewers, so
this functionality hasn't been incorporated.

Even though ImageMagic is a heavy chunk of compiling, it
is acceptably easy to use when installed (display command).
Enabling DJVU option and recompiling it shouldn't pull too
many dependencies into the system.


-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: User can't login but /etc/(master.)passwd OK

2012-05-09 Thread Polytropon

On Wed, 9 May 2012 14:04:35 +0200, n dhert wrote:
 the entries for 'THATUSER in /etc/passwd and /etc/master.passwd are still
 the same as from a backup
 of 14 days ago (no change in encrypted passwd)
 The /etc/pwd.db and /etc/spwd.db  are binary files so I can't check..

You can easily rebuild them from the text files using pwd_mkdb.
Is /ect/group also okay?



 The home-directory of THATUSER is still present and contents looks normal ..

Does the home directory itself (the path leading to it)
also look correct (owner  permissions)?



 All users have quota, but for this particular user:
 # quota -v THATUSER
 responds
 quota: THATUSER: unknown user
 # edquota -u THATUSER
 edquota: THATUSER: no such user
 # repquota /home does not show that user anymore

Maybe a side effect? Can you provide more error messages
maybe? Does /var/log/messages or /var/log/auth.log show
something relevant when the user in question attempts an
login?




-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: UFS Crash and directories now missing

2012-05-09 Thread Polytropon

On Wed, 9 May 2012 09:30:37 -0400, Alejandro Imass wrote:
 On Wed, May 9, 2012 at 8:53 AM, Erich Dollansky
 er...@alogreentechnologies.com wrote:
  For your recommendation above, what are the advantages or differences
  of slicing the disk versus partitioning on a single slice?
 
  it could be a misunderstanding. What is a partition? What is a slice. I 
  have to look always into the handbook. Anyway, as long the OS see different 
  units which have to be mounted independent of each other, it all does not 
  matter what is what.
 
 
 I meant in Unix terms of course. Slice is slice (partition in other
 OS) and partition a thru h
 
 The question is if it has any advantage of using a slice to mount the
 basejail in RO as opposed to doing the same thing on a partition.

The answer is: It it not possible. :-)

You cannot mount a slice.

Given the BSD terminology: A slice _has_ to contain partitions.
You cannot format a slice, you can only format partitions. A
formatted partition carries a UFS file system. (However, it's
possible to omit the slice, and partition the whole disk instead,
this is called dedicated mode). A third method is formatting
the whole disk (the 'c' device), in that case the 'c' is omitted.

The _only_ time you can mount a slice is when it is used in its
common meaning, being a DOS primary partition; in this case,
a FAT or NTFS file system will be placed directly into a slice,
as those do not support any (BSD-style) partitioning.

/dev/ad0- the disk
/dev/ad0s1  - 1st slice
/dev/ad0s1a - 1st partition on 1st slice
   THIS is something you can mount.
-or-
/dev/ad0a   - 1st partition on disk (dedicated)
   THIS can also be mounted.
-or-
/dev/ad0- the whole disk (equals /dev/ad0c)
   Even THIS can be mounted.

In case I'm misunderstanding your question, could you alter the
expression?



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: djvu viewers from ports or add capability to view to xpdf, gv or other

2012-05-09 Thread Polytropon

On Wed, 9 May 2012 16:35:03 -0500, Antonio Olivares wrote:
 Dear folks,
 
 How can I add the capability of viewing djvu files to say xpdf or gv
 without installing evince?
 
 Or is there a small djvu viewer available in ports that can be installed 
 easily?

You can install the port ImageMagick with support for djvu
format. To view a file, simply call display file. Not tested.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: maybe not truly freebsd related

2012-05-08 Thread Polytropon

On Tue, 8 May 2012 16:28:08 +0200 (CEST), Wojciech Puchar wrote:
 But yet - what graphical mail program can you recommend that have such 
 simple basic functionality of local mail support in Maildir format?

I'm using Sylpheed here. It requires Gtk 2 (which should be
fine when you're using Gnome anyway), and it stores mails in
MH format (quite comparable to Maildir). Related to TB, it's
still very lightweight.

There has also been a Gtk 1 version (much more lightweight),
but I think it's already out of ports, and its UTF-8 support
does not exist. However, it's even faster than the current
version. :-)

Remember that it's a MUA. It's not a calendar, not a web
browser, not a multimedia player and not a PDF viewer. (But
you can interface it to open content based on file type by
using external programs, such as xmms, xpdf, xzgv etc.).


-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: error: libm.so.4 needed by libspeex may conflict w libm.so.5

2012-05-08 Thread Polytropon

On Tue, 8 May 2012 21:04:18 -0400, David Banning wrote:
 If it's libspeex / libm.so.4 that is causing the error (thats what it looks 
 like to me)
 I wonder how to find out what libspeex is being used for - or for that matter 
 what 
 libm.so.4 is needed for.

I see you start exploring the joy of front page decisions based
upon information provided exactly there. :-)

Speex is a codec intended for speech compression (and libspeex is
its corresponding library implementation), and libm is the math
library of your FreeBSD system (OS, not a port).

However, I have mplayer 1.0.r20110329_3 installed here on FreeBSD
8.2-STABLE (i386), using speex 1.2.r1_3,1. If you have problems
installing it, and you know you're not going to need it, just do
a make config in the mplayer port's directory and deselect SPEEX
option, then it shouldn't be built.

Is your ports tree up to date? Maybe there's a newer version of
speex or mplayer that will happily work with the system's libm v5?



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: kernel configuration file

2012-05-07 Thread Polytropon

On Mon, 7 May 2012 15:01:31 -0500 (CDT), Robert Bonomi wrote:
 
 Bernt Hansson b...@bananmonarki.se wrote:
  2012-05-06 20:23, Robert Bonomi skrev:
 
   Including *every* loadable module, whether or not you actually use it.
 
  That's not really true, at least not for me, and I have not made any
  changes to the build environment. The loadable module that I actually
  use is bktr.ko, that one among others does not get built.
 
 I'd guess that bktr.ko is a 'third-party' module, found in a port, and not
 part of the base system.

No, it's part of the base system. I've been using bktr _in_ kernel
for many years (FreeBSD 5 and 7), but since 8.0, it does not build
anymore. However, the module _does_ correctly build.

The documentation is in man 4 bktr.

A typical use (with the PAL option, because I don't have
Never The Same Color here), did work in the past like this:

device  bktr
options BROOKTREE_SYSTEM_DEFAULT=BROOKTREE_PAL
options BKTR_USE_PLL
options BKTR_GPIO_ACCESS
options BKTR_USE_FREEBSD_SMBUS

Today, I need to use /boot/loader.conf with those entry

bktr_load=YES

Works for my Haupauge WinTV PCI video + tuner card, even
the options (PAL) seem to magically work! :-)



 I found that every loadable kernel module in the base system is, or at least
 was, rebuilt. 

That's correct so far. Additionally, all components specified
by the kernel configuration file will be rebuilt, which in case
of _no_ alteration is the content of GENERIC. As I said, there
may be parts that one can safely drop (e. g. WLAN, floppy, ISDN
or sound for a server).





-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: kernel configuration file

2012-05-06 Thread Polytropon

On Sun, 6 May 2012 13:23:08 -0500 (CDT), Robert Bonomi wrote:
  From owner-freebsd-questi...@freebsd.org  Sun May  6 08:36:52 2012
  Date: Sun, 6 May 2012 09:34:12 -0400
  From: Carmel carmel...@hotmail.com
  To: FreeBSD freebsd-questions@freebsd.org
  Subject: Re: kernel configuration file

  On Sun, 6 May 2012 08:08:31 -0500 (CDT)
  Robert Bonomi articulated:

  If you use the traditional kernel-huild 'Configure/make depend/make'
  sequence, to rebuild the kernel -only-,  its a matter of one minute or
  so on a _slow_ (486-class) machine.

  you'll either get a Configure error, a linker error, or it 'just
  works'.

  OK, now you lost me. I use the following basic sequence:

  make buildworld
  make buildkernel KERNCONF=CARMEL
  make installkernel KERNCONF=CARMEL
  make installworld

  I am sorry, but I am not fully comprehending what commands you want me
  to enter.

 That's the 'modern' way. 

The /usr/src/Makefile contains a comment header which
explains the purpose of the make targets the current
way supports. One should read it before starting, because
it's quite informative on _that_ way of doing things
(e. g. make kernel = make buildkernel installkernel).

 Note: make buildkernel forcibly rebuilds everything, *EVERY* time.
 Including *every* loadable module, whether or not you actually use it.
 Which can be *really* painful on slow hardware  (like 20+ *hours*, on a 
 486-class machine).

Maybe it's worth mentioning /etc/src.conf and /etc/make.conf
and the man src.conf manpage. That is a comfortable means
to avoid building (and therefore also installing) modules one
does not need. The approach to configure all and _only_ the
stuff I need in a custom kernel can be followed this way,
and it will even work with the current make target way.
Have no WLAN? So why bother building it? No ISDN? Omit it!
For minor kernel changes (e. g. if you want to try some
compile-time settings), this approach is really handy as
it minimizes the time required.

This consideration should _boost_ build+install times on
current plentycore multiprocessors with tons of RAM! :-)

-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Off topic: NetBSD or OpenBSD for Alpha server ?

2012-05-05 Thread Polytropon

On Sat, 05 May 2012 19:20:10 +0200, Kenneth Hatteland wrote:
 The idea of installing FreeBSD 6.4 and experiment with upgrading to7.x 
 and above appeals to quite a lot. If anyone have tried this I`d like to 
 know if it is doable. I guess I`ll pick up the server one of the coming 
 days.

It should be useful to pay attention to all security considerations,
and of course to features that the _software_ you want to run might
require from the OS.



 The tip on using OpenVMS is okay, I googled it. But this seems to be a 
 commercial OS, and I have no money to spend on it, and I get the server 
 for free to play with. So BSD will be fine.

OpenVMS offers, if I remember correctly, hobbyist licensing
which is less expensive than the commercial licensing.

Additionally, I've heared of FreeVMS, but I'm not sure if it's
still in development and will run on your hardware. It's supposed
to be a free (of costs) VMS-compatible operating system, if I
remember correctly.



 I`ll try FreeBSD first, and OpenBSD next I think if the experience of 
 FreeBSD 6.4 and above is not totally pleasant...

Try installing the OS, then continue with finding out what
specific software (from ports or packages) you'll need. Update
the system if needed, or if you're okay with a not so current
system, just leave the software as-is, if it fits your needs.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Best mail setup for home server?

2012-05-05 Thread Polytropon

On Sat, 05 May 2012 10:21:10 -0500, Joshua Isom wrote:
 I currently use my FreeBSD system as my generic unix server and some 
 coding, along with occasional multimedia.  I'd installed postfix years 
 ago and kept using it.  Right now, I use getmail with cron, dspam, and 
 dovecot to handle my gmail account.  I've never set up outgoing mail 
 which makes changing email clients, or devices, annoying.  Currently 
 postfix is set to use dovecot's deliver command so that dovecot can sort 
 and handle it.  Before I deal with setting postfix to relay the mail, 
 dealing with firewalls and other possible issues, is there a better 
 alternative?  I'd prefer that local mail just works even if I lose 
 internet, and any email that gets as far as my server will at least 
 eventually mail.  The archlinux wiki seems to suggest ssmtp doesn't work 
 properly with attachments.  Instead it recommends msmtp, which requires 
 an active internet connection to use.  Dragonfly's dma is local only to 
 the computer and not the LAN.  Are the only options configuring sendmail 
 or configuring postfix?

As it has been explained already, home _server_ in regards
of e-mail makes certain assumption on what you _should_ do.
Since dynamic IPs have become the main source of spam (and
spam the main amount of e-mails transferred), sending from
a dynmic IP might fail due to mail servers refusing to talk
to your box. Furthermore, connection might drop is also
a bad idea for a server. If problems in mail transmission
occur on the way, notifications will be addressed to your
server, and if it's currently not reachable, a problem for
the other mail server arises, maybe even in blacklisting
your machine.

I've had a comparable solution when I was at university,
behind a static IP: directly sending mail was no problem,
and for receiving I did use fetchmail. That combination
made me fully independent in choice of MUAs (and when paying
attention to local storage formats, they all could work on
the same mail data). I've been using an external server
for actually hosting the mailbox (emptied by POP), so
_that_ functionality (receiving messages on my _own_
system) was not in my scope at that time. However, with
proper masquerading _any_ MUA could send to localhost,
and even ls /some/stuff | mail -s stuff b...@example.com
was possible.

After moving, I only had dynamic IP, resulting in the
observation that my setup didn't work for _some_ targets
anymore, as they refused to accept messages from dynamic
IPs. So I reconfigured sendmail to just send the messages
to my ISP's MX. That mail relay _has_ a static IP. The
downside: You won't be able to control the arrival of your
messages; only successfully transmitted to relay will
be in the logs. You can see advantages and disadvantages
in this approach: local storage, requirement for permanent
and reversable connection (proper DNS records highly
suggested!) and being tied to ISP's MX.

Maybe you should rething your operations ideas with the
suggestions given on the list. There are some things to
consider, but what you're basically planning is possible
without much trouble, as long as you pay attention to
the protocol. :-)



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: freebsd-update not updating reported patchlevel

 freebsd-update), or wider steps, e. g. from
8.3 to 8.4 (using sources per CVS)?



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: freebsd-update not updating reported patchlevel

First of all, thanks for explaining your point of view.
Allow me to add a few thoughts:

On Fri, 4 May 2012 11:44:49 -0500 (CDT), Robert Bonomi wrote:
 
 Polytropon free...@edvax.de wrote:
  On Fri, 4 May 2012 04:14:05 -0500 (CDT), Robert Bonomi wrote:
   What is required is a differentation between the _kernel_ revision level,
   and the patchlevel of the entire base system.
   
   Store the kernel revision level -in- the kernel.  Use the 'standard'
   THREE-level version numbering  {Major}.{Minor}.{revision} for the kernel.
   Bump 'revision' for each set fo kernel patches.
   
   The patchlevel info for the base system can be a simple data file.
   I'd suggest a dotfile' in /etc, mode 644, with the followig flags
   set: 'system append only', 'system undlink'.
   
   Bump 'patchlevel' every time -anything- in the base system changes,
   regardless of whether it is part of the kernel or the 'world'.
 
  Interesting approach. Both files could also be header files
  in /usr/include to store this information per #define. But
  in fact, I like the /etc idea better.
 
 The 'state of the kernel' _belongs_ in /usr/src/sys, or similar. to be
 included in kernal builds, and where the *handful* of utilities -- e.g. 
 lsof -- that are intimately coupled to the exact O/S version are already
 picking up 'system specific' gory details. 

Correct. I appreciate the idea of having _one_ centralized
point for that information that is authoritative regarding
all queries. Like uname displays several aspects of the
kernel's data, it is limited in some regards:

For example, if you have updated the system the binary
way to -p3 which included a kernel change, uname will
report that -p3 properly. If you follow -STABLE, you
don't get the information of what build you currently
have, so you cannot put it into relation after what
-plevel we currently are.

% uname -r
8.2-STABLE

I know there is some file in /usr/src where the build
number can be obtained from (I think it's a #define),
but it's not included in the kernel queryable data.



 /usr/include is definitely a 'wrong place'.  Arguably, so is /etc. 
  From the standpoint of 'a single place' for critical data, anything other 
 than a kernel build should use what is in the 'uname' output. (See the
 notes on O'Brien, below.)

 _Very_few_ applications are concerned with the patchlevel of 'world'.
 rebuilding everything that #included a 'world patchlevel' file, when
 the only thing that changed was the patchlevel, is just plain silly.


Oh, I didn't think about recompiling any stuff against
such a header file. I did primarily assume it as a kind
of purely informative source, which could also be provided
by a plain text file.




