from:"Polytropon"

On Thu, 7 Mar 2013 10:01:03 GMT, Anton Shterenlikht wrote:
 I have a process that eats up al memory,
 in my case science/paraview if I try to
 analyse a large model. What should FreeBSD
 do when a process tries to use all RAM or more?

In this case, the swap space would be used, until the
system runs out of swap space.



 I my case I get a complete freeze, can't even
 login from the console, and requiring a cold
 reboot. I guess this is not supposed to happen,
 but what is supposed to happen in situations like this?

A normal reboot (including a proper shutdown) should
at least be possible. If the machine seems to freeze
entirely, this simply looks wrong, so maybe it's more
than just eating all the RAM?

You could try to impose a resource limit, see man limits
for details, so you could trigger the undesired behaviour
while e. g. only 50% of the available RAM is being used
by _that_ process (and therefor still leaving enough
resources for other system and user processes). You could
also monitor resource consumption with tools like top,
htop, vmstat or systat in adjacent xterms while you run
the test, seeing trouble pile up...



-- 
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: Raspberry Pi

On Thu, 7 Mar 2013 10:34:37 +, Nick Pettefar wrote:
 Hi, where can I download the Raspberry Pi version please?

I think you'll need the arm version of FreeBSD.
Check those links for more information:

http://www.raspberrypi.org/archives/3094

http://www.freebsdarm.org/

http://www.freebsd.org/platforms/arm.html

https://wiki.freebsd.org/WhatsNew/FreeBSD10#Overall_system_.2F_architectural_changes

http://people.freebsd.org/~dmarion/beaglebone/creating_bootable_sd_card/

http://www.raspberrypi.org/phpBB3/viewtopic.php?f=85t=30148

You can obtain more references by employing a google search. :-)



-- 
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: fetchmail/sendmail: Domain of sender address does not exist

On Thu, 7 Mar 2013 09:40:47 GMT, Anton Shterenlikht wrote:
 How do I set fetchmail and sendmail to fetch
 such emails?

Maybe it helps if you add the options fetchall flush to
your .fetchmailrc configuration file? I've had a similar
problem some years ago and I think this was the solution.
See man fetchmail for the corresponding command line
options (and you could probably add -v to see what's
actually going on).



-- 
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: fetchmail/sendmail: Domain of sender address does not exist

On Thu, 7 Mar 2013 21:55:57 GMT, Anton Shterenlikht wrote:
 And Matthias already helped me sort it out.

Could you write to the list how you solved the problem?
I think it would be interesting to those running into
similar problems.

I remember that in the end, my clever solution involved
logging into the ugly webmailer of my ISP and deleting
the few offending messages manually. It should be easier
than that. :-)

Thanks in advance.



-- 
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: backups using rsync

2013-03-05 Thread Polytropon

On Mon, 04 Mar 2013 12:19:09 -0800, Ronald F. Guilmette wrote:
 
 In message 20130304125634.8450cfaf.free...@edvax.de, 
 Polytropon free...@edvax.de wrote:
 
 On Mon, 04 Mar 2013 03:35:30 -0800, Ronald F. Guilmette wrote:
  Now, unfortunately, I have just been bitten by the evil... and apparently
  widely known (except to me)... ``You can't use dump(8) to dump a journaled
  filesystem with soft updates'' bug-a-boo.
 
 There are other tools you can use, for example tar or cpdup
 or rsync, as you've mentioned in the subject.
 
 tar I already knew about, but I think you will agree that it has lots of
 limitations that make it entirely inappropriate for mirroring an entire
 system.

That's true. If your purpose is backup of data files,
tar is a good tool, especially for cross-platform use.
But if you need to deal with exceptional things like
extended permissions, ACL, sparse files and such, you
will quickly see its limits. On the other hand, it can
be used for multi-volume savesets, but this is not your
intention.



 This cpdup thing is entirely new to me.  Thanks for mentioning it!  I really
 never heard of it before, but I just now installed it from ports, and I'm
 perusing the man page. 

It's a little bit comparable to rsync and can also do
things like only add (so you won't lose any files:
if they are removed in source, they will be kept in
backup). It also has limitations that rsync will not.



 It looks very promising.  Too bad it doesn't
 properly handle sparse files, but oh well.  That's just a very minor nit.
 (Does it properly handle everything else that rsync claims to be able to
 properly handle, e.g. ACLs, file attributes, etc., etc.?)

That's something you should check with an example
dataset you back up, restore, and compare. I've been
using it for normal files successfully.



 The same problems that apply when dumping live systems can
 bite you using rsync,
 
 What problems are we talking about, in particular?

The problems I'm refering to is the kind of _possible_
trouble you can get into when backing up files that
keep changing. The ability to make a snapshot prior
to starting the backup is a great help here (if you
don't have the chance to unmount the partitions to
backup). I can't imagine _how_ programs will react
if they start reading a file, prepare sufficient space
in some kind of TOC, then continue reading while the
file grows... or if a file is being read which is
removed during read... If you minimize the writing
activity to the (still) _live_ data you're dealing
with, that could be a benefit.




 I am guessing that if I use rsync, then I *won't* encounter this rather
 annoying issue/problem relating to UFS filesystems that have both soft
 updates and journaling enabled, correct?
 
 but support for this on file system
 level seems to be better in rsync than what dump does on
 block level.
 
 What exactly did you mean by this ?

As mentioned above: Unexpected and unpredictable results,
strange kinds of inconsistency, may they appear during
backup or later on restore.



  If I use all of the following rsync options...  -a,-H,-A, -X, and -S 
  when trying to make my backups, and if I do whatever additional fiddling
  is necessary to insure that I separately copy over the MBR and boot loader
  also to my backup drive, then is there any reason that, in the event of
  a sudden meteor shower that takes out my primary disk drive while leaving
  my backup drive intact, I can't just unplug my old primary drive, plug in
  my (rsync-created) backup drive, reboot and be back in the sadddle again,
  almost immediately, and with -zero- problems?
 
 You would have to make sure _many_ things are consistent
 on the backup disk.
 
 Well, this is what I am getting at.  This is/was the whole point of my post
 and my question.  I want to know:  What is that set of things, exactly?

The backup disk (or failover disk, as I said) needs to be
initialized properly prior to the first backup run: Make
sure it's bootable. Depending on how you handle identification
of the disk (by device name, by label or UFSID) and how
you're going to boot from it (by selecting the failover
disk in some post-BIOS/POST dialog or by swapping cables
or bays), you should check it actually starts booting.



 Regarding terminology, that would make the disk a failover disk
 
 OK.  Thank you.  I will henceforth use that terminology.

Just a suggestion from how you described you will be
using the disk, which isn't what commonly (or mostly)
is expressed by the term backup (also cf. archive
which is something entirely different).



 The disk would need to have an initialized file system and
 a working boot mechanism, both things rsync does not deal with
 
 Check and check.  I implicitly understood the former, and I explicitly
 mentioned the latter in my original post in this thread.
 