 *PROPERLY* USED, CVS keywords provide automatic inclusion of this 
 information -- for _every_ source module (.c or .h, and equivalents for
 other languages) in every executable build.

Correct, but obtaining such data is often not possible by the
application itself (except it has an extended version option
or it includes that info in a help screen).

For the kernel, uname prints various information (which are
obtained from the kernel directly, which is good), but what
program can do the same for the system?



 See above.  
 Done 'right', this stuff is already all there, with _existing- tools.

Not fully, if I see it correctly. E. g., what build number
has a particular -STABLE installation? Or, if kernel and world
are able to be updated independently - no kernel change, but a
program change from -plevel to -plevel+1 will leave the
kernel's uname -r at -plevel, so how to tell easily that
the world is at -plevel+1?



  Also very nice. By simply _viewing_ the file, the most non-current
  version will be discovered, so maybe (just _maybe_) re-ordering them
  in upside-down (newest version on top) would be better?
 
 Definitely -not-.  grin
 
 You obviously didn't notice that the file flags are 'sysem append only'.

Oh, I noticed that, and I know appending on top is always
more complicated than appending (in the precise sense of what
to append means). :-)



 The entire point of my proposal is to make it an IMMUTATABLE RECORD of
 'what was done'.  'add to top' has several disadvantages.  First,
 a performance issue, you do have to read down the log to find the 
 first 'END' line rather than being able to seek directly to it.
 Second, and the *BIG* one, you risk destroying the prior information
 by re-writing the file.  Third, it makes it easier for a 'malicious'
 update to cover it's tracks.

Additionally, _undoing_ operations would also be logged - not
by omitting lines, but by a proper record that states how things
have been reverted to a previous level, which is also very good
for diagnostics.



 Until you learn to think like O'Brien, staying ahead of him requires a
 -lot- of forethought.

Oh, I often think like O'Brien, and I don't remember, especially
when I'm talking to 6079 Smith W., machen Sie die Augen auf! :-)

On topic again

Re: freebsd-update not updating reported patchlevel

On Fri, 4 May 2012 16:45:51 -0500 (CDT), Robert Bonomi wrote:
 
 Polytropon free...@edvax.de wrote:
 
  First of all, thanks for explaining your point of view.
  Allow me to add a few thoughts:
 
  On Fri, 4 May 2012 11:44:49 -0500 (CDT), Robert Bonomi wrote:
   
   Polytropon free...@edvax.de wrote:
On Fri, 4 May 2012 04:14:05 -0500 (CDT), Robert Bonomi wrote:
 What is required is a differentation between the _kernel_ revision 
 level,
 and the patchlevel of the entire base system.
 
 Store the kernel revision level -in- the kernel.  Use the 'standard'
 THREE-level version numbering  {Major}.{Minor}.{revision} for the 
 kernel.
 Bump 'revision' for each set fo kernel patches.
 
 The patchlevel info for the base system can be a simple data file.
 I'd suggest a dotfile' in /etc, mode 644, with the followig flags
 set: 'system append only', 'system undlink'.
 
 Bump 'patchlevel' every time -anything- in the base system changes,
 regardless of whether it is part of the kernel or the 'world'.
   
Interesting approach. Both files could also be header files
in /usr/include to store this information per #define. But
in fact, I like the /etc idea better.
   
   The 'state of the kernel' _belongs_ in /usr/src/sys, or similar. to be
   included in kernal builds, and where the *handful* of utilities -- e.g. 
   lsof -- that are intimately coupled to the exact O/S version are already
   picking up 'system specific' gory details. 
 
  Correct. I appreciate the idea of having _one_ centralized
  point for that information that is authoritative regarding
  all queries. Like uname displays several aspects of the
  kernel's data, it is limited in some regards:
 
  For example, if you have updated the system the binary
  way to -p3 which included a kernel change, uname will
  report that -p3 properly. If you follow -STABLE, you
  don't get the information of what build you currently
  have, so you cannot put it into relation after what
  -plevel we currently are.
 
  % uname -r
  8.2-STABLE
 
 uname -v, maybe ??

Like uname -a (maximum output), only the date of the kernel
build is present. I'd like to know that strange number and
how it relates (pre-/postdates) -plevel patch levels.



 If you're talking about trying to associate a particular patch/revison
 level of a particular program with a partiular 'world' patchlevel.  That
 is a very different problem, and requires a separate separate solution,
 something like a 'correlation' database.

Yes, that was my primary intention.



  For the kernel, uname prints various information (which are
  obtained from the kernel directly, which is good), but what
  program can do the same for the system?
 
 For kernel info, any program that can 'popen' for write uname -a.  *grin*
 For the patchlevel of the 'world', TTBOMK it isn't recorded anywhere
 conveniently accessible. 

I know that this build number is stored somewhere (I found it
once!), I think it was a header file. Sure, you can grep for it,
but it would be easier to make this information better accessible
(and maybe even to put it into relation to a patch level number).



  Not fully, if I see it correctly. E. g., what build number
  has a particular -STABLE installation? Or, if kernel and world
  are able to be updated independently - no kernel change, but a
  program change from -plevel to -plevel+1 will leave the
  kernel's uname -r at -plevel, so how to tell easily that
  the world is at -plevel+1?
 
 It doesn't presently exist.
 
 That's precisely what the solution I proposed addresses.
 
 In the complete solution I proposed, 
'tail -1 /etc/{patchlog'
 
 Or, for a program,
one can popen() that command, and read the output
 or even
 #include sys/patchlog
 #include stdio.h
 
 fd=fopen(PATCHLOG,r);
 fseek(fd,PATCHLOG_LAST,SEEK_END);
 fgets(line,sizeof(line),fd)

So when does it arrive in -CURRENT? :-)



   The entire point of my proposal is to make it an IMMUTATABLE RECORD of
   'what was done'.  'add to top' has several disadvantages.  First,
   a performance issue, you do have to read down the log to find the 
   first 'END' line rather than being able to seek directly to it.
   Second, and the *BIG* one, you risk destroying the prior information
   by re-writing the file.  Third, it makes it easier for a 'malicious'
   update to cover it's tracks.
 
  Additionally, _undoing_ operations would also be logged - not
  by omitting lines, but by a proper record that states how things
  have been reverted to a previous level, which is also very good
  for diagnostics.
 
 If it's being done by automation, it can either log all the individual
 'undo' changes, or just log a 'reverting to patchlevel {foo} line.
 There are benefits to both approaches.
 
 If it's a 'manual' reversion, there's no way to guarantee anything gets
 added to the log.

Let's assume that the standard ways (freebsd-update, make installworld

Re: Off topic: NetBSD or OpenBSD for Alpha server ?

On Fri, 4 May 2012 17:11:00 -0500 (CDT), Robert Bonomi wrote:
 For obselete hardware one frequetly has no alternative but to run an
 obselete operating system.

Depending on the actual intention of use, it _may_ be no
problem to use obsolete operating systems and software.
(For example, I still have a FreeBSD 5.4 system with lots
of applications installed, perfectly working on a 300 MHz
system, intended for special purposes; I would _never_
use that as a server facing the Internet!)



 The OP has already decided on a *BSD.  Recommending VMS, of any form, is 
 not a 'helpful'/'responsive' response to his questions.  You *don't*know*
 _why_ he has selected *BSD, so you have _no_ idea whether VMS is viable
 or his needs.
 
 Given that he -needs- a *BSD on _that_ hardware which which 'flavor' would
 you recomend?  Or would you insist he discard that hardware and replace
 it with something current?   inquiring minds want to know.  *grin*

It there is a _required_ reason to run Alpha hardware, an
older FreeBSD OS isn't a bad choice. Depending on the
availability of sources (per /usr/ports of _that_ version)
or of packages (from the installation media of _that_ version,
or $PACKAGESITE pointing to the correct archives on the FreBSD
FTP server), software can be installed. There's also the
excellent tool portdowngrade. However, it may be a try  miss
to find out what software still runs, what _current_ software
can be made running, and what operation procedures still work.
This _ALL_ depends on what the system should be used for.
Only the OP can decide about what applies, and what doesn't.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: laptop very hot and noisy

2012-05-02 Thread Polytropon

On Wed, 2 May 2012 06:19:50 +0700, Erich Dollansky wrote:
 Hi,
 
 On Tuesday 01 May 2012 20:52:11 Polytropon wrote:
  On Tue, 1 May 2012 13:41:11 +0100, Anton Shterenlikht wrote:
   On Tue, May 01, 2012 at 08:25:11AM -0400, Alejandro Imass wrote:
  
  Not a big issue. Make sure you can remember which parts belong where.
  Make photos if it helps you, or draw some notes. If possible, find
  the service manual of the device and use it as orientation. But I
  think such kind of documentation is no longer part of the end user
  book present. :-)
  
 you cannot say this in general.

I didn't intend to. From my limited experience (considering modern
home consumer throw-away laptops and netbooks) there's hardly any
usable documentation. A start-up guide is among the few printed
materials. DVDs often contain drivers and a few instructions (e. g.
how to plug in the power supply), but things starting with opening
the device are typically left out. However, I welcome manufacturers
providing service manuals so a skilled user can use them. A typical
problem (as you described) can appear when special screwdrivers,
glue, spare parts or other tools are needed for repair that cannot
be purchased freely (or easily). In such cases, repair attempts
would often be more expensive than replacing the whole device.



  I've been lucky exploring that my new Lenovo Thinkpad T61p can be
 
 This is a different class of machines. They are made to be
 repaired and they are very large.

Both is correct, and I'm happy of that. :-)



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: UFS Crash and directories now missing

2012-04-30 Thread Polytropon

On Mon, 30 Apr 2012 13:23:40 -0400, Eitan Adler wrote:
 On 30 April 2012 07:36, Robert Bonomi bon...@mail.r-bonomi.com wrote:
  A competennt, not stupid, sysadmin would know these things.  And not
  'remove all doubt' (in the words of Abraham Lincoln), by raising such
  nonsense questions.
 
 A competent sysadmin would ask questions when they don't know the
 answer bringing up possibilities they thought about.
 A stupid sysadmin would yell at someone asking a question claiming
 they should have known the answer.

I know I don't add anything substantial by the following
statement, but allow me to post it anyway in addition to
your statement:

There is no problem in mentioning thoughts, possibilities
and options. It's also not a problem to admit a lack of
knowledge in certain fields (e. g. how UFS, journaling,
nullfs and fsck do interact with each other).

Things start to be problematic when conclusions are made
out of untrue assumptions or expectations. It must be
a system error, as I don't see a human error here.
The problem is: don't see != doesn't exist, and of
course != can't be proven. Such kinds of conclusion
often lead into wrong directions.

Of course it's hard to narrow down possibilities. A test
bed with limited variables is neccessary to have. Also
the proper tools and procedures of testing are important.
That's the ONLY way to be sure - by eliminating one
possibility after the other. What's being found in the
end - and even if it's regarded unprobable from the
beginning - must be the reason.

Robert mentioned important things to consider. If you
(unintendedly) destroy evidence for a forensic analysis
of what happened (whatever it may be), you'll have a
hard time finding out _what_ happened - except you can
get it to happen again. In case of security breaches
this is something you _don't_ want to risk IN PUBLIC
just to be able to observe it.

At this point, one could argue politeness vs. importance
of arguments. From what I've seen on other lists, Robert's
statements are still polite and full of things you can
take as a start to what to additionally learn. You should
concentrate on that essence. If you take the time to do
your homework, you'll be better prepared _if_ such thing
should ever happen again. Finding out _what_ has happened
is very hard (which I admit), and maybe it's even impossible.
You would have needed a more verbose auditing facility to
find out what program (user) caused a mv-like syscall.
Command logs can be altered, logged syscalls... yes, it's
not impossible, but magnitudes _harder_ to remove trails.

By the way, I can understand the frustration when something
impossible happened and you never can _really_ say what
it was, hoping it would not happen again. I've experienced
such kinds of trouble myself. (That's why I'm on this list.)



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: UFS Crash and directories now missing

2012-04-29 Thread Polytropon

On Sun, 29 Apr 2012 00:26:50 -0700, per...@pluto.rain.com wrote:
 Alejandro Imass a...@p2ee.org wrote:
 
  3) the directories were moved at reboot by journal recovery,
  fsck or something else
 
 I think it's *extremely* unlikely that fsck was involved, because
 it just doesn't do things like that. 

The point is: fsck moving directories looks different. In
case inodes get de-connected (their reference entries on
level n-1 are gone, their data on level n is still present),
fsck will access the lost+found/ directory in the corresponding
partition's root directory (or create it, if not present) and
write _new_ directory entries with the inode as their name,
because that's the only naming information possible (as the
original names on n-1 aren't accessible anymore). So those
directories will have names like #177628676/ and they _can_
contain subtrees full of data, _including_ names from levels
n+1 and onward. Files also are named #4767667892 and their
names can _maybe_ identified from their content (the file
command is helpful, and if they are textfiles containing
a CVS or other revision control system data tag, it's possible
to find out what they've been in their previous life).

However, as it has been explained, fsck will _not_ do so
unless being _allowed explicitely_ to do that kind of
MODIFICATION to the file system. Flags like -yf can do
that, but they are _not_ the default. This is due to the
fact that _any_ critical modification of file systems
requires the _responsible administrator_ to give permission.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: editor that understands CTRL/B, CTRL/I, CTRL/U

2012-04-28 Thread Polytropon

On Fri, 27 Apr 2012 18:36:13 -0600, Chad Perrin wrote:
 On Fri, Apr 27, 2012 at 06:00:51PM -0400, Jerry wrote:
  On Fri, 27 Apr 2012 14:33:29 -0700 David Brodbeck articulated:
  
  Again, this is one of the reasons credit scoring is becoming so
  popular -- it's an almost automatic way to narrow down the pile.
  Another method in common use right now is to throw out applications
  from anyone who's currently unemployed, and only look at ones who
  already have a position and are looking to change jobs.
  
  I have been told by several people in HR that the trend to give
  preference to those all ready working as opposed to the unemployed is
  based on the philosophy that if no one else will hire them, then why
  should we. While we could argue whether that logic is flawed, it is
  never-the-less presently in use. However, it doesn't really pertain to
  entry level openings. With the glut of individuals entering the job
  market, for an applicant to not be proficient in the skills being
  advertised for by the prospective employer is just a waste of time. If
  the employer is looking for skill A and B, crying to him/her that
  you have skill C is just a waste of both your times.
 
 It *does* pertain to entry level positions, because (from what I have
 seen) most entry level positions come with an experience requirement of
 at least two years.

But then this would invalidate ENTRY level. How exactly is
an applicant supposed to get a job from that entry level pool
when he doesn't have previous experience because he simply wants
to ENTER that field of profession?



 You speak as though you think they're correctly identifying the skills
 they actually need from their employees.  A big part of this entire
 discussion has been about the fact that many responsible parties in the
 hiring process are utterly without capacity for correctly identifying the
 skills they actually need to optimally fill the open positions.

Correct, at least that's my experience. To give you _few_ examples
which are more the norm than exceptions:

good MS standart knowledge
(Yavoll mein Hare Heiny Standart-Leader von Sowercrowd!)

programming knowledge in established programming languages, e. g. OS2
(cc hello.os2, and it's OS/2 with slash)

modern Microsoft operating systems (Windows 98 and XP)
(yes, _very_ modern and current; hey, it's more than 10 years old!)

extended basic knowledge
(so what, basic or extended?)

autonomous team-oriented working
(maybe as a one man team!)

It's funny when you encounter job offers by recruiters and HR
services who _fail_ to properly spell our native language, but
think they are in a positition to place _you_ (as a professional)
into a good job! Okay, it's NOT funny. It's also not funny if you
have to explain to such a senior consultant permanent placement
how to open a PDF file containing your application documents, and
it's even worse when they try to trick you to do their work, e. g.
enter all your data again into their (!) HR database.

As I said, the problem of the unclear expression _what_ skills
actually are needed can make it hard to properly apply for a job.
This problem isn't only present for written application, it's also
there if you get invited to an interview and the guy across the
table is simply asking the wrong questions, or unable to understand
your answers.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: editor that understands CTRL/B, CTRL/I, CTRL/U

2012-04-28 Thread Polytropon

On Sat, 28 Apr 2012 07:36:03 -0400, Jerry wrote:
 On Sat, 28 Apr 2012 07:41:18 +0200
 Polytropon articulated:
 
 On Fri, 27 Apr 2012 16:46:52 -0400, Jerry wrote:
  On Fri, 27 Apr 2012 13:58:40 -0600
  Chad Perrin articulated:
  
  On Fri, Apr 27, 2012 at 01:57:10PM -0400, Jerry wrote:
   On Fri, 27 Apr 2012 10:32:24 -0600 Chad Perrin articulated:
   On Thu, Apr 26, 2012 at 06:43:06PM -0400, Jerry wrote:
On Thu, 26 Apr 2012 15:52:56 -0600 Chad Perrin articulated:
On Thu, Apr 26, 2012 at 02:45:53PM -0700, David Brodbeck
wrote:
 
 Generic skills aren't recognized because they're hard to
 judge and test for.  People want quantifiable, objective
 things to weed out applicants.  This is also why credit
 scoring has become so popular -- sure, someone's credit
 score may not tell whether they'd be a good employee or
 not, but it's a convenient, objective way to throw out a
 bunch of resumes.

Indeed -- and the employer who bucks this trend does him/her
self a huge service, because large numbers of very skilled
and/or talented people are being rejected on entirely
arbitrary criteria that have little or no correlation to
their ability to do the job.  People who use such critera are
forcing themselves to compete with everyone else in the
industry using the same criteria, leaving a glut of job
candidates who would be great at the job waiting for someone
else to give them a chance.

Wouldn't it be far easier for this glut of job applicants to
either become proficient in the skills stated in the job
description for which they are applying or do what everyone
else does; i.e. lie on their résumé. If the mountain will not
come to Mahomet, Mahomet must go to the mountain.
   
   1. Pretty much every employer has a slightly different list of
   keywords. I guess you think all these job candidates should learn
   every skill in the world.
   
   No, I think they should learn the one(s) most sought after in
   their chosen field. If 90% of the potential openings in a
   specific field are requesting proficiency with MS Word, what do
   you think any legitimate applicants should become proficient in?
  
  Right -- because all the keywords you need will always be Microsoft
  Word.
  
  Admit it: you're just making up half-baked excuses to disagree now.
  
  If the requirement is for proficiency in MS Word, Excel or whatever
  and you lack those skills then you are not qualified for the job.
  Period.
 
 There are two problems hidden:
 
 1. You typically cannot learn proprietary products for free.
 Of course there are books and online material to help you, but
 you cannot try the software. You have to buy it, and you have
 to buy the OS that supports it. There is no (legal) way for
 autodidacts to make theirselves familiar by learning and doing.
 
 Irrelevant. You cannot learn to be a doctor, lawyer, physicist,
 etcetera sans an education. Unless you have managed to acquire a free
 ride, i.e. you are getting the education on someone elses dime, you
 will need to pay. Quite frankly Poly, I would have expected a better
 argument from you than that. It was really quite bogus.
 
 2. There are many different versions, so when you encounter
 Microsoft Word as a required skill, you cannot be sure that
 the skill _you_ have will be the right one. You know that
 products like Word differ from version to version. And of
 course they highly differ from established and standardized
 ways of doing things, so your generic knowledge (e. g. acquired
 by learning and doing OpenOffice or StarOffice or Abiword)
 isn't fully portable simply because of the arbitraryness of how
 Word does things.
 
 arbitraryness [sic} is one way of describing it. Since MS Office is
 the de facto  standard it can be stated that the other entries in the
 word processing field are guilty of arbitrariness in their approach to
 the matter.

I don't agree here. The history in UI and behavioural changes
in prograns like Word made whole generations of its users
nearly completely RE-learn what they already could do before,
worse or better. During the many versions things massively
changed, and there is no _the_ Word version you find un
business.

Putting formatting options into the File menu is one of such
things that I call arbitrary, because logic dictates that it
would be expected to be where the other formatting options
(typeface, selection, paragraph - page) are found. Something
similar can be seen for visualisation settings: some of them
are in View, some other aren't.

Standard (at least in my idealized opinion) also includes
file formats. Instead of memory dump blobs, programs like
OpenOffice use a publically documented format which makes
it easy to implement output processors for OO-files without
further problems.



 For the record, would you please point me to the RFC that
 gives the requirements for a word processor. I must have missed it
 somewhere

Re: UFS Crash and directories now missing

2012-04-28 Thread Polytropon

On Sat, 28 Apr 2012 13:52:02 -0400, Alejandro Imass wrote:
 On Sat, Apr 28, 2012 at 1:31 PM, Robert Bonomi bon...@mail.r-bonomi.com 
 wrote:
 
  Alejandro Imass aim...@yabarana.com wrote:
  On Sat, Apr 28, 2012 at 11:39 AM, Robert Bonomi
  bon...@mail.r-bonomi.com wrote:
    Alejandro Imass aim...@yabarana.com wrote:
   After a little more research, ___it it NOT unlikely at all___ that
   under high distress and a hard boot, UFS could have somehow corrupted
   the directory structure, whilst maintaining the data intact.
  
   This is techically accurate, *BUT* the specifics of the quote 
   corruption
   unquote in the case under discussion make it *EXTREMELY* unlikely that 
   this
   is what happened.
  
   99.99+++% of all UFS filesystem corruption' issues are the result of a
   system crash _between_ the time cached 'meta-data' is updated in memory
   and that data is flushed to disk (a deferred write).
  
   The second most common (and vanishingly rare) failure mode is a powerfail
   _as_ a sector of disk is being written -- resulting in 'garbage data'
   being written to disk.
  
   The next possibility is 'cosmic rays'.  If running on 'cheap' hardware
   (i.e., without 'ECC' memory), this can cause a *SINGLE-BIT* error in
   data being output.
  
   The fact that the 'corrupted' filesystem passed fsck -without- any 
   reported
   errors shows that everything in the filesystem meta-data was consistent
  
  [...]
 
   I think it is safe to conclude that the probabilities -greatly- favor
   alternative #1.
  
 
  OK. So after your comments and further research I concur with you on
  the mv but if it wasn't a human, then this might be exposing a serious
  security flaw in the jail system or the way EzJail implements it.
 
  BOGON ALERT!!!
 
 
 I admit my ignorance on how the filesystem works but I don't think
 your condescending remarks add a lot of value. The issue here is this
 actually happened and there is a flaw somewhere other than the stupid
 administrator did it.

If you search the archives of this list, you'll find my _first_
post to that list: I've had a similar problem, df shows data
must be there after crash (panic - reboot - fsck trouble), but
files aren't there (even _not_ in lost+found). It's quite possible
that in _exceptional_ moments this can happen. The fsck program
is intended to repair the most typical file system faults, but
nothing complicated will be done without interaction: Altering
data on disk will _always_ involve the responsible (!) admin to
check if it is really intended to do so.

There can be many reasons. I've never found out what was the
reason for the trouble I've had. Some years ago, I found a make
failing because /uss/src/blah... something not found, and
a quick memtest revealed the secret: defective RAM module that
caused a bit error, and the difference between r and s
is just one bit. Replaced the module - everything worked.
Mean soldering rays from outer space. :-)

You'll find many useful forensic tools in the ports collection
that might help locate lost data (quotes intended as long as
the data is still on the disk). The more complex your setting
is (e. g. striped disks, or ZFS), this can be nearly impossible.
Plain old UFS can sometimes be your saviour (but BACKUP should
be your real friend).





-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: editor that understands CTRL/B, CTRL/I, CTRL/U

2012-04-27 Thread Polytropon

) resume, written by a professional resume writer many
  years ago, absolutely astounded me. I had no idea I was as
  proficient and skilled in so many areas. As the writer explained, it
  is not what you say but how you say it. Just because I once wrote a
  two page article that got published in a cheap magazine does not
  mean that I am an accomplished author with numerous credits to my
  name -- or does it?
 
 No, it doesn't.  Maybe an accomplished author with one credit to your
 name.  Amusingly, that'll turn out to be a great way for employers to
 notice you're exaggerating with that accopmlished author bit, too.
 Only by lying (numerous credits) can you allay suspicions for a
 moment in those credulous enough to not ask for samples (which
 absolutely does not make it okay).
 
 Now you are being naive. There are numerous examples of people in both
 corporate and government jobs that have made out right lies as to
 their education, etcetera. Some of those frauds have gone undetected
 for years. The majority of resumes for entry level jobs are rarely if
 ever given more than a perfunctory look.

Again, I fully agree with you. Selling yourself on the HR market
includes the typical aspects of selling you'll find in consumer
products. For example, in marketing... let's say a tablet, the
manufacturer doesn't say you cannot remove the battery (which will
be flat line after 1 year of use), and the device will be unsupported
after 2 years of use; no, the manufacturer will only show positive
aspects of the tablet: it's shiny, slim, lightweight, entertaining
and so on. He will also exxagerate, e. g. it's the world's most
popular, future-proof, revolutionary and so on.

Doing something _comparable_ is fully valid in applications. Of
course there's also fraud to be noticed, e. g. doctors who haven't
studied medicine (happened in Germany), people who are dumb as
bread and too stupid to hammer a nail into the wall, but being
awarded manager of the year and applying for an important
position. In the end, maybe they'll be successful in their
positions, but in many cases (and I also wrote this before)
you'll find people in workplaces they are _not remotely_
qualified for.

Employers have recognized that. They've risen the barrier for
entry. Even lower-end jobs now require higher levels of education.
For example, I've recently encountered a job offer for a thing
called virtualization administrator (system administrator)
which turned out to be phone 1st level support. The requirement
however was: university degree or professional education with
experience. Interesting for something that even Timmy Dumbass
could do: Read questions from a flowchart and mark the YES/NO
answers before transfering the call to 2nd level. Of couse I
don't have to tell you that this particular job won't be paid
as other jobs typically done by people successfully leaving a
university. This particular job was underpaid.



 The bottom line is if you want a job, you either learn or acquire the
 criteria required for the job, or find a way to BS your way into it
 and hope you can pull it off. No legitimate employer is going to change
 his criteria to accommodate your skills.

Employers often have strange expectations. I also wrote that
some of them, because a shortage of skilled programmers,
suddenly want the geek (who trained himself lots of programming
languages and development methods in his free time), but they
want him to have certificates and university degrees. Reality
shows that the _really_, I mean ***REALLY*** good programmers
often don't have any degrees at all, sometimes even no professional
education! Those promising candidates drop out at the beginning if
they don't improve their CV or resume. It's the only chance they
can turn their knowledge and experience into money (by being
employed by a boss who _recognizes_ what he can get). Needless
to say that such skills aren't taught in schools, universities,
professional education and IT courses. You can only teach them
to yourself.


-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: need help on installing bsd in virtual box

2012-04-27 Thread Polytropon

On Fri, 27 Apr 2012 23:01:28 +0100 (BST), dhillon sandeep wrote:
 Hi,
    I am struggling in installing the bsd in virtualbox
 i am totally new to unix, but have previously installed
 Ubuntu linux in my virtual box, i have dounloaded both the
 images bootonly and release iso, after creating the new
 virtual machine and starting it for first time and by
 selecting the iso image nothing is happening or getting
 installed in the virtual machine. I downloaded the iso
 images from your BSD website. Please let me know what i
 am doing wrong or what do i need to do to install bsd.

As you have experiences with Ubuntu, maybe you're interested
in giving VirtualBSD a try? It's a preinstalled and preconfigured
image containing a FreeBSD installation. You can play it
with Virtualbox.

http://www.virtualbsd.info/

You can find instructions and screenshots on that web page.

Regarding the installation of a normal FreeBSD OS, refer
to the handbook with explains the basic steps of installation
and configuration.

http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/

http://www.freebsd.org/doc/en_US.ISO8859-1/books/faq/index.html

You need to properly configure your virtual environment and
pay attention to 32/64 bit when doing so. The bootonly image
is typically used to install the system via network, there are
no installation datasets on that media. The CD1 and DVD1 media
images will be the ones used in typical installations.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Python module wnck?

2012-04-26 Thread Polytropon

On Thu, 26 Apr 2012 13:37:44 + (UTC), Walter Hurry wrote:
 I'm trying to get screenlets up and running on 9.0, but I'm getting the 
 following Python error:
 
 ImportError: No module named wnck
 
 I believe that this should be supplied by a a package or port named 
 something like py27-wnck, but am unable to trace any such.
 
 Can anyone point me in the right direction please?

Use the ports, Luke. :-)

% cd /usr/ports
% make search name=wnck

And a result:

Port:   libwnck-2.30.6
Path:   /usr/ports/x11-toolkits/libwnck
Info:   Library used for writing pagers and taskslists

More checks:

% cat less x11-toolkits/libwnck/pkg-descr 

libwnck is a Window Navigator Construction Kit, i.e. a library used
for writing pagers and taskslists.  It is needed for the 
GNOME 2.0 desktop.

Does this look like what you're searchin for? I know it's
not specified to be a Python module, but maybe it interfaces
with Python somehow?

You can also use make search key=wnck to bring up a list
of keyword search results.


-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: editor that understands CTRL/B, CTRL/I, CTRL/U

2012-04-26 Thread Polytropon

On Thu, 26 Apr 2012 14:45:53 -0700, David Brodbeck wrote:
 On Tue, Apr 24, 2012 at 11:55 PM, Polytropon free...@edvax.de wrote:
  Thanks for that article, it's really sad. One of the main
  problems is (in my opinion) that GENERIC SKILLS aren't
  recognozed with the big importane they have.
 
 This applies to hiring as well as education.  When they read a job
 application, HR people seem to basically do keyword matching.  They
 don't know or care about generic skills. 

That's a shortsightet view, especially when you consider the
typical lifecyle of software. Being educated on one specific
version that doesn't share many similarities with competitor's
products or own follow-up versions, you're lost.

Generic skills (such as generic Linux and UNIX skills) enable
you to become familiar with _any_ Unix-like operating system
very quickly, and in a world software changing dayly this is
an important skill.

Additionally, generic skills enable you to learn _anything_
quickly, such as a new scripting language, or a DTP application.
They all share generic concepts (like some kind of syntax for
a programming language, or some kind of UI design for a GUI
based program).

And you're right: HR people don't do more than keyword matching.
That's the only thing they have time for.



 If the posting says
 'Microsoft Word experience' the words 'Microsoft Word' better appear
 somewhere in the resume. 

It's even worse. There are some standardized skill profiles
(which aren't standardized) that one is expected to include.
I currently have an example here. It contains 100 times the
word Microsoft, but lacks essential stuff that one would
assume when applying for a job as a virtualisation / system
administrator. Some non-MICROS~1 stuff is mentioned in footnotes,
most of it even improperly spelled or not attributed to the
proper company.

For example, if you're familiar with StarOffice, OpenOffice and
LibreOffice (which you can acquire knowledge in _for free_), you
should be able to conclude how the MICROS~1 products work, any
version of them (even though they are very different and incon-
sistent, and you _cannot_ learn them for free). So this would
match the skill office applications, but maybe because the
word Microsoft doesn't appear several times, this skill is
rejected.

This also works with commercial UNIXes that are hard to try
for free. But with your generic skills, you can find out how
things work, because the basics are the same everywhere. You
can even install Hercules on your FreeBSD machine and find
out how an IBM /360 mainframe is operated - teaching you basic
skills how to deal with z/OS, CMS, TSO, REXX, ISPF and other
(primarily commercial) applications you might encounter).



 Likewise, if they want experience with a
 particular programming language, you'd better have experience with
 THAT SPECIFIC LANGUAGE...never mind if you already know five and can
 pick up another in a week's time.