 But is there anything else, other than those two things (which, just as
 you say, are both clearly outside of the scope of what rsync does

Re: Grepping though a disk

On Mon, 4 Mar 2013 10:09:50 +0100, Damien Fleuriot wrote:
 Hey that's actually a pretty creative way of doing things ;)

It could be more optimum. :-)

My thought is that I could maybe use a better bs= to make
the whole thing run faster. I understand that for every
unit, a subprocess dd | grep is started and an if [] test
is run. Maybe doing this with 1 MB per unit would be better?
Note that I need to grep through 1 TB in 10 kB steps...



 Just to make sure, you've stopped daemons and all the stuff
 that could potentially write to the drive and nuke your
 blocks right ?

Of course. The /dev/ad6 disk is a separate data disk which
is not in use at the moment (unmounted). Still it is possible
that the block has already been overwritten, but when the
search has finished, it's almost certain in what state the
data is.

I would rewrite the file, but my eidetic memory is not working
well anymore, so I can only remember parts of it... :-(




-- 
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: Grepping though a disk

On Mon, 04 Mar 2013 04:15:48 -0600, Joshua Isom wrote:
 I'd call bs= essential for speed.  Any copying will be faster with 
 something higher.

I thought about that. Narrowing down _if_ something has
found is easy, e. g. when the positive 1 MB unit is dd'ed
to a file, further work can easily be applied.



 Also, there's the possibility, very annoying, that 
 your search string overlaps a place where you read.  I'd probably check 
 1M blocks, but advance maybe 950k each time.

I also thought about that, that's why the distinctive
phrase I'm searching for is less than 10 characters long.
Still it's possible that it appears across a boundary
of units, no matter how big or small I select bs=.

But I don't know how to do this. From reading man dd
my impression (consistent with my experience) is that
the option skip= operates in units of bs= size, so I'm
not sure how to compose a command that reads units of
1 MB, but skips in units of 950 kB. Maybe some parts of
my memory have also been marked unused by fsck. :-)



 Make sure you're reading 
 from block offsets for maximum speed.

How do I do that? The disk is a normal HDD which has been
initialized with newfs -U and no further options.

ad6: 953869MB Hitachi HDS721010DLE630 MS2OA5R0
at ata3-master UDMA100 SATA 1.5Gb/s

The file system spans the whole disk.



 I know disk editors exist, I 
 remember using one on Mac OS 8.6 for find a lost file.  That was back on 
 a 6 gig hard drive.

Ha, I've done stuff like that on DOS with important business
data many years ago, using the Norton Disk Doctor (NDD.EXE)
when Norton (today: Symantec) wasn't yet synonymous for The
Yellow Plague. This program actually was quite cool, and you
could search for things, manipulate disks on several levels
(files, file system and below). I had even rewritten an entire
partition table from scratch, memory and handheld calculator
after an OS/2 installation went crazy. :-)



 Depending on the file size, you could open the disk in vi and just 
 search from there, or just run strings on the disk and pipe it to vi.

You mean like strings /dev/ad6 | something, without dd? That
would give me _no_ progress indicator (with my initial approach
I have increasing numbers at least), but I doubt I can load a
1 TB partition in a vi session with only 2 GB RAM in the machine.

If I try strings /dev/ad6 I get a warning: strings: Warning:
'/dev/ad6' is not an ordinary file. True. But this opens a
useful use of cat: cat /dev/ad6 | strings. (Interesting idea,
I will investigate this further.)

The file size of the file I'm searching for is less than 10 kB.
It's a relatively small text file which got some subsequent
additions in the last days, but hasn't been part of the backup
job yet. I can only remember parts of those additions, because
as I said my brain is not good with computer. :-)

Or do you think of something different? If yes, please explain.
The urge to learn is strong when something went wrong. :-)


-- 
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: Grepping though a disk

On Mon, 4 Mar 2013 11:29:00 +, Steve O'Hara-Smith wrote:
 On Mon, 4 Mar 2013 12:15:24 +0100
 Polytropon free...@edvax.de wrote:
 
  But I don't know how to do this. From reading man dd
  my impression (consistent with my experience) is that
  the option skip= operates in units of bs= size, so I'm
  not sure how to compose a command that reads units of
  1 MB, but skips in units of 950 kB. Maybe some parts of
  my memory have also been marked unused by fsck. :-)
 
   Not too hard (you'll kick yourself when you read down) - translation
 to valid shell script is left as an exercise for the reader :)
 
  bs=50k count=(n*20) skip=(n*20 - 1)
 
   Probably nicer to use powers of 2
 
 bs=64k count=(n*16) skip=(n*16 - 1)

Thanks for the pointer. I was so concentrated on finding
the answer within dd that I hadn't thought about that.
It's easy to write this in shell code. As a conclusion,
I will apply for further IQ reduction, seems that I have
enough spare brain power I don't use 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: backups using rsync

On Mon, 04 Mar 2013 03:35:30 -0800, Ronald F. Guilmette wrote:
 Now, unfortunately, I have just been bitten by the evil... and apparently
 widely known (except to me)... ``You can't use dump(8) to dump a journaled
 filesystem with soft updates'' bug-a-boo.

There are other tools you can use, for example tar or cpdup
or rsync, as you've mentioned in the subject.



 I _had_ planned on using dump/restore and making backups from live mounted
 filesystems while the system was running.  But I really don't want to have
 to take the system down to single-user mode every week for a few hours while
 I'm making my disk-to-disk backup.  So now I'm looking at doing the backups
 using rsync.

The same problems that apply when dumping live systems can
bite you using rsync, but support for this on file system
level seems to be better in rsync than what dump does on
block level.



 If I use all of the following rsync options...  -a,-H,-A, -X, and -S 
 when trying to make my backups, and if I do whatever additional fiddling
 is necessary to insure that I separately copy over the MBR and boot loader
 also to my backup drive, then is there any reason that, in the event of
 a sudden meteor shower that takes out my primary disk drive while leaving
 my backup drive intact, I can't just unplug my old primary drive, plug in
 my (rsync-created) backup drive, reboot and be back in the sadddle again,
 almost immediately, and with -zero- problems?

You would have to make sure _many_ things are consistent
on the backup disk.

Regarding terminology, that would make the disk a failover
disk, even if the act of making it the actual work disk
is something you do manually.

The disk would need to have an initialized file system and
a working boot mechanism, both things rsync does not deal
with, if I remember correctly. But as soon as you have
initialized the disk for the first time and made sure (by
testing your first result of a rsync run), it should work
with any subsequent change of data you transfer to that
disk.



 P.P.S.  Before anyone asks, no I really _do not_ want to just use RAID
 as my one and only backup strategy. 

RAID _is_ **NO** backup. It's for dedundancy and performance.
If something is erased or corrupted, it's on all disks. And
all the disks permanently run. A backup disk only runs twice:
when backing something up, or when restoring. In your case,
restoring means that the disk is put into operation in its
role as a failover disk.