That is correct. But being able to do so depends on the
employer to _publish_ his expectations in an understandable
format. In a setting where job applications are typically
filtered by an external HR company which _also_ makes the
job announcement, you'll hardly find them. Instead, there's
lots of blahblah like we're an established company, a
prominent market leader or young and dynamic expanding
service provider - and then programmer or system
administrator. You often don't find any hint who the _real_
employer would be. And in the end, it turns out that they
are searching for a phone monkey in 1st level customer
support. :-)



 Generic skills aren't recognized because they're hard to judge and
 test for. 

Hard to judge - no, but only by try and watch which often
is not possible or not intended.

Hard to test for - true, as proper test would have to be
developed first, and I assume that's rather expensive.
There are generic tests like FizzBuzz, but it doesn't say
_that_ much, and it's not enough to use _only_ this test.
However, it's a nice fall-through test if you want to
hire a programmer and he doesn't get it done by any programming
language _he_ may choose. :-)

Generic skills are _the_ skills you need to learn something
new. Stupidly repeating things doesn't work. Being tied to
the one way of doing things doesn't fit a quickly changing
world. You can't rely on vendor lock-in everywhere.



 People want quantifiable, objective things to weed out
 applicants. 

They often _assume_ that this is provided by colorful paper,
typically hanging on a wall in your back, the wall of fame.
There are many certificates that state you actually know
something, but there are more than enough that just cost
money, and you get them, no matter what you know (certificate
spam, if I may say that) - those are _worthless_.

I think objective is very hard to find here. Many considerations
depend on assumptions and expectations. For example, you want
a programmer. You don't state for what precisely (kind of project
and programming language

Re: Was..... Lots of lagging after upgrade of xorg. Now keyboard layout is lost

2012-04-24 Thread Polytropon

On Tue, 24 Apr 2012 07:13:20 +0200, Leslie Jensen wrote:
 
 
 2012-04-23 19:56, Leslie Jensen skrev:
 
 
  2012-04-23 18:29, Warren Block skrev:
  On Mon, 23 Apr 2012, Leslie Jensen wrote:
 
 
   Use Option
  AutoAddDevices Off to disable HAL input device detection.
  ___
 
 
  http://www.wonkity.com/~wblock/docs/html/aei.html
 
 
 
 
 After adding the above Option I lost the Swedish layout of my keyboard.
 
 Following the instructions and editing the
 
 /usr/local/etc/hal/fdi/policy/x11-input.fdi
 
 
 I already have the file in place with the following setup:
 
 
 ?xml version=1.0 encoding=ISO-8859-1?
 deviceinfo version=0.2
device
  match key=info.capabilities contains=input.keyboard
 merge key=input.x11_options.XkbOptions 
 type=stringterminate:ctrl_alt_bksp
 /merge
merge key=input.x11_options.XkbModel 
 type=stringlatitude/merge
merge key=input.x11_options.XkbLayout type=stringse/merge
  /match
/device
 /deviceinfo
 
 
 Where else can I control the setting for Swedish?

You could use the default method: /etc/X11/xorg.conf which is
designed to _centralize_ X-related settings. Keyboard settings
can also be put there.

For example, this is what I use to define a german keyboard
layout (and which applies everywhere in X, as intended):

Section InputDevice
Identifier  Keyboard0
Driver  kbd
Option  XkbModel  pc105
Option  XkbLayout de
Option  XkbOptionsterminate:ctrl_alt_bksp
EndSection

You can employ this approach, changing it to swedish language.
Note that I'm using the X setup without HAL and DBUS here.

Additionally, there's the method of using xmodmap with a
custom ~/.xmodmaprc file which can be used to make keyboard
language settings work _independently_ from both xorg.conf
and XML files scattered across the local/ subtree. :-)



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Lots of lagging after upgrade of xorg.

2012-04-23 Thread Polytropon

On Mon, 23 Apr 2012 13:57:29 +0200, Leslie Jensen wrote:
 
 Hello list.
 
 I'm experiencing a lot of lagging after I've upgraded xorg. I use XFCE 
 and just changing from one desktop to another now feels really slow.
 
 If I click on a button in an application I have to move the mouse 
 pointer before there's a reaction to the click.
 
 In a terminal window I also have to move the pointer outside before the 
 input from the keyboard is registered.
 
 Maybe it's a mouse problem. Anyway I need some help if it's a setting 
 that has to be changed.
 
 My system, 8.2-RELEASE-p6, worked well before the upgrade of xorg.

Did you done any changes to the possible HAL settings (with
or without - needs to be set at compile time, and maybe xorg.conf
with some options)?


-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: converting UTF-8 to HTML

2012-04-22 Thread Polytropon

On Sun, 22 Apr 2012 11:45:45 +0100, Matthew Seaman wrote:
 On 22/04/2012 10:17, Erik Nørgaard wrote:
  UTF-8 is variable with, ascii characters are stored as single bytes (not
  sure about iso-8859-1) while other characters are stored as two byte chars.
 
 ascii uses the low 128 values that you can assign to an unsigned char,
 ie. those where the high-order bit is zero.
 
 iso-8859-1 and the various other iso-8859-X character sets fill in the
 remaining 128 characters with various other glyphs useful in latin
 alphabets, so it's still one char per glyph.  Other alphabets (greek,
 cyrillic, etc) have similar one byte-per glyph encodings. But you have
 to know what the encoding is to display the content correctly, and it is
 difficult to mix chunks of text in different encodings in the same document.

How about the extended ASCII character set that has a mixture
of non-US glyphs and semi-graphic symbols?

http://asciiset.com/extended.gif

This default layout isn't tied to a specific encoding, if I
remember correctly, or is it? Accessing the set as seen in the
picture allows using special character from many languages,
such as german umlauts and eszett, greek gamma and phi,
danish o-slash, swedish a-circle and even the yen symbol.
And the nice semi-graphic symbols to draw boxes and backgrounds,
as well as card deck symbols or the lazy L.

Of course, there are no arabic or chinese letters in there,
so it can be seen as a roman-derived language centrism
(targeting europe and america in the first place). All of
them are natively supported by graphic cards when running
in text mode, if my assumption is correct. So this extended
set of capabilities still is the most-minimum common
functionality that one can rely on.

(FreeBSD remaps some of the characters in text mode to display
the semi-graphic mouse pointer, so the full set cannot be
used all the time.)



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Dual monitors ok, but no mouse and keyboard action on the slave screen

2012-04-22 Thread Polytropon

On Sun, 22 Apr 2012 13:43:31 +0200, Kenneth Hatteland wrote:
 I`ve gotten a 17 inch monitor in addition to my 22 inch working with 2 
 separate desktops. I plan to have stuff like wireshark etc on the 
 smallest. But I have a problem, I can get no work done since I have no 
 mouse or keyboard working on the 17...
 
 Anyone have somewhere with a solution to point me towards ?

There are basically two kind of two-monitor settings: One
is to have the WM manage them, the other one is to concatenate
them to one logical screen.

I've been using the concatenated screen with two 21 CRTs,
each running at 1400x1050, so the result was a 2800x1050
ultra extended extraordinary super hyper big wide screen. :-)

You can configure this in your /etc/X11/xorg.conf (which you
can have X auto-generate).

For example, ServerLayout could contain

Screen 0 Screen0 0 0
Screen 1 Screen1 LeftOf Screen0
Option Xinerama on

Then add the two Monitor sections according to the screen
parameters (in my case, identical data).

In the final Screen section, you can then experiment with

Option TwinView
Option TwinViewOrientation LeftOf
Option ConnectedMonitor CRT, CRT

depending on your actual connection setup.



You can find more inspiration here:

Dual head issues, non-xinerama setup possible?
http://forums.freebsd.org/showthread.php?t=11567

Dual monitor setup
http://lists.freebsd.org/pipermail/freebsd-mobile/2005-January/005613.html

Dual monitors xorg.conf
http://lists.freebsd.org/pipermail/freebsd-questions/2005-May/087929.html

Using two monitors with X.org
http://www.freebsddiary.org/xorg-two-screens.php



Many things to consider depend on your actual setting (which
hardware you have, what WM you use and which behaviour you
want).



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: find sources to build Handbook and FAQ for FreeBSD?

2012-04-21 Thread Polytropon

On Sat, 21 Apr 2012 00:50:20 -0500, Antonio Olivares wrote:
 On Fri, Apr 20, 2012 at 8:45 PM, Warren Block wbl...@wonkity.com wrote:
  On Fri, 20 Apr 2012, Antonio Olivares wrote:
 
  Does anyone know where the source(s) for the FreeBSD Handbook and
  FreeBSD FAQ are found?
 
 
  SGML source is in /usr/doc/en_US.ISO8859-1/books/handbook/ and
  /usr/doc/en_US.ISO8859-1/books/faq/, or other subdirectories under /usr/doc
  for other languages.  There's build infrastructure in /usr/doc/share.
 
 [olivares@tricorehome ~]$ cd /usr/doc/share
 bash: cd: /usr/doc/share: No such file or directory
 [olivares@tricorehome ~]$ cd /usr/doc/
 bash: cd: /usr/doc/: No such file or directory

See the /usr/ports/misc/freebsd-doc* ports. They will install
the documentation in a freebsd/ subtree at the obvious location.

% ls /usr/local/share/doc/freebsd
de@  en@  faq@
de_DE.ISO8859-1/ en_US.ISO8859-1/ handbook@

As you can see from this example, I have the en and de
languages installed. The articles/ and books/ subtrees will
contain the HTML files.



  Some description about the doc tools is in the FreeBSD Documentation Project
  Primer at
  http://www.freebsd.org/doc/en_US.ISO8859-1/books/fdp-primer/index.html
 
 Thanks, but I am looking for \TeX{}/\LaTeX{} source files that are
 used to build the *.pdf versions of HANDBOOK,  FAQ.  If one does a
 properties on a PDF, we can see maker dvips + ghostscript 8.71.  This
 is what I am looking for, the files to produce that document[sources
 in tex/latex form] and see if I can produce it with what is readily
 available in kertex now.
 
 If I look in /usr/local/share/doc, these are not there either

I think that's because of the move of documentation out of the
base system, into separate ports for the supported languages.

SGML source files are in /usr/src/release/doc (part of
the system sources).


-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: converting UTF-8 to HTML

2012-04-21 Thread Polytropon

On Sat, 21 Apr 2012 09:10:03 -0500 (CDT), Lars Eighner wrote:
 On Sat, 21 Apr 2012, Erik Nørgaard wrote:
 
  When characters show up wrong in the users browser it's usually because the 
  browser is set to use a non-UTF-8 charset by default such as windows-1252, 
  the web server sends the charset=ascii in the http header and there is no 
  or 
  incorrect meta tag to resolve the problem. Non UTF-8 charsets are a 
  leftover 
  from last millenia that we sometimes still choke on .. sorry the rant ;)
 
 UTF-8 is a waste of storage for most people [...]

Disks and RAM are huge and cheap. Plenty of space that is
going to be used. Nobody cares.



 [...] and is incompatiple with
 text-mode tools: it's simple another bid to make it impossible to run
 without a GUI.

Again, nobody cares - until, of couse, it's too late and you
need to do some recovery or analytic tasks in a limited
environment or via a connection with limited means.

Regarding the fun of encodings, endianness, representation,
use (fi the two letters vs. fi the ligature, or ß
the 1-byte sequence vs. ß the two-byte sequence), see
the following document:

Matt Mayer: Love Hotels and Unicode
http://www.reigndesign.com/blog/love-hotels-and-unicode/

And finally it offers an interesting attack vector, given
the fact that several unicode characters look the same,
but in fact are different. So two files with the 'same'
name is a possible means that malware implementers can
utilize to mislead the users.

Short example from MICROS~1 land here:
http://blogs.technet.com/b/mmpc/archive/2011/08/10/can-we-believe-our-eyes.aspx

But this all doesn't negate the usefulness of unicode / UTF-8
in general. Especially when you have collaborative settings
with multi-language document processing requirements, it
is a helpful thing, as working with normal (ASCII) letters,
cyrillic ones, chinese and japanese symbols, arabic writing
is no big deal as long as all the tools do properly support
it the _same_ way.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Changing psm (mouse) resolution?

2012-04-19 Thread Polytropon

On Thu, 19 Apr 2012 14:25:48 -0700, Ronald F. Guilmette wrote:
 
 
 The man page for the psm driver says:
 
 ... The current resolution can be changed at runtime.
 
 Unfortunately, it fails to mention any sort of command line utility
 that would provide this functionality.

It's mentioned as

SEE ALSO
 ioctl(2), syslog(3), atkbdc(4), mouse(4), mse(4), sysmouse(4), moused(8),
 syslogd(8)

in the manpage, even though it doesn't explicitely state that moused
is the binary to run.



 Is there a command line utility that provides this functionality?  Or do
 I need to write one from scratch, using the ioctl calls that are documented
 in the man page?

See man moused for details, especially the -r resolution option
should be useful. You can combine it with -f for testing. According
to the manpage, something like

# moused -f -r 300 -a 2.0 -p /dev/psm0 -t ps/2

should be good for testing. To make the settings permanent, you can
code them into /etc/rc.conf, using

moused_enable=YES
moused_port=
moused_type=
moused_flags=

with the required values.


-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: FreeBSD's backwards webdesign / corporate identity

2012-04-13 Thread Polytropon

On Fri, 13 Apr 2012 22:59:41 +0700, Erich Dollansky wrote:
 On Friday 13 April 2012 20:56:35 Sean Cavanaugh wrote:
  
 Hi,
 
   -Original Message-
   From: owner-freebsd-questi...@freebsd.org [mailto:owner-freebsd-
   questi...@freebsd.org] On Behalf Of Erich Dollansky
   Sent: Friday, April 13, 2012 9:12 AM
   To: freebsd-questions@freebsd.org
   Cc: Julian H. Stacey; Tony; Steffen Daode Nurpmeso
   Subject: Re: FreeBSD's backwards webdesign / corporate identity
   
   Hi,
   
   On Friday 13 April 2012 18:44:07 Steffen Daode Nurpmeso wrote:
Julian H. Stacey wrote [2012-04-13 13:13+0200]:
 The 1000 year Reich lasted 6.
   
13.
Not for all, though.
   
   1945 - 1933 gives 12.
   
   Do I have to start a calculator now?
   
  
  Its 13 INCLUSIVE. You're calculating exclusive
 
 it also fits better to today's date.

Fits even better next Friday! ;-)



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: link_elf: symbol ata_controlcmd undefined

2012-04-13 Thread Polytropon

On Fri, 13 Apr 2012 23:11:22 +0200 (CEST), Marco Beishuizen wrote:
 Hi,
 
 I noticed some messages when booting:
 ...
 link_elf: symbol ata_controlcmd undefined
 KLD file atapicam.ko - could not finalize loading
 ...
 
 My /boot/loader.conf includes:
 ...
 atapicam_load=YES
 hw.ata.atapi_dma=1
 ...
 
 Entering kldload atapicam gives:
 kldload: can't load atapicam: No such file or directory
 
 Has anyone an idea how to load atapicam.ko?
 
 I'm running FreeBSD-9.0-STABLE.

In FreeBSD 9, loading atapicam should not be neccessary,
as it is now part of the GENERIC kernel and has merged
the ATA and SCSI functionality for disks and optical
devices. Try removing it from loader.conf and try again.
(Note possible device name changing ad - ada unless
you're using labels.)




-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: FreeBSD's backwards webdesign / corporate identity

2012-04-13 Thread Polytropon

On Sat, 14 Apr 2012 07:49:40 +0700, Erich Dollansky wrote:
 Hi,
 
 On Friday 13 April 2012 23:37:16 Polytropon wrote:
  On Fri, 13 Apr 2012 22:59:41 +0700, Erich Dollansky wrote:
   On Friday 13 April 2012 20:56:35 Sean Cavanaugh wrote:
   
 -Original Message-
 From: owner-freebsd-questi...@freebsd.org [mailto:owner-freebsd-
 questi...@freebsd.org] On Behalf Of Erich Dollansky
 Sent: Friday, April 13, 2012 9:12 AM
 To: freebsd-questions@freebsd.org
 Cc: Julian H. Stacey; Tony; Steffen Daode Nurpmeso
 Subject: Re: FreeBSD's backwards webdesign / corporate identity
 
 On Friday 13 April 2012 18:44:07 Steffen Daode Nurpmeso wrote:
  Julian H. Stacey wrote [2012-04-13 13:13+0200]:
   The 1000 year Reich lasted 6.
 
  13.
  Not for all, though.
 
 1945 - 1933 gives 12.
 
 Do I have to start a calculator now?
 

Its 13 INCLUSIVE. You're calculating exclusive
   
   it also fits better to today's date.
  
  Fits even better next Friday! ;-)
  
 oh, yeah, the big birthday bash. Is it organised via facebook?

Who with a sane mind would press his face into a book? :-)



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Sendmail recommended permissions for apache/php server

2012-04-12 Thread Polytropon

On Thu, 12 Apr 2012 08:17:33 +0100, Matthew Seaman wrote:
 On 12/04/2012 02:49, Polytropon wrote:
  On Wed, 11 Apr 2012 23:57:51 +, Ian Lord wrote:
   I then got a different error in /var/log/messages
   Apr 11 19:38:40 dev sendmail[41170]: NOQUEUE: SYSERR(www): can not write 
   to queue directory /var/spool/clientmqueue/ (RunAsGid=0, required=25): 
   Permission denied
 
   I found very old threads saying to change the group of apache
   to smmsp but I doubt it's a good idea.
 
  No, not change to, but you can _add_ apache (or whatever is
  originating the error) to the smmsp group. Add it to smmsp:*:25:
  in /etc/group.
 
 You should not be changing the ownership and permissions on any of the
 directories used by sendmail(8), or the group membership of any of the
 groups used by sendmail.  Not even if you think you know what you are
 doing.  This is extremely security sensitive, and getting it wrong means
 at minimum unprivileged users can forge e-mails untraceably[*].

You're right - as long as sendmail works properly (and is invoked
by whatever means sends e-mail out of apache / PHP), the present
group settings and permissions should be okay. Sendmail will
then properly run as the smmsp group member which will enable
it to properly access the queue directory.



 There is no reason for apache to have any sort of write permissions to
 /var/spool/clientmqueue -- that should only be accessible to sendmail,
 and sendmail is the only program that should ever use it.

I'm not aware of why a program should directly access the mail
queues, but maybe that's a special PHP feature. :-)




-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: install IDS from port , but can not find the Ports security/acid

2012-04-11 Thread Polytropon

On Wed, 11 Apr 2012 15:41:10 +0800, PstreeM China wrote:
 hi everyone :
 
 i want use the snort like IDS in my network ,  google the document from
 internet .
 there is ACID (analyst center) used provided web UI , can install from
 security/acid .
 
 but after i fetch port (portsnap fetch ; portsnap extract ) , i can not
 find the ports security/acid

There is an entry in /usr/ports/MOVED:

security/acid||2008-04-04|Has expired:
development has ceased, use security/base

From base's description:

BASE is the Basic Analysis and Security Engine. It is based on the code
from the ACID project. This application provides a PHP-based web front-end 
to query and analyze the alerts coming from a Snort IDS system.  

BASE is a web interface to perform analysis of intrusions that Snort has
detected on your network. It uses a user authentication and role-base
system, so that you as the security admin can decide what and how much
information each user can see. It also has a simple to use, web-based
setup program for people not comfortable with editing files directly. 

Maybe you can check this one in relation to your requirements?


-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: How to successfully enable HP LaserJet Professional m1212nf MFP,

2012-04-11 Thread Polytropon

On Wed, 11 Apr 2012 14:17:01 -0500, Edwin L. Culp W. wrote:
hpcups 3.12.2, requires proprietary plugin that seems to not be available
in the HP site. I have tried to get it using hplip-3.12.2 with no
success. I have tried with both cups and hplip and can't get it going.
Any suggestions appreciated. Maybe the official hplip-3.12.4 might work
but hasn't been updated yet.I tried to compile it but wasn't able to
adapt the patches.

I have checked the printer's specification, but I can't
find any mentioning about if it supports one of the
standard languages PS or PCL (as one would assume for
a product that HP markets as Pro(fessional)). However,
the documentation states that it accepts PDF - so maybe
you can try to feed a PDF file to the printer directly?
You can use nc (netcat) to do this, I assume you already
have the printer networked.

I'm not sure how the other functionality relates to the
network connection (or maybe it is only availabe for the
local USB connection?), check the documentation that came
with the printer to find out more.

For example, my Samsung color laser printer (MFC) has no
networking functionality, but is represented by /dev/ugen0
for the scanner part and /dev/u(n)lpt0 for the printer part.
Maybe something similar is possible with your printer?

I'm using that kind of setup with my HP Laserjet 4000 duplex,
a _real_ professional (office-class working horse) printer.
It's accessed per its IP and fed PS, which is the default
output format of any application that wants to print something.
The printer spooler is inside the printer and can be queried
via CUPS (and also by its command line tools).

P.D. Is there a better way to use hp equipment than cups?

Yes, base system's printer spooler (lpr) that simply hands
the print jobs to the printer and manages them remotely.
This assumes the printer has its internal print server
(which should be normal for anything professional). CUPS
can also deal with that if needed, as more and more applications
rely on its presence.

Finally I _assume_ the printer sadly is not that professional
and doesn't support a lot of standards, depending on what I
found on this page:

http://h10010.www1.hp.com/wwpc/us/en/sm/WF06a/18972-18972-3328064-12004-3328083-3965847.html?dnr=1

Good luck anyway! :-)

--
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Sendmail recommended permissions for apache/php server

2012-04-11 Thread Polytropon

On Wed, 11 Apr 2012 23:57:51 +, Ian Lord wrote:
 I then got a different error in /var/log/messages
 Apr 11 19:38:40 dev sendmail[41170]: NOQUEUE: SYSERR(www): can not write to 
 queue directory /var/spool/clientmqueue/ (RunAsGid=0, required=25): 
 Permission denied
 
 I found very old threads saying to change the group of apache
 to smmsp but I doubt it's a good idea.

No, not change to, but you can _add_ apache (or whatever is
originating the error) to the smmsp group. Add it to smmsp:*:25:
in /etc/group.

See the error message above:

can not write to queue directory /var/spool/clientmqueue/

Check:

% ls -ld /var/spool/clientmqueue
drwxrwx---  2 smmsp  smmsp  512 Apr 12 03:12 /var/spool/clientmqueue/
^^^
This directory can be read, written and entered/searched by
_members_ of the smmsp group.

Back to the error message:

(RunAsGid=0, required=25)

It is indicated that group #25 (smmsp) is the required GID, not 0.

And:

Permission denied

which is the logical conclusion.

Conclusion: You must make sure that whatever needs to access
this directory is in the smmsp group (25).



 Chmodding 777 the /var/spool/clientmqueue/ fixed the problem,
 I can now send emails, but I wonder if this is the way to fix
 the issue correctly.

You souldn't need to do that. Now this directory can be modified
by anyone, that's not good.



 Is that the official fix or did I missed some configuration
 somewhere ? Sending emails from php using mail or sendmail
 should be something working out of the box I guess, I doubt
 we're supposed to change permissions to make it work

Correct. In regards of _security_, it's required to _allow_ the
corresponding program / functionality / part of apache / mailer
or whatever the access to the mail queue. This is something that
is _not_ possible out of the box because there are many possi-
bilities and security considerations.



 Any help would be appreciated.

Try to add apache (or whatever part of it, or PHP subsystem called
by it that needs to access the mail queue) to the required group
to give it the proper permission to do so.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: FreeBSD's backwards webdesign / corporate identity

2012-04-10 Thread Polytropon

On Tue, 10 Apr 2012 15:29:42 +1000, Da Rock wrote:
 On 04/10/12 21:32, per...@pluto.rain.com wrote:
  Mark Felderf...@feld.me  wrote:
 
  Python on Planes is the future, mn.
  Shouldn't that be spelled plains, as in the places where the
  snake-containing grass grows?
 
 
 
  :-)
 
 Ha! One would think so, but with ruby on rails one would think that 
 python on plains wouldn't sound anywhere near as exciting or appear too 
 quick. That and a shaded reference to a certain similarly titled movie 
 with Samuel L Jackson- corny! :D

Should we modernize programming languages by putting
them on something? Like awk on a anchor, C on a
chimney or Java on Jambalaya? :-)



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: FreeBSD losing market share?

2012-04-09 Thread Polytropon

Tony, I'm always fascinated how people consider market share the
purpose of everyone and everything. FreeBSD is not a profit-oriented
company (it's not even a company in this regards), and you can
hardly _measure_ its market share. Hell, you can't even measure
its _usage share_! Unlike corporations with a certain income model
where unit sales can be counted, you cannot count them for FreeBSD
as anyone can download and install as many copies of it as he
likes. Due to the licensing model, derived works that are turned
into a closed-source project can even be attributed to a different
company (e. g. a FreeBSD-derived OS that is installed into an
embedded system acting as a firewall will sales_units++; for that
company, not for FreeBSD). You have _no_, I repeat NO means
to determine how many FreeBSD systems are currently up and running.
That would be usage share. Market share is a measuring model that
you can't even apply to FreeBSD in my opinion.



On Sun, 8 Apr 2012 15:22:47 +0200, Tony wrote:
 Imagine how FreeBSD's market share and popularity would skyrocket once
 regular people gets access to it.

FreeBSD has no market share, if you apply the term correctly,
as it is not part of the market.



 Low-cost hosting definitely is the way of
 the future.

I'm not sure it is. Even by the means of cloud computing prices
are still rising (due to energy costs increasing), and only efficiency
is a way to chance this trend. Sadly, requirements to not follow this
approach, which makes things becoming more expensive in the future.
Unlimited data is also a thing that, in my opinion, will disappear
in the future. Lean and fast applications will have a renaissance.



 Just look at how well low-cost
 airlineshttp://en.wikipedia.org/wiki/List_of_low-cost_airlinesare
 doing.

Are _currently_ doing, but they will sooner or later be out of fuel.
Fuel is becoming more expensive as the available amount is limited.
If you consider such things on the long run, you will surely have
to admit that a short-time strategy (being cheap right now) does
not pay.




-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: FreeBSD's backwards webdesign / corporate identity

2012-04-09 Thread Polytropon

Allow me a few additions:

On Mon, 9 Apr 2012 13:58:43 -0500 (CDT), Robert Bonomi wrote:
   2) Make sure *every* 'img ' tag has a meaningful alt=' field.  (if it
  is just a filler/spacer image, _still_ put an  'alt= ' field on it.
  This explicitly declares that the image has 'no content')

And also note the longdesc= parameter to img. It should be
used to give a meaningful (!) description of an image's
content.



   6) Try resizing the browser panel -- wide (side-to-side) and short (top-
  to-bottom), and narrow (side-to-side) and tall (top-to-bottom).
  Does it work _acceptably_ in a 640x480 window?  (why not?  There are 
  users out there with VGA-only displays.)

Also note that following this auto-resize approach enables
a web page to be used in all its beauty on mobile devices,
such as tablets and smartphones. If done right, no separate
version is needed there.



  *IDEALLY* the content should expand to fill whatever width there is
  available (I have a display 1920 pixels wide, the current webpage
  =refuses= to use more than about 1/2 of it.

Excellent idea for improving. Many web pages could learn from
that simple idea. The habit of being optimized for 1024x768
is very annoying, especially if you _have_ a large display
where you arrange non-fullscreen windows on. Pages that fail
to accomodate to the window's size are very annoying. In fact,
they are the majority.



  Get this right, and you show you understand that HTML is -not-
  a 'page layout' language, That one is merely providing 'hints'
  for the browser to 'do with as it sees fit'.  Web layout *is*
  a very different discipline from layout for the printed page.

Many web developers seem to be unable to see HTML as a markup
language that defines structure (indead of layout). If you
want to have pixel-precise 1:1 reproduction, use PDF.

With HTML5 (but also applies to HTML4), it's easy to use the
HTML tags to define what text _is_ instead of what it should
look like. A usable approach is to use CSS for styling, and
HTML for content-oriented structural description. That way,
from the HTML content one can determine what's a heading,
what's a paragraph, what's an address - instead of thinking
in terms like this is in bold text or this is underlined
text with 18px in 'Comic Sans'. The idea of semantic browsing
can be utilized.



 Once you have -that- 100% functional with no errors, -then-, and *only*
 *then* is is appropriate to add *optional* (for the user) 'enhancements' 
 in addition to the basic functionality.

I think there _is_ potential to improve the FreeBSD website,
in layout, content and structure. But it should be done
carefully so it doesn't break anytime soon (such as modern
web pages do when a new version of some arbitrary extension
is out or when the underlying implementation language breaks
things due to updates).



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: FreeBSD's backwards webdesign / corporate identity

2012-04-08 Thread Polytropon

On Sun, 8 Apr 2012 14:40:12 +0200, Tony wrote:
Hello!

As much as I love FreeBSD, I'm a bit alarmed by its webdesign / corporate
identity.

What's wrong with it? It's very accessible (especially for
blind users) and it presents the availabe information in a
structured way. Sure, it does not use many of today's modern
extensions to get simple things done, as it uses _simple_ things
to get them done (e. g. a href= for a link istead of Flash),
but that's not a problem in my opinion. The pages load fast,
they display well in all four major browsers (Firefox, Chrome,
Safari, Opera), and it even renders properly in browsers with
limited abilities (lynx, links, w3m, dillo).

Your next point: Corporate identity. I sadly fail to see where
FreeBSD can be seen as a corporation. It's rather a community,
having some core installations, but it's not a company that has
to maintain a specific design across all its products. However,
FreeBSD's projects are consistent regarding naming and logos.

Since FreeBSD is the world's best OS, I believe it should have a
design that reflects this.

In your opinion, what would you consider a good example to
imitiate, or at least to consider as a source of inspiration?
I don't say design couldn't be improved - but what are _your_
opinions in how it should be done? Can you be more specific?

A design that is so neutral and stripped of any
unnecessary details that the user's attention is directed straight on to
the content as opposed to how the content looks. A design that you can look
at over and over without getting annoyed.

Hmmm... I always thought exactly that is what the landing page
of the FreeBSD project already is. You're describing the status
quo.

The current design is an uneven mix of various styles, and seems more
forced than well thought out.

That may be due to the reason that those are different project
which are somewhat independent. The home page is a different
thing than the documentation (different maintainers, different
projects), and the wiki, as well as other sources, are associated
projects not governed by the the core team of the FreeBSD
project.

First you have the shiny Satanic 3D-lookalike
logo (yes, despite what y'all say, it's still Satanic) [...]

Despite you say the opposite, it's not. :-)

[...] that might look cool
the first few times one looks at it. Now though it's more like what the
hell *is* that thing anyway?

It's a logo, nothing more, nothing less. There are many logos
in the corporate world that raise the question of What _is_
this?!, and even in some cases, it cannot be answered because
it's _nothing_ (except a graphical exercise).

See the logo in that way, and see the mascot. Both of them are
not satanic. They have horns, the mascot has a tail. A bull
also has horns and a tail. Is it satanic therefore? Would you
refuse to eat a steak because it might be satanic as well?

(ref: Tres
Logoshttp://www.amazon.com/Tres-Logos-Robert-Klanten/dp/3899552679/ref=sr_1_1?ie=UTF8qid=1332777820sr=8-1
)

This reference cannot be reviewed for free, so I sadly have to
discard it. If you could be more specific on the FreeBSD case,
please _be_ more specific.