 RAID is swell if your only problem
 is hardware failures. 

Still hardware failures can corrupt data on all participating
disks.



 As far as I know however it will not save your
 bacon in the event of a fumble fingers rm -rf * moment.  Only frequent
 and routine actual backups can do that.

Correct. It's important to learn that lesson _before_ it is
actually needed. :-)




-- 
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

Grepping though a disk

2013-03-03 Thread Polytropon

Due to a fsck file system repair I lost the content of a file
I consider important, but it hasn't been backed up yet. The
file name is still present, but no blocks are associated
(file size is zero). I hope the data blocks (which are now
probably marked unused) are still intact, so I thought
I'd search for them because I can remember specific text
that should have been in that file.

As I don't need any fancy stuff like a progress bar, I
decided to write a simple command, and I quickly got
something up and running which I _assume_ will do what
I need.

This is the command I've been running interactively in bash:

$ N=0; while true; do echo ${N}; dd if=/dev/ad6 of=/dev/stdout 
bs=10240 count=1 skip=${N} 2/dev/null | grep PATTERN; if [ $? -eq 0 ]; 
then break; fi; N=`expr ${N} + 1`; done

To make it look a bit better and illustrate the simple
logic behind my idea:

N=0
while true; do
echo ${N}
dd if=/dev/ad6 of=/dev/stdout bs=10240 count=1 skip=${N} \
2/dev/null | grep PATTERN
if [ $? -eq 0 ]; then
break
fi
N=`expr ${N} + 1`
done

Here PATTERN refers to the text. It's only a small, but
very distinctive portion. I'm searching in blocks of 10 kB
so it's easier to continue in case something has been found.
I plan to output the resulting block (it's not a real disk
block, I know, it's simply a unit of 10 kB disk space) and
maybe the previous and next one (in case the file, the _real_
block containing the data, has been split across more than
one of those units. I will then clean the garbage (maybe
from other files) because I can easily determine the beginning
and the end of the file.

Needless to say, it's a _text_ file.

I understand that grep operates on text files, but it will
also happily return 0 if the text to search for will appear
in a binary file, and possibly return the whole file as a
search result (in case there are no newlines in it).

My questions:

1. Is this the proper way of stupidly searching a disk?

2. Is the block size (bs= parameter to dd) good, or should
   I use a different value for better performance?

3. Is there a program known that already implements the
   functionality I need in terms of data recovery?

Results so far:

The disk in question is a 1 TB SATA disk. The command has
been running for more than 12 hours now and returned one
false-positive result, so basically it seems to work, but
maybe I can do better? I can always continue search by
adding 1 to ${N}, set it as start value, and re-run the
command.

Any suggestion is welcome!



-- 
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: rm -R

2013-03-02 Thread Polytropon

On Sat, 2 Mar 2013 18:27:15 -0800, Gary Kline wrote:
 On Sat, Mar 02, 2013 at 07:43:50PM -0600, Joshua Isom wrote:
  On 3/2/2013 2:43 PM, David Tilbrook wrote:
  The problem (and its solution) have been
  raised for at least 39 years.
  
  But google and other search engines treats words([A-Za-z0-9])
  starting with - as meaning exclude results with this, even when
  obvious it's about unix commands.  It can be rather annoying when
  searching for help.
  ___
 
 
   sorry to be so dense; can you give me a few concrete examples? ya
   lost me!

Just as one of many examples, google for find -name (without
the quotes of course), a typical combination which the AND NEAR
search should return sufficient results for. Compare the list
of results to what you would expect.

Now, google for find -name _including_ the quotes (!)
in order to have google treat the search string literally.
The results will be different, as you would expect.

In ye olde times when search engine meant altavista.digital.com
(if I remember correctly), search strings usually needed
to include + (AND) and - (AND NOT) if you wanted to construct
a search term consisting of more than one word. With google
implying a + prefix for every search word _and_ assuming
they should be as near as possible to each other (therefor
the name AND NEAR for the kind of search) this became
obsolete. Almost, as it seems...



-- 
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: rm -R

2013-03-02 Thread Polytropon

On Sat, 02 Mar 2013 23:45:30 +0100, Julian H. Stacey wrote:
 Hi,
 Reference:
  From:   Jos Chrispijn ker...@webrz.net 
  Date:   Sat, 02 Mar 2013 18:44:22 +0100 
  Message-id: 51323a76.2040...@webrz.net 

 Jos Chrispijn wrote:

  Teske, Devin:
   rm -R -- -S

   The -- tells it here's the end of the options, here come the 
   file/directories

  Almost:

  rm -R --  -S;

  did it, thanks very much for you help!

 This also works
   rmdir ./-S
 ( is probably the best generic naming method,  was valid decades
 ago, before rm got the luxury of modern stuff eg --  would work
 for other commands that might not have delimieters such as -- )
 This also work but is over kill :
   rmdir './-S'

Just note that the ; has been part of the name in question,
so the end of command sign would have to be part of the
directory name: rmdir ./-S; and rmdir './-S;' would be
the alternatives to rm -R -- -S;.

I'd be interested in what happens when you have such a
directory name and press PF8 in the Midnight Commander
in order to delete it.

Now go ahead and create a file * in / and tell a junior
sysadmin to remove it. ;-)

-- 
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: https://wiki.freebsd.org/ certificate error

2013-03-01 Thread Polytropon

On Fri, 01 Mar 2013 10:42:58 -0500, Fbsd8 wrote:
 Javad Kouhi wrote:
  Also no problem with FreeBSD 9.1 and chromium. But sometimes ago I have
  this problem with all https sites. because the government forged the wrong
  SSL certificate and my browser and my browser warned me about it. Do you
  have this problem with other websites?
  
  On Fri, Mar 1, 2013 at 6:02 PM, Zyumbilev, Peter 
  pe...@aboutsupport.comwrote:
  
 
  On 01/03/2013 16:14, Ralf Mardorf wrote:
 
  [1] $ firefox -version
  Mozilla Firefox 19.0
 
  No problem with SeaMonkey 2.16.
 
 
  Peter
 I use xp browser and it's certificate checking is enabled.

You are sure using a more than 10 year old system should
be considered safe enough to provide a reference?



 Maybe the browsers running from xorg desktops are NOT certificate aware 
 so them not getting the error warning would be expected.

They are. Or to be correct: The most prominent ones are,
like Firefox, Chrome, and Opera. More lightweight browsers
like dillo actually might not have this functionality.



 The fact remains, the ms/browsers do find the wiki.freebsd.org wedsite's 
   certificate invalid because the certificate ip address does not match 
 the ip address the public dns points to.

As it has been mentioned, one certificate can be used for
several IP addresses. Both www and wiki are located at
8.8.178.110 (returned by host command), so there might
be a DNS issue or something comparable strange...

I've checked with Opera 11.50 here, no problems.