Then you have a surrounding layout trying to cater to that logo, but fails
miserably as it was made by programmers as opposed to people with an actual
education in design http://www.royalacademy.org.uk/.

No, the design caters the _content_. That is the purpose of
design aiming at the designated target audience.

There is no natural
flow http://en.wikipedia.org/wiki/Laminar_flow and the whole thing just
comes off as corny http://www.urbandictionary.com/define.php?term=corny -
and this makes us all look bad. I also hear
PostgreSQLhttp://www.postgresql.org/is planning to sue FreeBSD for
stealing its design.

Let's watch the result of the lawsuit then. :-)

I propose a new, supersimple look for FreeBSD based on
Helveticahttp://www.youtube.com/watch?v=wkoX0pEwSCw.

I don't think you can consider a font being the center of a
web design. What if that font isn't even installed? What if
a blind user accesses the page? Even though I like the
Helvetica font, I believe it's not enough, and even not
possible to design around a font.

Or am I misunderstanding your intention?

No devil logo, [...]

There already is no devil logo.

[...] no bells and whistles, [...]

Fully agreed, and already present.

Perfection is achieved, not when there's nothing left to add, but when
there's nothing left to take away.

Very good.

Re: define a default username for logging in

2012-04-07 Thread Polytropon

On Sat, 7 Apr 2012 12:21:57 +0430, takCoder wrote:
 Hi All :)
 
 i'm trying to find a way to enable a required feature : to set *default
 username *in my Freebsd 8.2 server..
 
 i mean, i wanna be able to login with just entering My Master Password(no
 usernames needed.. also prefer it to be per tty), which is *not related to
 my root account,  *but is the password of a user which i have defined as my
 default user..
 
 is it possible for, e.g. pam_login module (i couldn't find any manuals on
 such feature yet..), to have such a config or is there any other ways to
 set such default username for login?

It is, but I assume my answer will just be a half of the
whole story. The problem will be: no password. But maybe
you can find some inspiration and then extend the procedure
to fit your needs.



1. Modify /etc/gettytab as follows:

default:\
...

localautologin:\
:al=USERNAME:tc=Pc:

a|std.110|110-baud:\
...

where USERNAME is the name of the user you want to login as
(given by the al= parameter, and inheriting the tc= settings).
Make sure the user does exist in the system.



2. Modify /etc/ttys as follows:

ttyv0  /usr/libexec/getty localautologin  cons25  on  secure

and maybe change cons25 to cons25l1 (or any other value that might
be required).



As I said initially, this does _not_ prompt for a password!
Maybe /etc/passwd's shell field allows you to add the password
protection.

If you're logging in remotely, ssh USERNAME@yourserver.qw.er.tzu
will only prompt for a password. This idea offers an opportunity
to something overcomplicated:

Create a user for localautologin that is _not_ your default
user name. Make this user login automatically, and into his
~/.login, place the command ssh USERNAME@localhost so
right after performing the localautologin, ssh will attempt
to connect to localhost _as USERNAME_ and _prompt for_ the
password. Terrible, I know. :-)

To milden the pain of this approach, you could allow telnet
for localhost, i. e. from 127.0.0.1 to 127.0.0.1 _ONLY_ and
nothing more, and use telnet instead of ssh in the ~/.login
command.




-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Fast question abount EDITOR

2012-04-05 Thread Polytropon

On Thu, 05 Apr 2012 08:57:55 +0200, Andrea Venturoli wrote:
 Hello.
 
 This might be a stupid question... however...
 
 %setenv EDITOR emacs -nw
 setenv: Too many arguments.
 
 %setenv  EDITOR emacs -nw
 %crontab -e
 crontab: emacs -nw: No such file or directory
 crontab: emacs -nw exited with status 1
 
 
 Is there a way I can easily achieve the above?
 Do I really need a script which in turns call emacs -nw?

According to crontab's source, the editor is invoked
by an execlp() call, usr/src/usr.sbin/cron/crontab/crontab.c
in line 404:

execlp(editor, editor, Filename, (char *)NULL);

The synopsis of this function can be found in man 3 exec:

int execlp(const char *file, const char *arg, ... /*, (char *)0 */);

The manpage contains this information:

The initial argument for these functions is the pathname
of a file which is to be executed.

and:

The first argument, by convention, should point to
the file name associated with the file being executed.

as well as:

The functions execlp(), execvp(), and execvP() will
duplicate the actions of the shell in searching for
an executable file if the specified file name does
not contain a slash ``/'' character.  For execlp()
and execvp(), search path is the path specified in
the environment by ``PATH'' variable.  If this
variable is not specified, the default path is set
according to the _PATH_DEFPATH definition in paths.h,
which is set to ``/usr/bin:/bin''.

That means that $EDITOR has to contain the file name of the
editor (as its location will be determined automatically).
When options are added, this requirement isn't met anymore.
This seems to imply that you cannot use an alias - you'd
have to provide a script as you initially did assume. :-(



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: modem

2012-04-04 Thread Polytropon

On Wed, 4 Apr 2012 15:10:47 -0700 (PDT), tim smith wrote:
 
 Well, I checked the log for ppp, nothing I could see.
 There's not much as I still can't send the modem an AT, so...

Could you verify the presence of the cuau* file in /dev?
Maybe you can post the essential parts of your ppp.conf
as well as significant changes to your kernel (in case
you're not using the GENERIC kernel; check device uart),
and /boot/loader.conf?

And finally, can you show the _full_ error message that
you receive, maybe also informational parts of the ppp
log file?

What steps of diagnosis and testing could you already
accomplish (with which results)?


-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: modem

On Tue, 3 Apr 2012 09:39:38 +0700, Erich Dollansky wrote:
 Hi,
 
 On Tuesday 03 April 2012 06:49:55 tim smith wrote:
  
  My us robotics serial modem worked without issue on previous freebsd 
  versions. With 9, user ppp term, I get /dev/cuau0/ device failed to open
  
  Suggestions?
  
 what does 
 
 ls /dev
 
 say?
 
 Is the modem at least seen by FreeBSD?

Erm... seen by FreeBSD? I have never noticed something
like that. The OS sees the _serial port_ devices and assigns
a device in /dev, but the modem itself does not cause any further
action.

I've been using an Elsa Microlink serial modem in the past.
By the time the serial subsystem in FreeBSD changed, I didn't
use it anymore, but I assume /dev/cuau0, /dev/cuau0.init and
/dev/cuau0.lock should be present when the serial port is
configured correctly. See man 4 sio for details.

A ppp session protocol would also be interesting for
diagnosis purposes.


-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Printer recommendation please

 the firmware have the latest manufacturer
patches? Is there a password in the administration interface?
What traffic is running across the printer? While many sysadmins
(even in MICROS~1 environments) are aware of checking and
cleaning (and reinstalling) the Windows PCs frequently,
the things hidden in the printer are often left out. So
right after cleaning the PCs, the network could be re-initialized
by an attacker who lives inside the printer.

After all, I think social engineering based attacks will become
much more popular than addressing printers. I do _not_ say to
keep ignorant and carry on, but there are higher threats than
the PDF-capable laser printer in room 101. :-)



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Printer recommendation please

On Tue, 03 Apr 2012 11:22:24 -0700, per...@pluto.rain.com wrote:
Jerry je...@seibercom.net wrote:

Obviously you are not aware of the latest trend towards the
movement to standardize PDF as the standard print format. I would
recommend you start by reading the documentation located at:
http://www.linuxfoundation.org/collaborate/workgroups/openprinting
and continue on from there.

That page seems to be concerned with using PDF, rather than PS, as
a common intermediate print language in CUPS. I see nothing there
relevant to sending PDF directly to a printer.

See this page:

http://www.linuxfoundation.org/collaborate/workgroups/openprinting/pdfasstandardprintjobformat

It discusses (quite short, I admit) programs outputting PDF
instead of PS when generating printing data. Handing that
data over to the printer does not involve any conversion
or intermediate formats anymore.

The functionality of CUPS would then be reduced to what
the system's default printer spooler does (and did since
the 1970's): Read data from a program and send it to the
printer. Only the format of data has changed: pure text,
text with control characters, PS, PCL, PDF. It starts at
the application front.

While there might be some rational for your security concerns on
a business network in regards to wireless networks, they are not
really relevant on a home networks. The simple ease of use that a
wireless network gives a user on a home network far outweigh any
pseudo claims of espionage.

Following that line of reasoning to its logical conclusion would
lead one to believe that home networks have no need of any malware
protection, e.g. anti-virus. Any ISP which has had to deal with
incidents precipitated by customers' infected machines -- including
but likely not limited to DDoS and spambots -- would likely disagree.

Home networks and carelessly treated corporate networks
make the majority of what causes trouble on the Internet.
Don't notice == doesn't exist. :-)

I maintain that an attacker can more easily trick a less-than-
paranoid user into sending a malware print file to a PDF-accepting
printer than to a non-PDF-accepting printer, simply because PDF
is such a commonly used distribution format.

In regards of the web being a main source of attacks, few
lines of Javascript would allow to automatically access the
printer and send it some PDF data, drive-by attacks made
simple.

If someone prints a
malware PDF file that they have downloaded, and the process of
printing it does not require that it be transformed in any way (such
as conversion to PS) before being sent to the printer, their only
protection from disaster is whatever validation may be built into
the printer itself. (Keep in mind that what started the malware
discussion was Poly's link to a report stating that some printers
do not sufficiently validate an update firmware job.)

That's why I _hope_ printer manufacturers will take care
about that topic. As far as it's _possible_ to validate
PDF data that _might_ be a threat, it should be done, and
in worst case, malicious portions of the data should be
ignored.

Granted the identical exposure exists for a PS printer if the
downloaded malware file is identified as a PS file, however the
risk is much less in practice because distribution of PS files
is sufficiently uncommon that most unsophisticated users would
have no idea what to do with one if they were to come across it.

Furthermore, PS files would - on most cases - undergo another
conversion, for example to PCL. A PS interpreter would have to
be exploited to carry malicious code from PS to PCL (if the
PCL language allows the same kind of hostile manipulation as
the PS language would). In this case, FOSS is a good shield.
Code that gets many reviews by the _public_ is less prone to
contain backdoors (phrase incorrectly used) that would allow
such mis-interpretation.

Re: current pids per tty

On Tue, 3 Apr 2012 16:47:57 -0400, Matthew Story wrote:
 Across all TTYs, something like this would probably work:
 
 sudo fstat | awk '$5 ~ /^\/dev/  $8 ~ /tty/ { printf %s %s %s\n, $1,
 $8, $3; }' | sort -k1,2
 
 from there, if you think you need to trace the process trees down, you can
 use this list of pids and ps to do the rest ...

Or use pstree from ports.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Printer recommendation please

2012-04-02 Thread Polytropon

On Mon, 2 Apr 2012 07:33:03 -0400, Jerry wrote:
Obviously you are not aware of the latest trend towards the movement to
standardize PDF as the standard print format. I would recommend you
start by reading the documentation located at:
http://www.linuxfoundation.org/collaborate/workgroups/openprinting
and continue on from there.

Seconded, good introductional read.

Addition:

PDF as Standard Print Job Format
http://www.linuxfoundation.org/collaborate/workgroups/openprinting/pdfasstandardprintjobformat

While there might be some rational for your security concerns on a
business network in regards to wireless networks, they are not really
relevant on a home networks. The simple ease of use that a wireless
network gives a user on a home network far outweigh any pseudo claims of
espionage.

I think you're underestimating the threat coming from hijacked
home consumer networks. Of course, business networks are more
interesting, as they might contain data one could sell (personnel
data, inventions, business figures, pricing, internal products
calculations and so on), but home networks seem to be more
easily to crack. The typical point of attack is a Windows PC
in such a network, and the result is a machine controlled by
a criminal, acting as a spam server, as part of a botnet, as
a participant in illegal file sharing or as a storage point
for child pornography. The user itself often doesn't recognize
any of those activities.

In today's Internet, more than 90% of the traffic generated
in email is spam. What do you think they come from?

Now let's assume printers are easily exploitable because
manufacturers are careless when implementing the PDF printing
standard, or they leave extensions active that can be
abused. While average Windows users are more and more
aware of caring about viruses, trojans, malware and other
attacks for their _own_ security, such considerations
about a printer aren't wide spread. But it's only a
printer, it can't do anything!

What I want to say: Printers _are_ and _will be_ attack
vectors that need attention. If the manufacturers provide
a good basis, that would be great. For example, if a PDF
file contained malicious code, the printer accepts it,
prints it, but doesn't do anything more, it would be a
safe procedure. But as PDF is _known_ to be unsafe in
regards that it _can_ contain stuff to attack a computer,
the conclusion is that (depending on what manufacturers
actually implement) it might do so to a printer too.
The danger of PDF is comparable to the danger of Office
files (typically macros as hooks for malicious code).
Now add some auto-opening functionality to a MUA, and
you're done.

Summary: PDF as a printing standard is very welcome, as
long as it takes the chance to be a secure thing.

Furthermore, there are means of encrypting print data. I
leave the mastery of that matter up to the student.

That's interesting, I'll investigate on that further.

Re: Printer recommendation please

2012-04-01 Thread Polytropon

On Sat, 31 Mar 2012 21:20:41 +0200, Lino Miklav wrote:
 On 31.03.2012 00:16, Peter A. Giessel wrote:
  It doesn't surprise me that Gutenprint doesn't have a setting
  specifically for the 6500 because Xerox provides one:
 
 Uf, I have this idea to only use LPD and filters.

That should be no problem. If I read the specifications
for the Xerox Phaser 6280V DN correctly, it supports both
PS and PCL.

Here's an example for a PCL printer filter:

#!/bin/sh
printf \033k2G || exit 2
gs -q -dBATCH -dNOPAUSE -dPARANOIDSAFER -dSAFER -sPAPERSIZE=a4 
-r600x600 \
-sDEVICE=ljet4 -sOutputFile=- -  exit 0
exit 2

The ljet4 produces PCL, it can also be used to access
features like duplexer (add -dDuplex=true). It basically
does the same as the apsfilter filter, except that the
apsfilter one has support for pretty printing and
direct command line printing, so

% lpr foo.c

or

% lpr bar.png

can be issued directly, no need to create a PS stream
by another application.

You can easily add that filter to /etc/printcap's if= setting,
add rm= with the IP or hostname of the printer, prepare the
spool and it should work.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Printer recommendation please

2012-04-01 Thread Polytropon

On Sat, 31 Mar 2012 14:01:43 -0700, per...@pluto.rain.com wrote:
I personally don't trust wireless, because it's well nigh impossible
to truly secure it.

In that case, one should also pay attention to secure the
printer. Wait - secure the printer? What am I talking about?

Firmware attacks!

Yes - malware has already reached printers. As they contain
all typical parts of a computer and are equipped with net-
working capabilities, they can cause trouble in networks
the same way as what hujacked Windows PCs typically do.
They can be turned into networked allies, carrying out
the attackers orders within networks.

Those who are interested may find some information here:

Exclusive: Millions of printers open to devastating hack attack, researchers say
http://redtape.msnbc.msn.com/_news/2011/11/29/9076395-exclusive-millions-of-printers-open-to-devastating-hack-attack-researchers-say

ShmooCon 2011: Printers Gone Wild!
http://www.youtube.com/watch?v=GZgLX60U3sY#t=3m40s

ShmooCon 2011: Printer to PWND: Leveraging Multifunction Printers During
http://www.youtube.com/watch?v=MPhisPLwm2A

Printer malware: print a malicious document, expose your whole LAN
http://boingboing.net/2011/12/30/printer-malware-print-a-malic.html

Print Me If You Dare
Firmware Modification Attacks and the Rise of Printer Malware
http://events.ccc.de/congress/2011/Fahrplan/events/4780.en.html

HP firmware to 'mitigate' LaserJet vulnerability
http://news.cnet.com/8301-1009_3-57347817-83/hp-firmware-to-mitigate-laserjet-vulnerability/

It seems that printers can be infected via specific network
traffic or closed-source malicious drivers (that nobody
can examine content-wise) that will find their way to the
device. Depending on your local legislation, that can develop
into dangerous (and expensive) directions...

2. Standard language.
Postscript and PCL. Make sure the printer understands at least
one of them.

or, alternatively, PDF (which some of the newer printers are reputed
to take directly, rather than requiring the host to convert it to PS
or PCL).

Jerry mentioned this, and I think it's a feature worth demanding
when buying a new printer. Still if PDF input is not possible,
PCL or PS should be looked for. All those considerations make
sure you can use the printer with _any_ OS you like, and due
to this fact it will be usable even after the target OS will
be out of support (and follow-up drivers won't be provided).

Re: shutdown -p doesn't power-off USB

2012-04-01 Thread Polytropon

On Sat, 31 Mar 2012 17:38:11 +0200, Jens Schweikhardt wrote:
 I'm running 9-STABLE/amd64 and for a few months now, whenever I shut
 down with shutdown -p now, the USB devices still have power. This is
 most visible on the USB keyboard, where *all* LEDs are turned on and
 stay on.

That's not a bug, it's a feature. :-) Many mainboards allow
switching-on per keyboard. There's typically a toggle in
the board's CMOS setup. Maybe there's also an option to
turn USB power completely off. However, USB powered on
seems to be the default as soon as the machine's power
supply is on line.



 The MB is an ASUS P5Q3 Deluxe.

Also check its documentation, maybe USB power is mentioned?



 Any help appreciated in telling me how to turn off USB power with shutdown.

I don't think this is any option in the OS. You should check
this per hardware.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Printer recommendation please

2012-03-31 Thread Polytropon

On Sat, 31 Mar 2012 21:17:52 +1000, Da Rock wrote:
 Watch the older type fusers though- they can develop 'flat spots' on the 
 rollers. The newer printers use a ceramic type fuser which has fast 
 warm-up and no flat spot troubles.

But it's still possible to get replacement parts for older
office printers. I said _office_ printers, even used ones
that you can pick up for few dollars or a bottle of beer.
Spare parts aren't expensive, and in many cases, you can
install them yourself. The funny thing: Even for 10 years
old printers (and even older ones), they are available.

Try _that_ with a home consumer inkpee printer! :-)



 Also keep the dust low on _any_ printer and it will last longer and 
 perform better. Dusty paper can cause major issues (both printing and 
 mechanical) as well.

Sometimes rubber parts tend to harden. There are a few
tricks to make them soft again, but the typical solution
is to replace them for few dollars. Note that this isn't
something you'll notice in 2 - 5 years of use. You often
need 10 or more years to find fail and trouble in a good
printer. Good printer == office printer, as I said befire. :-)

-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Printer recommendation please

2012-03-30 Thread Polytropon

On Fri, 30 Mar 2012 17:38:36 +0200, Karel Miklav wrote:
 Could you please recommend me a home printer that works nicely with FreeBSD?
 
 HP inkjets aren't that bad, FreeBSD drivers are allright, but I'd like 
 to shift towards some kind of PostScript laser. Xerox Phaser 6500 looks 
 nice, but I can not economically justify my appetite. Is there a cheaper 
 alternative or maybe PostScript printers aren't that good idea anyway, heh?

Allow me to mention some things that are worth investing in.

1. Network connection.
Don't bother with USB stuff. Buy a printer that offers Ethernet
and maybe also WLAN, this will save you many trouble, and you
are free to put the printer wherever you want.

2. Standard language.
Postscript and PCL. Make sure the printer understands at least
one of them. PCL is very common among HP printers. Regarding
drivers - you don't need them. PS is the default output format
for printing from every application. Printer filter collections
such as apsfilter or CUPS tend to support non-PS printers very
well, and it's quite easy to write your own printer filter (may
even be a one-liner) using ghostscript. There's nothing wrong
with PS because (as I said) you don't need any drivers, but the
data transfer may need some time, and the processing speed
depends on how fast and how good (!) the PS interpreter in the
printer is. In my experience (with the printers I'm going to
mention at the end of this message) PCL is faster.

3. Laser printer.
Don't believe that inkpee printers are genereally cheaper. They
are not. The only excuse for using them is that you need photo
quality color prints (requiring the proper paper, too).

4. Additional functionalities.
Before buying something, ask yourself what you need. Does it
need to have a scanner? Does the scanner part support FreeBSD?
Is there a way to scan to local storage (e. g. USB stick)
in the printer? Does it need a sheet feeder for scan input?
Does it need to scan photo positive/negative films? Does it
need to fax?

I have had good luck with my army of laser printers here.
HP Laserjet II, 4, 4000 duplex, as well as a Samsung color
laserprinter CLX-2160. All this stuff works out of the box.
I don't have any need for inkpee. Photos can be printed at
much better quality at my local drugstore, if I need that.
The printer filters are gs one-liners I wrote myself, because
I speak PCL to the laser printer, and some splix gibberish
using foo2qpdl to the (sadly USB connected) color printer.




-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Printer recommendation please

2012-03-30 Thread Polytropon

On Sat, 31 Mar 2012 00:12:07 -0400, Matt Emmerton wrote:
 Toner really doesn't go bad, and good laser printers are built to last.  My
 first laser printer was an HP LaserJet 5P that my local bank branch was
 throwing away in 2003. It ran on its existing toner cartridge for 5 or 6
 years under light use - maybe 500 pages per year.

Ha, that's nothing!

I _still_ have a fully functional HP Laserjet 4. I got it
in a heavily used state in 1996, and I've never treated it
in a polite way: always quite heavy use. The printer is
still working today, after more than 15 years. It has
been on pause for some years, and right after plugging
it in again, it produced regular quality results. Just
try _that_ with typical home consumer inkpee stuff. :-)

I can't tell you how many pages the printer has done in
its life. The page counter must have encountered an overrun
and now says some 4 digit number which doesn't increase
anymore. So now I can sell it as only few pages printed,
like new... :-)

If durability is interesting, buying a laser printer will
be the right choice. Today's inkpee printers seem to be
the same price as a full ink cartridge refill (or even
lower), creating cheaper devices on one hand (good), but
creating more electronic waste at the other hand (bad).
So if you want to reduce garbage, get a printer that will
serve you for a long time.

When I was at university, there was a student, a rich one
as one could assume: When he had emptied a printer, he
bought a new one, dropping the old one into the garbage
can. He even bought a new printer when he failed to plug
in the one he just bought, and he also bought a new one
when he didn't get the drivers installed of another new
printer. He threw away two (maybe more?) fully functional
printers.

You see, money can compensate stupidity. His educational
result? He got a degree in computer science. :-)



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Vivaldi Tablet

2012-03-29 Thread Polytropon

On Wed, 28 Mar 2012 16:24:51 -0700, Gary Kline wrote:
   i dont have a clue what a chording keybd is;

This kind of keyboard uses key combination of its FEWER keys
to generate characters (or even syllables or words). The
name chorded is used synonymously with instruments like
the guitar where you use one hand to hold down certain
strings in a defined manner, and then it plays a chord
like A major or D minor.

There's an initial article about it on WP:

http://en.wikipedia.org/wiki/Chorded_keyboard

This kind of keyboard is typically used by court recorders
in the US. They are trained to record whole conversations
in real time directly onto paper. By bressing three, four
or more keys at a time, a specific output is generated by
the device. It's often called stenotype, because it's like
typing in stenography, emphasizing that's a phonetic code
in the foreground.

http://en.wikipedia.org/wiki/Stenotype

http://upload.wikimedia.org/wikipedia/commons/4/40/Stenkeys.gif

http://upload.wikimedia.org/wikipedia/en/c/cf/Steno-example.gif

Also typewriters for blind persons use this approach. The
model Erika Picht Portable (paper format DIN A5 I think)
is  still well known to me. There's also a regular (DIN A4)
model, produced by Schreibmaschinenwerke Dresden (type-
writer works Dresden), part of the combinate robotron.
Those machines are _stiill_ produced in Dresden.

http://www.aph.org/museum/images/braillewriters/30.jpg

http://petitmuseedubraille.free.fr/_machines-braille/images/_m15a.jpg

http://www.gfai-sachsen.de/images/Erika-Picht_MultiTech-E511_800.jpg

Input devices with comparable key layouts are also available
for the PC, but instead of stenotype, they generate regular
characters.



   i v much like this vivaldi 7 tablet, just as-is.  i wonder
   if a future 7inch model could have more memory Along with a
   slide-in kybd.  slide out and work: edit, use ffox,
   konsole or xterms, then slide back in place. this tablet
   could replace the ipad, nook, asus.  

Interesting thought. Maybe it wouldn't target home commodity
users in the first place, but a sliding keyboard could be
a benefit for professional users who want to do more than
just watching movies on such a thing. It would also help
to bring the concept of separating input and output to the
device in a physical manner (because it might be useful
in certain conditions when your fingers aren't located
at places where you are supposed to read something), and
STILL keeping the regular touch interface (no real separation)
available, intact and unbroken.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Vivaldi Tablet

2012-03-27 Thread Polytropon

On Tue, 27 Mar 2012 17:21:45 -0700, Gary Kline wrote:
   how about the eee-701s?  they are no mo' but used to have a
   70% of full size keyboard.  my eee-900A had All the std
   keys.  do we really need the F[n] keys?
 
 
   anyway, if not a  tiny kybd, maybe a small one.  

Maybe I can suggest the Happy Hacking keyboard here?
It's small in size and popular amnong hackers. It does
not have PF keys (those can be emulated by Fn+number,
comparable to Alt+number on early 3270's). Its dimensions
are about 11 x 4.5 x 1.5 at less than 1.5Lb weight.

However, this is an external (USB2) keyboard.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Off-Topic: Computing for the Blind

2012-03-27 Thread Polytropon

On Tue, 27 Mar 2012 08:21:04 -0500, Martin McCormick wrote:
   By the way, math done by any method other than Braille
 is darn next to useless. Equations in Braille can be formatted
 very much like they are in print and there is a whole Braille
 system for reading and writing math.

Interesting, I didn't know that. However, LaTeX allows
writing (and typesetting) math on a pure text basis
which may be interesting to authors who are unable to
access a GUI-driven formula editor. Of course there is
another learning courve here. But nothing does prohibit
a blind scientist to write his stuff himself, read it
himself; things as $\bar{x}=\frac{\sum_{i=1}^{n}({x_i})}{n}$
can be quite easily be used if you have learned few
relatively simple things: typing on the keyboard,
using a powerful editor, the LaTeX language, and
maybe Braille. This way, an author can concentrate
on content, while the tools step into the background
and let him just do his stuff.



 After all, it's unix which means one can expect
 certain behaviors regarding standard devices.

As long as the devices play nice... :-)


-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Vivaldi Tablet

2012-03-27 Thread Polytropon

On Tue, 27 Mar 2012 19:48:34 -0600, Chad Perrin wrote:
 On Wed, Mar 28, 2012 at 02:37:49AM +0200, Polytropon wrote:
  On Tue, 27 Mar 2012 17:21:45 -0700, Gary Kline wrote:
 how about the eee-701s?  they are no mo' but used to have a
 70% of full size keyboard.  my eee-900A had All the std
 keys.  do we really need the F[n] keys?
   
   
 anyway, if not a  tiny kybd, maybe a small one.  
  
  Maybe I can suggest the Happy Hacking keyboard here?
  It's small in size and popular amnong hackers. It does
  not have PF keys (those can be emulated by Fn+number,
  comparable to Alt+number on early 3270's). Its dimensions
  are about 11 x 4.5 x 1.5 at less than 1.5Lb weight.
  
  However, this is an external (USB2) keyboard.
 
 I was thinking of mentioning the Happy Hacking keyboard, but I see you
 beat me to it.  I have not used one for more than a few minutes once,
 though.  Does the Fn+number work with Ctrl+Alt+Fnumber combination to
 move around between TTY consoles?

As far as I remember, it does. I don't have a HHK here to check.
From what I know, the keyboard generates the proper codes
internally, so Fn+number is equivalent to PF number in
any regards, and therefore any combination with Ctrl and/or
Alt should also work as expected. To the computer, it should
be no difference from a real keyboard.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Off-Topic: Computing for the Blind

2012-03-26 Thread Polytropon

On Mon, 26 Mar 2012 14:21:08 -0500, Martin McCormick wrote:
   The easiest and most economical interface for computer
 users who are blind is spoken speach.

That's correct. However, unlike a Braille readout which
gives tactile information (through the reader's hands),
synthetic voice cannot easily accomodate to the reader's
habits and reading speed. Scanning text is not possible
as the generated voiced text is played in linear time,
which means you cannot easily skip forward and backward,
re-read a certain passage, and you basically do not come
down to the letter level, you only have a word level.
While this has benefits in unconcentrated reading (e. g.
reading an article or literature, it can be problematic
with scientific or technical text where a (healthy) reader
would let his eyes jump within the text stream.



   One can learn to type and touch-typing was tought in
 schools for the blind for scores of years before computers ever
 even came on the scene.

I also learned typewriting (mandatory!) in school, and
believe it or not, it comes handy every time I have to
deal with a computer. :-)



 We pounded on typewriters and our
 poor suffering typing teachers were the feedback mechanisms that
 told us how we were doing. So, a person who is blind needs to
 know how to type.

A good keyboard can help here. Keep in mind that a keyboard,
being a means of input, provides tactile feedback as output.
So without any visual confirmation you can detect when you
made a typing error, activating a motor program to correct
it on the fly.

At this point, I typically recommend using an IBM Model M
keyboard. But the Sun USB Type 7 is also good, as it provides
programmable keys for volume control, application interaction
and Braille readout control. (I use those keys primarily for
dealing with the window manager - no need to use the eyes!)



   None of these screen readers are perfect, but most
 computer users who are blind end up being reasonably happy with
 one of them.

Especially in combination with web browsers, they are prone
to fail. Where there's no text (as content) in a web page,
there's nothing to read to the user. The use of the HTML
tags alt= and longdesc= is a long forgotten art, and when
Flash enters the scene to replace few lines of HTML (as
for links or simple text), there's no easy way to determine
_what_ currently is on the screen.



   There are also Braille displays which some people use
 but they are extremely costly.

Sadly, that is correct. In my opinion this is because they
are a niche market. When purchasing one, you have to pay
attention to if it can capture normal text screen content.
How is it attached to the computer? Does it require proprietary
drivers? How long can it be used before an OS revision breaks
the drivers?

Those Braille readouts can be placed infront of the keyboard,
the primary means of input. Reading and writing isn't far
away from each other (finger travelling distance).

Classic Braille readouts didn't seem to require any driver.
I've seen such devices in the past. A slider on the side
simply defined the row of text which was then displayed on
the readout - one out of 25. I think it was plugged into
the VGA chain (PC - readout - screen), but I'm not that
familiar with this technology; I've seen it on a DOS PC.
However, as FreeBSD's default screen mode is 80x25 text
mode, it should be possible to use such a device. Maybe
it's possible to get a used one for cheap...



   I mentioned the speech recognition systems. Many of
 those actually present problems for those who are blind because
 you need to train them on your speech and the feedback is
 graphical so a good old keyboard is still the best input device.

Speech recognition requires training. Both the user and the
system have to learn from each other. But you have a learning
curve everywhere, be it typing, talking, or reading from a
Braille output.




-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Off-Topic: Computing for the Blind

2012-03-25 Thread Polytropon

further investigations.




-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: about change file mode

2012-03-22 Thread Polytropon

On Thu, 22 Mar 2012 11:44:18 +0100, Xavier FreeBSD questions wrote:
 Hi tot all,
 
 Why don't change the files mode ?
 
 casa# mount -t msdosfs /dev/da0s1 /mnt/JetFlash\ Transcend\ 1GB/

The answer is right in your first command: You're using
a MSDOS file system. That particular file system doesn't
know about rwx attributes. That's why files have +x by
default. You can not remove the x attribute from them
because they actually don't have one.