-- 
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: Fat Fingered An 'rm -rf' of Important Files

2013-02-28 Thread Polytropon

On Wed, 27 Feb 2013 21:08:58 -0600, Joseph A. Nagy, Jr wrote:
 Okay, I know I should pay more attention to what I'm doing, and having 
 separate partitions isn't an excuse for regular backups. If we can skip 
 the finger wagging on that part I'd appreciate it.

I've experienced similar and different moments of unintended
successful rm, so I won't mention missing backups. ;-)



 Is there any way to retrieve any of them?

Yes, but it's not easy. Prepare to go on a journey though
file system documentation, trial  error.

Obviously we're talking about a USB stick, so no TB amount
of data has to be processed. First of all: Do _not_ alter
the USB stick in any way. No matter what you do, it can
always get worse.



 I've not wrote any data to 
 either partition since the accidental deletion.

Very good.

You can first make a copy of the file system (the whole
stick) and use that: It will be faster to access and if
you do something wrong, the original data (which we can
assume is still there) won't be affected:

# dd if=/dev/da0 of=stick.dd

Now let me introduce you to the list of helpful programs
in case you've done something ultimately stupid which I
have already repeated several times on this mailing list.
I'm sure you can find some program that will help you.
See my individual notes regarding your specific situation.

I will refactor text from a previous message.

A worst-case tool to recover data (not file names, but file
content) is testdisk; in ports: sysutils/testdisk. It's also
on some diagnostics and recovery CDs like UBCD.

You can also try this:

# fetch -rR device

Also recoverdisk could be useful.

The ports collection contains further programs that might be
worth investigating; just in case they haven't been mentioned
yet:

ddrescue
dd_rescue   - use this to make an image of the stick!
magicrescue
testdisk- restores content
recoverjpeg
foremost
photorec

Then also

ffs2recov
scan_ffs

should be mentioned.

And finally, the cure to everything is found in The Sleuth Kit
(in ports: tsk):

fls
dls
ils
autopsy

Keep in mind: Read the manpages before using the programs. It's
very important to do so. You need to know what you're dealing
with, or you'll probably fail. There is no magical tetroplyrodon
to click ^Z and get everything back. :-)

Proprietary (and expensive) tools like R-Studio or UFS Explorer
can still be considered worth a try. Their trial versions are for
free. UFS Explorer even works using wine (I've tried it).

Note:

I've dealt with a comparable problem some months ago when
a Windows PC has repaired a FAT file system on a USB
stick, with the excellent result of all data being gone.
I could restore everything except the original file names
(which I wrote a script to conclude them from file metadata
and content).

So it should be possible.




Good luck!




-- 
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: Fat Fingered An 'rm -rf' of Important Files

2013-02-28 Thread Polytropon

On Thu, 28 Feb 2013 05:55:45 -0600, Joseph A. Nagy, Jr wrote:
 On 02/28/13 03:02, Polytropon wrote:
  On Wed, 27 Feb 2013 21:08:58 -0600, Joseph A. Nagy, Jr wrote:
  Obviously we're talking about a USB stick, so no TB amount
  of data has to be processed. First of all: Do _not_ alter
  the USB stick in any way. No matter what you do, it can
  always get worse.
 
 Oh no, not a USB stick, I'm talking hdd partitions (4GiB on one, 64GiB 
 on another) but it will be the same process. I'll see about investing in 
 an external hard drive.

Okay, than I must have misread something. But the same
idea applies: Do not alter the disk in question (or at
least not the partitions). If you have a spare disk, make
a dd copy of the partition and work with that. Disks are
cheap, your data is not. You're going to need a new disk
from time to time anyway, so enough space to try out things
is worth the purchase.



  I've not wrote any data to
  either partition since the accidental deletion.
 
  Very good.
 
  You can first make a copy of the file system (the whole
  stick) and use that: It will be faster to access and if
  you do something wrong, the original data (which we can
  assume is still there) won't be affected:
 
  # dd if=/dev/da0 of=stick.dd
 
  Now let me introduce you to the list of helpful programs
  in case you've done something ultimately stupid which I
  have already repeated several times on this mailing list.
  I'm sure you can find some program that will help you.
  See my individual notes regarding your specific situation.
 
  I will refactor text from a previous message.
 
  A worst-case tool to recover data (not file names, but file
  content) is testdisk; in ports: sysutils/testdisk. It's also
  on some diagnostics and recovery CDs like UBCD.
 
 I've tried test disk but it doesn't have a UFS option for some reason.

I also had this problem with testdisk on UFS (also with
accidentally deleted files). It might be that it needs
to operate on a slice, not on a partition, but I am not
sure about this.



  You can also try this:
 
  # fetch -rR device
 
 where would I fetch to?

It will fetch to the current directory, that's why my
suggestion with a disk big enough for restoring data.



  Also recoverdisk could be useful.
 
 also in /usr/ports/sysutils ?

No, it's already part of the OS: /sbin/recoverdisk,
and therefor man recoverdisk is available.



  The ports collection contains further programs that might be
  worth investigating; just in case they haven't been mentioned
  yet:
 
  ddrescue
  dd_rescue   - use this to make an image of the stick!
  magicrescue
  testdisk- restores content
  recoverjpeg
  foremost
  photorec
 
  Then also
 
  ffs2recov
  scan_ffs
 
  should be mentioned.
 
  And finally, the cure to everything is found in The Sleuth Kit
  (in ports: tsk):
 
  fls
  dls
  ils
  autopsy
 
 awesome

A suggestion: If you have understood the basics of a UFS
file system (e. g. by reading McKusick's famous article),
you'll see that data on disk basically equals inodes as
a form of database that point to blocks on disks, either
directly or to blocks pointing to other blocks. There are
tools to write the content of the blocks into files, for
example icat (from TSK). You can find out which blocks
are in use by files using the command blocks from
within fsdb (an OS tool); fsdb -r device to invoke,
and man fsdb for instructions. Note that deleted files
cause two things: The inode information will be cleared,
and the blocks will be marked unused. But chances are
good that the blocks theirselves are intact, i. e. they
still containt the data.



  Keep in mind: Read the manpages before using the programs. It's
  very important to do so. You need to know what you're dealing
 
 the testdisk man page is very unhelpful. ):

Sadly true. :-(

The program emphasizes a self-explaining interactive
way to be used and shortens the amount of information
provided in the manpage, because... well, nobody reads
them, and people expect to simply click on undelete
anyway. :-)

There's better information here:

http://www.cgsecurity.org/wiki/TestDisk

Go down to Documentation and see what you can use.
But note that TestDisk primarily is intended to deal
with partition tables and file tables (primarily FAT-
based). To recover files (as in: data that is somewhere
on the disk, but not in the file system anymore), give
PhotoRec a try.

More info than in its manpage here:

http://www.cgsecurity.org/wiki/PhotoRec

See my comment regarding MagicRescue later.



  with, or you'll probably fail. There is no magical tetroplyrodon
  to click ^Z and get everything back. :-)
 
 well, not so sure. Found a Windows program to recover deleted stuff on 
 UFS, but I imagine it's harder touse then the above (and more 
 expensive).

There are recovery tools for Windows, but as you said,
those that actually work are quite expensive. Tools like
TSK can do the same

Re: how to disable bluetooth

2013-02-27 Thread Polytropon

On Mon, 25 Feb 2013 14:56:46 +0100, CeDeROM wrote:
 hey, how can i disable bluetooth in freebsd (9.1)?

Kill it with fire! ;-)



 my device is visible to other devices whethever i switch the radio button,
 also the radio button seems to be the only waynto disable my computer from
 being visible to other devices...
 
 i hwve tried to disable bluetooth and ubt in loader.conf and various
 serives in rc.conf but still my computer was visible to other devices. i
 dont want it to be visible to other bluetooth computers but still i want to
 have radio switched on to use wifi.

You can try to omit Bluetooth entirely by defining the symbol
WITHOUT_BLUETOOTH=yes in /etc/src.conf and rebuilding your
system (kernel and world). See man src.conf for details.


-- 
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: BAD link

2013-02-25 Thread Polytropon

On Mon, 25 Feb 2013 12:30:02 -0500, Marv Graham wrote:
 The link to the BSD Toolkit is broken!?!?
 
 It's on this page: 
 http://www.freebsdmall.com/cgi-bin/fm/bsdpow?id=hztTbuMRmv_pc=664

Doesn't seem to be broken here. A page containint a version
6 of FreeBSD is presented.

Can you be more specific with what you mean by broken?



 Will it's 'parts' run on a MAC, which has Apple-modified FreeBSD
 as its core ??

Probably not. However, there are means to run the PPC version
of FreeBSD on some Apple hardware, as well as it's maybe
possible to run FreeBSD's x86 version on some x86 Macs.

However, it should be possible to use VirtualBSD within a
virtualisation environment (i. e. in a VM) on Mac OS X and
other operating systems. It even comes with the currently
popular Mac look  feel.

See: http://www.virtualbsd.info/



 What EXACTLY is PC-BSD?

It's an operating system and applications package with preinstalled
and precondigured desktop environment and custom installer, using
FreeBSD at its heart.

See: http://www.pcbsd.org/



 Does it install on a WINDOZE OS ?

No. You cannot install an operating system on Windows.




-- 
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: Why not simplify Copyright at boot/dmesg?

2013-02-23 Thread Polytropon

On Sat, 23 Feb 2013 16:47:10 +0100, vermaden wrote:
 Why not simplify that:
 
 | Copyright (c) 1992-2013 The FreeBSD Project.
 | Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
 | The Regents of the University of California. All rights reserved.
 | FreeBSD is a registered trademark of The FreeBSD Foundation.
 | (...)
 
 ... into that:
 
 | Copyright (c) 1992-2013 The FreeBSD Project.
 | Copyright (c) 1979-1994 The Regents of the University of California.
 | FreeBSD is a registered trademark of The FreeBSD Foundation.
 | (...)

Because you need to exclude 1981, 1982, 1984, 1985, 1987 and 1990
which are missing in list of years. :-)



-- 
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: Why not simplify Copyright at boot/dmesg?

2013-02-23 Thread Polytropon

On Sat, 23 Feb 2013 18:14:48 +0100, vermaden wrote:
 
 
 Od: Polytropon free...@edvax.de
 Do: vermaden verma...@interia.pl; 
 Wysłane: 17:11 Sobota 2013-02-23
 Temat: Re: Why not simplify  Copyright at boot/dmesg?
 
  On Sat, 23 Feb 2013 16:47:10 +0100, vermaden wrote:
   Why not simplify that:
   
   | Copyright (c) 1992-2013 The FreeBSD Project.
   | Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
   | The Regents of the University of California. All rights 
   reserved.
   | FreeBSD is a registered trademark of The FreeBSD Foundation.
   | (...)
   
   ... into that:
   
   | Copyright (c) 1992-2013 The FreeBSD Project.
   | Copyright (c) 1979-1994 The Regents of the University of California.
   | FreeBSD is a registered trademark of The FreeBSD Foundation.
   | (...)
  
  Because you need to exclude 1981, 1982, 1984, 1985, 1987 and 1990
  which are missing in list of years. :-)
 
 It may sound like ignorance, but why we need to exclude them?

To be honest: I have no idea. It's just that I noticed
this at first sight.

I would assume there is some specific legal sense behind
this naming and counting convention; two lawyers, three
opinions might apply. :-)



 We do not exclude any years for FreeBSD Project ;)

But The FreeBSD Project as a copyright holder covers a
different time frame than The Regents of the University
of California, so this seems to be some specific difference
causing two lines of copyright information.



-- 
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: cdrecord: Please report.

2013-02-21 Thread Polytropon

On Wed, 20 Feb 2013 15:52:39 -0800, Ronald F. Guilmette wrote:
 
 
 I am not a happy camper.
 
 Now that I've ``upgraded'' from 8.3-RELEASE to 9.1-RELEASE it appears that
 good old burncd no longer works, apparently because the CD/DVD drive is
 now exclusively handled as an ATAPICAM device.

That was to be expected.




 So I try to use cdrecord and I get this:
 
 # cdrecord dev=0,0,0 driveropts=burnfree FreeBSD-9.1-RELEASE-i386-disc1.iso

Try instead:

# cdrecord dev=0,0,0 speed=12 -v -eject -tao -data file

Some drives get crazy when you don't provide a slow burning
speed. Add other options (burnfree) if needed.

Just for comparison: If you want to burn a music CD:

# cdrecord dev=0,0,0 speed=12 -v -eject -dao -audio track

You can see the obvious difference -data vs. -audio, but also
-tao (track at once) vs -dao (disk at once).



 cdrecord: The current problem looks like a buffer underrun.

Is your drive faster than your disk which hands in the data?
Try to reduce the writing speed.



 cdrecord: It looks like 'driveropts=burnfree' does not work for this drive.

A clear message. You can use the -prcap option to find out
what your recorder will support; see man cdrecord for
details.



 So, um, WTF?  To whom am supposed to report this failure?

To your superior. :-)



 And more to the point, how can I burn a simple damn CD now?

Try the command mentioned above. I've left burncd somewhere
in FreeBSD 5 and I'm now using cdrecord exclusively. When run
as a !root user, make sure permissions to access the recorder
are set properly. Maybe add an alias for your dialog shell.



-- 
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: sh script files

2013-02-20 Thread Polytropon

On Wed, 20 Feb 2013 07:40:08 -0500, Fbsd8 wrote:
 # write to file
   ${file}
 
 I'm thinking the file is never closed so on power failure I loose the 
 contents of the file.
 
 How would I code a command to close the file?

The file is closed when the write operation has been
finished. You can use the sync command to flush
pending writes to the file (as writing is handled
asynchronously by the system). When the program
that writes to the file exits, it will close the
file it's writing to. This depends on the command
you're using infront of , as the command you've
shown will simply generate a null file (file with
no actual content).