However, you can mask that false-positive attribute
by using the -m option. See man mount_msdosfs for
details. You can also use it in your /etc/fstab's
options field to make it a default.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Convert mp3 to audio CD

On Tue, 20 Mar 2012 23:10:16 -0400, Steve Bertrand wrote:
 I know this is a backwards request, as I haven't had to go from mp3 to 
 audio CD format in at least 10 years, but I do now.
 
 What is available to do so?

This script, ugly as hell, but works. :-)

You need to install madplay or mpg123, as well as sox.
If you also want to convert Ogg/Vorbis files, you need
ogg123 (vorbis-tools). And of course you need a cd
burning program such as cdrdao or cdrecord. See the
examples for calling them at the end of the script.







#!/bin/sh

echo === MP3 + OGG/Vorbis - Audio CD Converter 
===
echo -n MP3 decoders found:
DECODER=0
which mpg123  /dev/null 21
if [ $? -eq 0 ]; then
DECODER=2
echo -n  mpg123
fi
which madplay  /dev/null 21
if [ $? -eq 0 ]; then
DECODER=1
echo -n  madplay
fi
if [ $DECODER -eq 0 ]; then
echo none, aborting.
exit 1
fi
echo -n , using 
if [ $DECODER -eq 1 ]; then
echo madplay.
else
echo -n mpg123
which madplay  /dev/null 21
if [ $? -eq 0 ]; then
echo  and sox.
else
echo , but sox is not available, exiting.
exit 1
fi
fi
echo -n OGG/Vorbis decoders found:
which ogg123  /dev/null 21
if [ $? -eq 0 ]; then
echo  ogg123.
else
echo none, aborting.
exit 1
fi

TOCFILE=audiocd.toc
echo CD_DA  $TOCFILE

FILELIST=`ls *.[mo][pg][3g]`
if [ $FILELIST =  ]; then
echo No files (*.mp3 and/or *.ogg) found, aborting.
exit 1
fi

for FILE in *.[mo][pg][3g]; do
echo -n $FILE - 
echo $FILE | grep/dev/null 21
if [ $? -eq 0 ]; then
echo filename contains spaces, cannot proceed.
exit 1
fi
TYPE=
echo $FILE | grep .mp3  /dev/null 21
if [ $? -eq 0 ]; then
TYPE=mp3
fi
echo $FILE | grep .ogg  /dev/null 21
if [ $? -eq 0 ]; then
TYPE=ogg
fi
if [ $TYPE = ogg ]; then
ogg123 -q -d raw -f $FILE.raw $FILE
sox -r 44100 -c 2 -w -s $FILE.raw $FILE.cdr
rm $FILE.raw
elif [ $TYPE = mp3 ]; then
case $DECODER in
1)
madplay -q -o cdda:$FILE.cdr $FILE
;;
2)
mpg123 -sq $FILE  $FILE.raw
sox -r 44100 -c 2 -w -s $FILE.raw $FILE.cdr
rm $FILE.raw
;;
esac
elif [ $TYPE =  ]; then
echo file type unknown, aborting.
exit 1
fi
echo $FILE.cdr
echo TRACK AUDIO COPY AUDIOFILE \$FILE.cdr\ 0  $TOCFILE
done
echo Your recording command could be:
echo  # cdrecord -eject -v -dao -audio *.cdr
echo  # cdrdao write --eject audiocd.toc
echo After finished, don't forget to delete converter garbage:
echo  # rm *.cdr audiocd.toc
echo 
==
exit 0




-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: dbus, epiphany, rekonq

On Wed, 21 Mar 2012 09:29:09 +, Anton Shterenlikht wrote:
 On Wed, Mar 21, 2012 at 04:12:50AM +0100, Bernt Hansson wrote:
  You do have this in /etc/rc.conf
  
  dbus_enable=YES
 
 I didn't think it was necessary, as firefox3
 launches dbus-daemon on startup. But I'll give
 it a go.

I think those are mandatory when running X + { KDE | Gnome }.
Maybe it's even required for running KDE or Gnome applications
without the whole desktop environment? Maybe the dependencies
are that deep that they affect the libraries used...

Always check for

hald_enable=YES
dbus_enable=YES

in your /etc/rc.conf.

See:
http://www.freebsd.org/gnome/docs/halfaq.html


-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Convert mp3 to audio CD

On Wed, 21 Mar 2012 07:07:40 -0400, Rod Person wrote:
 On Tue, 20 Mar 2012 23:10:16 -0400
 Steve Bertrand steve.bertr...@gmail.com wrote:
 
  I know this is a backwards request, as I haven't had to go from mp3
  to audio CD format in at least 10 years, but I do now.
  
  What is available to do so?
  
 
 Basically the same as other, but just using lame to convert a directory
 full of mp3s
 
 
 #!/bin/sh
 
 for a in *
 do
   OUTF=`echo $a | sed s/\.mp3/.wav/g`
   lame --decode -q 0 $a $OUTF
 done

Just note that those *.wav files will have to be in the
correct format (44.1 kHz two-channel 16 bit) and maybe
require byte order reversal as well as stripping the
WAV headers to record them as a music CD. It seems that
some recording programs already contain this step. Refer
to audio CD specifications for why pure WAV files don't
make an audio CD.




-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Convert mp3 to audio CD

On Wed, 21 Mar 2012 07:25:11 -0400, Rod Person wrote:
 On Wed, 21 Mar 2012 12:16:24 +0100
 Polytropon free...@edvax.de wrote:
 
  On Wed, 21 Mar 2012 07:07:40 -0400, Rod Person wrote:
   On Tue, 20 Mar 2012 23:10:16 -0400
   Steve Bertrand steve.bertr...@gmail.com wrote:
   
I know this is a backwards request, as I haven't had to go from
mp3 to audio CD format in at least 10 years, but I do now.

What is available to do so?

   
   Basically the same as other, but just using lame to convert a
   directory full of mp3s
   
   
   #!/bin/sh
   
   for a in *
   do
 OUTF=`echo $a | sed s/\.mp3/.wav/g`
 lame --decode -q 0 $a $OUTF
   done
  
  Just note that those *.wav files will have to be in the
  correct format (44.1 kHz two-channel 16 bit) and maybe
  require byte order reversal as well as stripping the
  WAV headers to record them as a music CD. It seems that
  some recording programs already contain this step. Refer
  to audio CD specifications for why pure WAV files don't
  make an audio CD.
  
 
 I've used this for years and never had an issues, but to accomplish
 removing the header you would use the -t option along with --decode for
 lame and -x does a bit swap, but not sure if that is the same byte order
 reversal.

Yes, I think that's the correct approach. My old script,
written when I was new to FreeBSD (at v4.0), uses the
unelegant sox -r 44100 -c 2 -w -s $FILE.raw $FILE.cdr
to convert to the proper CD audio format which could
also be used by burncd (deprecated?), but also by cdrdao
and cdrecord with no additional conversion parameters.

Thanks for the hint about lame --decode!



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Dualboot with Windows 7

On Sun, 18 Mar 2012 23:05:58 +0100, David Demelier wrote:
 Hello,
 
 I try to create a dualboot with Windows 7, I set up partitions like that :
 
 ada0s1 - NTFS (windows recovery)
 ada0s2 - NTFS (windows main partition)
 ada0s3 - BSD
   ada0s3a - freebsd-swap (3G)
   ada0s3b - freebsd-ufs / (remaining space from drive)

Erm... according to traditional partitioning, isn't
the 'a' partition reserved for booting, 'b' for swap?
I see you have installed everything into one / partition
which technically is no problem and should work, but
it's not on the boot partition.



 And then I let the installer complete the step, because FreeBSD didn't 
 let you (since 9.0) choose between the boot manager nothing was 
 installed and the boot directly goes to Windows 7.

You need to install all the required stages for booting.
If I understand the process correctly, the slice 's3' needs
code to branch to the boot partition (which is supposed
to be the 'a' partition), and the boot selector needs to
be accessed from the beginning of the disk - you said
you're using EasyBCD for this which is okay.



 I installed EasyBCD to add a new entry to FreeBSD on the third 
 partition, but when I choose the FreeBSD entry nothing happens, only the 
 _ character blinking.

I assume missing boot characteristics as described above.
Please review your installation process and maybe re-do it.
In worst case, drop to command line for using the traditional
toolset to apply the proper slicing and partitioning.
According to man fdisk and man bsdlabel, you should be
able to write the required boot characteristics to allow
the correct boot process.



 Thus I tried bsdlabel -B ada0s3 from the FreeBSD iso shell but it didn't 
 solve. What can I do to boot FreeBSD now?

As this part is done, I suppose incorrect partitioning.

2.6.5 Creating Partitions Using Disklabel
http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/install-steps.html

Refer to table 2-2: Partition Layout for First Disk.

Boot manager and MBR handling are also covered in this chapter.



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Can't install WindowMaker

On Mon, 19 Mar 2012 07:51:04 +0100, Sabine Baer wrote:
 On Sat, Mar 17, 2012 at 07:07:33PM +0100, lokada...@gmx.de wrote:
 
 [...]
  I see that you have a german mailadress, so i will write in germany.
 
 Dann kann ich ja auch mal auf Deutsch antworten, ich hoffe, man sieht
 uns das nach.
 
  Guck mal nach, ob du unter /etc/ eine Datei mit dem Namen
  portsnap.conf hast.
 
 Ja, habe ich, habe ich aber bisher noch nie benutzt, ich nehme immer
 cvsup -g ports-supfile (mit host=cvsup.freebsd.org). Das dauert zwar
 ewig, aber ist doch gleichermassen aktuell, oder? 

Nein, es ist oftmals noch aktueller (more current). :-)

Explaination: The portsnap provides a snapshot of the ports
tree. It's _not_ updated as fast as what you can get via
differentials using csup (which can change a file within
half an hour). When many file changes are scheduled, portsnap
is faster, and it's especially useful to download a complete
ports tree. For being bleeding-edge current, csup is the
better method, especially if you update your ports tree
regularly.


  Wenn du eine Flatrate hast, kannst du vielleicht auch mal dein
  ganzes System auf 9.0 aktualisieren (freebsd-update ist für die
  Standardinstallation ohne Kompilierung ganz gut).  
 
 Ich habe bisher davon abgesehen, 7.n zu verlassen, weil ich aus dem
 Augenwinkel gesehen habe, dass bei Versionen 7.n die Einstellungen
 fuer die seriellen Schnittstellen veraendert werden muessen, und da
 hatte ich keine Lust zu. Dass die funktionieren, ist fuer mich aber
 absolut unabdingbar, ich habe naemlich ein echtes Terminal daran
 haengen

You should note that 7.4 is a legacy release which is rather
old, but I can understand the urge _not_ to fiddle with things
that just work (and especially if it's something _that_ special
like a _real_ serial terminal).



  Einige Probleme treten auf, wenn die Libarys auf dem System zu alt
  sind, aber in den Ports aktuelle benötigt werden.
 
 Vielleicht ist das bei windowmaker tatsaechlich so. Ich habe noch mal
 alles 'runtergeschmissen, ein make clean in ?usr/ports gemacht, alle
 distfiles 'rausgeschmissen und /usr/local/* komplett geloescht. Und
 auf einmal konnte ich Ports installieren, die zuvor alle Fehler
 erbracht haben - ImageMagick, dia, gimp, /lang/lua (fuer nmap).

Ist manchmal echt das beste. Hier zeigt sich mal wieder, daß
die Trennung von OS und Programmen sehr sinnvoll sein kann!



 Ich habe es ohne 'sophisticated tools', nur mit cd
 /usr/ports/*/$ANWENDUNG, make config und make install clean gemacht,
 es hat lange gedauert, aber war erfolgreich.

See man 7 ports, the target make config-recursive to get
rid of build interruptions due to forced interactivity.



 Nur der windowmaker weigert sich nach wie vor zu kompilieren. Da ich
 aller 'rausgeschmissen habe, weiss ich auch nicht mehr, ob ich zuvor
 eine aeltere Version installiert hatte.

Im Zweifelsfall installier Dir mal portdowngrade und zieh
eine ältere Version von WindowMaker, vielleicht compiliert
die ja?

Ich halte es in solchen Fällen oft für sinnvoll, nach der
Komplettreinigung und dem mtree-Lauf mit _dem_ Port zu
beginnen, den man eigentlich haben will. Alle Dependencies,
die der braucht, sollten automatisch gezogen werden, so hat
man eine minimale (kontrollierte) Umgebung, in der möglichst
wenig Stolpersteine liegen sollten.



 Ich bedanke mich jedenfalls bei Dir und 'Polýtropon' fuer die Hilfe,
 sie hat mir wirklich geholfen, wenn ich auch immer noch keinen
 windowmaker wieder habe.

On my voluminous voyage through window manager and desktop
environemnts, I've always come back to WindowMaker. As I
said, installing it on v8 hasn't been a problem, but when
I did use it on v7 (with the sources being current at _that_
time) it also worked. Maybe the current sources are really
too new, and a portdowngrade could work?

At least the port is not maked _not_ to build on v7, so it
basically _should_ work.

So, nun haben wir hier schönes Sprech-Mischmasch, mashed
language so to say... :-)



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: Dualboot with Windows 7

On Mon, 19 Mar 2012 08:29:22 +0100, David Demelier wrote:
 On 19/03/2012 07:28, Polytropon wrote:
  On Sun, 18 Mar 2012 23:05:58 +0100, David Demelier wrote:
  Hello,
 
  I try to create a dualboot with Windows 7, I set up partitions like that :
 
  ada0s1 -  NTFS (windows recovery)
  ada0s2 -  NTFS (windows main partition)
  ada0s3 -  BSD
 ada0s3a -  freebsd-swap (3G)
 ada0s3b -  freebsd-ufs / (remaining space from drive)
 
  Erm... according to traditional partitioning, isn't
  the 'a' partition reserved for booting, 'b' for swap?
  I see you have installed everything into one / partition
  which technically is no problem and should work, but
  it's not on the boot partition.
 
 
 
 You're right, but I made a mistake while writing, my a partition is / 
 and b is swap.

Okay.



  And then I let the installer complete the step, because FreeBSD didn't
  let you (since 9.0) choose between the boot manager nothing was
  installed and the boot directly goes to Windows 7.
 
  You need to install all the required stages for booting.
  If I understand the process correctly, the slice 's3' needs
  code to branch to the boot partition (which is supposed
  to be the 'a' partition), and the boot selector needs to
  be accessed from the beginning of the disk - you said
  you're using EasyBCD for this which is okay.
 
 
 
 I followed the part 13.3.2 from 
 http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/boot-blocks.html
 
 I think this should be enough, isn't it? it says bsdlabel -B will 
 replace the boot1 and boot2 stage so all of them are installed.

Looks correct.



 Now the question is how to branch the a partition as the boot partition ?

No need. As soon as the branching from ada0-start - ada0s3
has been processed, the 'a' partition ada0s3a will be accessed
as it is the boot partition. It will then continue stage 1 and 2
and finally access the loader, which will load the kernel.

In 13.3.2 it is explained as follows:

They [Stage One, /boot/boot1, and Stage Two, /boot/boot2]
are located outside file systems, in the first track of
the boot slice, starting with the first sector. This is
where boot0, or any other boot manager, expects to find
a program to run which will continue the boot process.
The number of sectors used is easily determined from the
size of /boot/boot.

In your case, the boot slice (for FreeBSD) is ada0s3 where the
boot manager EasyBCD will branch to.

Getting just a cursor (as you described) makes it hard to
identify where the process hangs. If EasyBCD is the last
thing you see, I assume the FreeBSD boot process isn't even
initiated. Every part of it (MBR boot manager, boot0, boot1,
boot2 and loader) would issue some kind of text when accessed.


-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org

Re: dbus, epiphany, rekonq