-- 
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 add zfs support to FreeBSD?

2013-02-18 Thread Polytropon

On Mon, 18 Feb 2013 16:57:14 -0500, Fbsd8 wrote:
 Fbsd8 wrote:
  The handbook does not cover how to add zfs support.
  
  How is it done?
 
 
 Let me reword. If zfs is in the base system why does it not show up
 when I look for it this way?
 
 if config -x $( sysctl -n kern.bootfile ) | grep -q 
 '^[[:space:]]*options[[:space:]]\{1,\}ZFS\'; then
echo yes zfs is in the kernel
 fi

Without the ability to check this, I strongly assume that
if you enable ZFS as described in the Handbook, the module
/boot/kernel/zfs.ko (part of the default system) will be
loaded. That's why it won't show up in a sysctl query
aimed at the _kernel_ itself -- because it isn't in the
kernel.

Also, sysctl -n kern.bootfile will return the actual
kernel file, /boot/kernel/kernel, which is a binary. If
the exact config list (from the kernel _configuration_
file) is not plain-text part of that file, grep will not
find the text you're grepping for.



-- 
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 add zfs support to FreeBSD?

2013-02-18 Thread Polytropon

On Mon, 18 Feb 2013 20:41:19 -0500, Fbsd8 wrote:
 So the next question is there any sh script code I can use to
 check if zfs has been enabled by the rc.conf zfs_enable statement.
 
 I need to determine if zfs is enabled on the host.

Even though the statement zfs_enable=YES may be part of
/etc/rc.conf, it's still possible that the ZFS subsystem
is _currently_ not running. So in my opinion you should
check with something that relies on ZFS actually running.
So there could be a difference between is enabled and
is running.

Regarding is enabled, you can easily check if the
corresponding line in /etc/rc.conf is present. This could
result in a false-positive answer regarding is running.

For example, if I add zfs_enable=YES to /etc/rc.conf
and check its presence (e. g. using grep), but I don't
do anything else; I'll get this:

% zfs
internal error: failed to initialize ZFS library

Here's an example from a system not running ZFS:

/sbin/zfs  /dev/null 21
if [ $? -eq 0 ]; then
echo ZFS is currently running.
else
echo ZFS is currently _not_ running.
fi

Check on a system running ZFS. :-)


-- 
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: clicking on URL'S in mutt.

2013-02-18 Thread Polytropon

On Mon, 18 Feb 2013 21:04:05 -0800, Gary Kline wrote:
   for some reason, kmail is busted and I need a way of getting
   mutt to spawn firefox.  Anybody?

I can only suggest the most basic method (which should work
with any text mode MUA, even those without any mouse support).
It requires that you already have a Firefox running, e. g. on
a 2nd workspace. First select the URL in the mail message with
the left mouse button, then switch over to the Firefox window
and click the middle mouse button (or press down the mouse
wheel if you don't have a normal 3 button mouse). If the URL
is encapsulated in spaces, a double-click would select it,
so you don't need to manually select it from its beginning
to its end (or vice versa).



   ps:  this is from a linujx desktop running kde... 

If your mail storage is intact, can't you (temporarily) use
Thunderbird to access it? If I remember correctly, KMail and
Thunderbird are using the same storage format (mbox, I think)...




-- 
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: bsd lost partition recovery

2013-02-17 Thread Polytropon

On Sun, 17 Feb 2013 18:30:28 +0700, Erich Dollansky wrote:
  On Sun, Feb 17, 2013 at 2:31 PM, Erich Dollansky 
  erichsfreebsdl...@alogt.com wrote:
  
   On Sun, 17 Feb 2013 12:56:54 +0330
   takCoder tak.offic...@gmail.com wrote:
  
*here's the question:*
how to restore lost data of a formatted bsd partition?!
   
   restore it from a backup?
  
  
  no, from the formatted hard drive.. if only i had a backup..
  
 You did it the hard way.
  
  
*and here's what has happened to me:*
i was trying to install windows xp sp2 on a HD on my system to
move it to another hardware after that for some reasons, but i
had my main HD attached to may system as well..
   
all of a sudden i made the worst misktake ever and formatted my
main HD's bsd partition instead!!!
  
   If you want to read the data from that partition it will depend very
   much on how far the Windows installation really went.
  
   i just deleted the partition and then found out what i just did,
   then i
  stopped the installation process and start looking ways for recovery..
  
  till now, somewhere in the web i saw that R-Studio supports ufs
  recovery but it's a windows application.. i'm preparing a system for
  checking it right now..
  
 Just try it.

You can try R-Studio Emergency, a live CD, to _check_ if there
is anything to restore. This step is for free, as you don't need
to buy it. Also there's no need for installing anything on some
Windows.

However, there are tools that can be used on FreeBSD (if you can
put the disk into another system). Note: Do not write to that
disk! If possible, make a dd image of the disk and work with the
image, do not _alter_ the original disk in any way. Every little
wrong step can decrease your chances for recovery.

I came to this list many years ago with a recovery problem, and
have learned a lot since that time (as I also had no backup and
still got my files back). So I'll try to give some suggestions.
But remember that it depends on your _specific_ setting on what
will possibly work and what won't.

As it has been questioned already, formatting can have several
meanings, especially in Windows land. The more actual content
of the disk got overwritten, the worse it gets. So if you've stopped
the installation at some point, it's still possible that your data
can be located.

Allow me to introduce a few tools to you (my meanwhile famous list
of data recovery tools which I've posted some times on this list):

A worst-case tool to recover data (not file names, but file
content) is testdisk; in ports: sysutils/testdisk. It's also
on some diagnostics and recovery CDs like UBCD.

You can also try this:

fetch -rR device

Also recoverdisk could be useful.

The ports collection contains further programs that might be
worth investigating; just in case they haven't been mentioned
yet:

ddrescue
dd_rescue   - use this to make an image of the disk!
magicrescue
testdisk- restores content
recoverjpeg
foremost
photorec

Then also

ffs2recov
scan_ffs

should be mentioned. It's possible that the disk contains some
information to restore the initial UFS file systems.

And finally, the cure to everything is found in The Sleuth Kit:

fls
dls
ils
autopsy

In worst case. Just in worst case.

Keep in mind: Read the manpages before using the programs. It's
very important to do so. You need to know what you're dealing
with, or you'll probably fail. There is no magical tetroplyrodon
to click ^Z and get everything back. :-)

Proprietary (and expensive) tools like R-Studio or UFS Explorer
can still be considered worth a try. Their trial versions are for
free. UFS Explorer even works using wine (I've tried it).

If you have one or two spare disks, initialize them in FreeBSD.
Make a dd image of your disk to the 1st disk. Use recovery tools
to write their results on the 2nd disk. Disks are cheap, at least
your data is worth more than two disks. And you're going to need
the disks anyway. :-)

I wish you best luck to get your data back. I can fully understand
how frustrating such a situation can be.


-- 
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: Quota as a boot time module

2013-02-16 Thread Polytropon

On Sat, 16 Feb 2013 09:32:32 -0500, Fbsd8 wrote:
 I was reading the handbook quota section and it says quota has to be 
 compiles into the kernel. I thought it can also be loaded as a boot time 
 module?

If I remember correctly, quota is still one of the few things
you cannot load as a module, which implies you have to compile
it into your (custom) kernel. In /usr/src/sys/conf/NOTES there
is an entry for quota.


-- 
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-sending selected e-mail messages

2013-02-13 Thread Polytropon

I need a way to automatically re-sent stored e-mail messages
according to some criteria and like to ask for advice or
suggestions for an already existing solution before I start
reinventing the wheel. :-)

The messages in question are stored in MH format. This is a
tree where a mailbox equals a directory, and the individual
files in that directory equal the messages. They are numbered
1, 2, 3, 4, ..., and so on. Each message is in

Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

or

Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

or maybe even

Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

text format. Some of them might contain an attachment, which
is included in the file with something like

--Multipart=...
Content-Type: ...;
 name=...
Content-Disposition: attachment;
 filename=...
Content-Transfer-Encoding: base64
--Multipart=...---

Some messages are fully multipart.

So when iterating on ~/Mail/sent/1,2,3,4,5,... I get all the
messages. Each third line, To:, is the criteria to look at.
If it matches a given recipient, the mail should be sent again.
This can easily be done by the system's mailer which is properly
configured (and uses ISP's MX), so | mail -s maybe new subject
is possible. The message should already be properly pre-composed.

What is the easiest way to do this without reinventing the wheel,
or should I? :-)







-- 
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: Windowmaker build error, freebsd 9.1 inotify_init

2013-02-13 Thread Polytropon

On Wed, 13 Feb 2013 21:17:10 -0500, sean wrote:
 How exactly do I properly format the fix stated in the closed
 174105 bug report to the Windowmaker Makefile?
 
 
 Fix
 Adding -linotify to LD_FLAGS in the Makefile fixes the problem.
 

Just add the text -linotify to the LD_FLAGS variable defined
in the Makefile. It's /usr/ports/x11-wm/windowmaker/Makefile
in line 40. However, the name of the variable is LDFLAGS (no
underscore). I have v 1.171 of this file here, port version is
0.95.2 rev. 4 -- just for reference.

Original line:

LDFLAGS+=   -L${LOCALBASE}/lib

To be changed into:

LDFLAGS+=   -L${LOCALBASE}/lib -linotify

Then try to build WindowMaker again.


-- 
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: Chmod of /dev/ulptN

2013-02-12 Thread Polytropon

On Mon, 11 Feb 2013 18:33:12 -0800, Ronald F. Guilmette wrote:
 I exactly followed the directions here:
 
   
 http://lists.freebsd.org/pipermail/freebsd-questions/2012-February/238118.html
 
 Nontheless, my /dev/lpt0 node still only has permissions set to 0644.

In the subject you've mentioned /dev/ulpt (USB LPT port), and
now /dev/lpt (conventional LPT port), still the same applies
for both.



 Why?

Because the content of /dev is dynamically generated and maintained
as a virtual file system. This service is controled by devfs (devices
created at startup time) and devd (devices dynamically created during
runtime), both with their respective configuration files devfs.conf
and devfs.rules, and devd.conf. Those files need to be adjusted. To
make the change become active, restart the service.



 What did I do wrong?

You did not use the current proper method. :-)

Make the change you need in /etc/devfs.conf, e. g.

perm ulpt0 0664

if you need rw/rw/r. You can also use own user:group in
a similar way if you need to change file owner or group.
See man devfs.conf for details.




-- 
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 achieve E-Mail Notification on root login?

2013-02-12 Thread Polytropon

On Tue, 12 Feb 2013 13:24:52 +0100, Matthias Petermann wrote:
 
 Hello,
 
 given there is a FreeBSD system with users in the wheel group, what is  
 the best practise
 to send out a notification via E-Mail if one of them becomes root via  
 su? In an ideal
 case the E-Mail would contain the user name and the time.
 
 I thought about using sudo but this is not in the base system which I  
 would prefer.

I'm not sure if there already is a solution (provided in the
base system) that offers this functionality, but the fact of
a user having used su to su root is logged by the system.
The line is appended to /var/log/messages:

Feb 12 14:40:57 r56 su: poly to root on /dev/pts/2

The information you want is in there, and you could either use
the whole line, or apply some sed, awk or even perl to form a
message with less information (only date and user).

A scripted solution could monitor /var/log/messages for changes
and use the system's builtin mailer to deliver the message. Tools
like tail -f, grep and | mail could be involved. It should
be quite trivial to implement this and add a custom rc.d-style
script (or even few lines in ye olde /etc/rc.local).



-- 
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: Upgrade causes loss of all firefox settings (?)

2013-02-12 Thread Polytropon

On Tue, 12 Feb 2013 23:40:41 +0100, Ralf Mardorf wrote:
 I don't like the new style of Opera and the
 style of Chromium. I prefer classic menus.

Luckily you can change that in Opera and give it the
traditional style. Still Opera suffers from ongoing
disimprovement (e. g. reduction of functionality in
file save dialog or printing support).



 I'm not a flash fan, but I already couldn't download
 the last Internet provider bill :D, perhaps a flash or some other script
 issue.

Maybe Java? Some old fasioned guys still use it... ;-)

Even though I don't like Flash especially because it
has become what animated GIFs have been used for in the
past - replacement for HTML, nagging, stealing focus,
aggressive advertising with sound and so on - it perfectly
works with Opera.



 Btw. the design of Opera still is closer to my workflow than the
 design of Chromium.

That's my problem with Chromium and Firefox too. I like
the possibility to remove visual controls (red X button
on tabs for example) without losing functionality (middle
click on tab closes tab). From my very individual experience,
Opera offers the best integration of mouse and keyboard. It's
sufficiently fast and renders stuff acceptably correct.




-- 
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: which X driver for NVIDIA Quadro FX 570M?

2013-02-11 Thread Polytropon

On Tue, 12 Feb 2013 00:30:59 +1100 (EST), Ian Smith wrote:
 On Mon, 11 Feb 2013 10:41:31 GMT, Anton Shterenlikht wrote:
  From: paranormal paranor...@isgroup.com.ua
  Subject: Re: which X driver for NVIDIA Quadro FX 570M?
  To: freebsd-questions@freebsd.org
  Date: Wed, 06 Feb 2013 03:23:40 +0200
   
  I have t61p with mentioned card.
  x11/nvidia-driver works well for me (at least quake, doom, compiz work).
   
   Thanks for all the replies.
   
   I bought a T61p for 220 GBP - what bliss!
   
   BIOS update - no problem
   HEAD r246552 - no problem
   wireless with iwn0: Intel Wireless WiFi Link 4965 - no problem
   sound with hdac0: Intel 82801H HDA Controller - no problem
   CD-RW with cd0: HL-DT-ST DVDRAM GSA-U10N 1.05 Removable CD-ROM SCSI-0 
 device
 and sysutils/cdrtools-devel - no problem
   X with nvidia0: Quadro FX 570M and x11/nvidia-driver - no problem
   flash as per 
 http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/desktop-browsers.html
 (7.2.1.2 Firefox and Adobe Flash Plugin) - no problem
   
   In fact, no problems at all!
   
   I can't recommend it enough.
   
   Anton
 
 Suspend and resume?

I'd be interested in that, too (to possibly transform this knowledge
to the T60p I have).

And docking station support (if you have it)?




-- 
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: which X driver for NVIDIA Quadro FX 570M?

2013-02-11 Thread Polytropon

On Mon, 11 Feb 2013 14:28:30 GMT, Anton Shterenlikht wrote:
I guess no... However, I'm very ignorant of suspend/resume,
so not sure I'm doing the right thing.

For those who use the laptop in transportable mode (i. e.
not on the desktop as a desktop-PC substitute), those
features might be interesting in order to save power.

- the T61p manual details standby and hibernation modes.
Is this what you refer to by suspend?

Both are _different_ kinds, if I remember correctly.

Standby stores machine states in RAM and buffers it with
the battery power. This mode still requires power. This
is ACPI states S2 and S3.

Hibernate stores machine data somewhere on hard disk or
SSD. This mode does not require power. This is ACPI state
S4.

I can go into standby with Fn+F4, or with acpiconf -s 3
but can't seem to get back. The disk starts, but the
screen is corrupted, kind of black with very few white dots.

That seems to indicate that the GPU memory data is lost.
A typical problem with those sleep states.

- I've had a quick look at acpi(4) and apm(8).

APM is not in use anymore. ACPI has taken that functionality
at the point in time when APM has been brought to a fully
functional state.

I have:

hw.acpi.supported_sleep_state: S3 S4 S5
hw.acpi.s4bios: 0

From
http://static.usenix.org/event/usenix02/tech/freenix/full_papers/watanabe/watanabe_html/node6.html
you can see:

S3: sleep states. In these states, memory contexts are held
but CPU contexts are lost. The differences between S2 and
S3 are in CPU re-initialization done by firmware and device
re-initialization.

S4: a sleep state in which contexts are saved to disk. The
context will be restored upon the return to S0. This is
identical to soft-off for hardware. This state can be
implemented by either OS or firmware.

S5: the soft-off state. All activity will stop and all contexts
are lost.

- Anything I should check/test in BIOS?

The obvious things, but I assume the presets are already
fine for a laptop.

I see that power management is enabled in BIOS.
Is that enough?

It should be. Note that PM can also include things like
spinning down disks or reducing CPU power.

--
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: mount: /dev/da0p1: Invalid argument

On Fri, 8 Feb 2013 12:30:51 GMT, Anton Shterenlikht wrote:
 So what is the advice for transferring data
 via USB in such cases? Any other gpart partition
 I could use?

Use the most universal file system which isn't even a
file system: tar.

First, create a tar archive (not a _file_!) to the USB
media as if it was a tape. On the sparc machine:

# tar cvf /dev/da0 your stuff

You can add compression flags like z and j if you need.
To list what you've got, use tar t accordingly.

Then uncompress directly from the media on the non-sparc
machine:

# tar xvf /dev/da0

See man tar for other options you might want to add.
Also note that in this case, no file system is involved,
so you can't mount anything here.


 In the end I burned a CD with the files in question,
 but it's a bit of a waste, as I only need to
 move over several KB of data (wireless setup).

That's true. Don't you have any floppies at hand? ;-)

(Note: The tar approach also works on floppies, and even
across OS borders, e. g. between Solaris and Linux.)



-- 
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: mount: /dev/da0p1: Invalid argument

On Fri, 8 Feb 2013 08:07:12 -0500 (EST), Chris Hill wrote:
 On Fri, 8 Feb 2013, Anton Shterenlikht wrote:
 
   [ snip ]
 
  So what is the advice for transferring data
  via USB in such cases? Any other gpart partition
  I could use?
 
 I've always used FAT32 for thumb drives and the like. I don't know if 
 the SPARC would be able to use it, but FAT32 seems like it's most likely 
 to be usable by the largest number of different platforms.

While this may work for ordinary purposes, it can cause
trouble due to limitations of file name length, possibly
also characters in file names, and maximum file sizes. If
those factors are not triggered, it seems to be okay.



-- 
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 custom kernel on -current: unknown option COMPAT_LINUX

On Sun, 10 Feb 2013 00:18:06 GMT, Anton Shterenlikht wrote:
 This is on amd64 r246552
 
 I added
 
 options COMPAT_43
 options COMPAT_LINUX
 options COMPAT_LINUX32
 
 to the kernel config,
 following sys/amd64/conf/NOTES
 
 On buildkernel I get:
 
 unknown option COMPAT_LINUX
 
 What am I missing?

Do you also have those (from a working i386 system):

# Linux support
options COMPAT_LINUX# Enable Linux ABI emulation
options LINPROCFS   # Enable the linux-like proc 
filesystemsupport (requires COMPAT_LINUX and PSEUDOFS)
options LINSYSFS# Enable the linux-like sys filesystem 
support (requires COMPAT_LINUX and PSEUDOFS)
device  lindev
options COMPAT_AOUT # Enable i386 a.out binary support

(note PSEUDOFS is also needed)



-- 
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 custom kernel on -current: unknown option COMPAT_LINUX

On Sun, 10 Feb 2013 00:31:44 GMT, Anton Shterenlikht wrote:
   From free...@edvax.de Sun Feb 10 00:29:36 2013
 
   On Sun, 10 Feb 2013 00:18:06 GMT, Anton Shterenlikht wrote:
This is on amd64 r246552

I added

options COMPAT_43
options COMPAT_LINUX
options COMPAT_LINUX32

to the kernel config,
following sys/amd64/conf/NOTES

On buildkernel I get:

unknown option COMPAT_LINUX

What am I missing?
 
   Do you also have those (from a working i386 system):
 
   # Linux support
   options COMPAT_LINUX# Enable Linux ABI emulation
   options LINPROCFS   # Enable the linux-like proc 
 filesystemsupport (requires COMPAT_LINUX and PSEUDOFS)
   options LINSYSFS# Enable the linux-like sys 
 filesystem support (requires COMPAT_LINUX and PSEUDOFS)
   device  lindev
   options COMPAT_AOUT # Enable i386 a.out binary 
 support
 
   (note PSEUDOFS is also needed)
 
 No, I haven't added those.
 Are these necessary to have
 the linux binary compatibility?
 The handbook only mentions COMPAT_LINUX.

I think I had the same question some time ago and found
out that if those options are present, the Linux functionality
will build properly in the kernel. So I assume they are
required.



-- 
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: packages listing

On Sat, 9 Feb 2013 22:52:37 -0800 (PST), Dánielisz László wrote:
 Hi Everybody,
 
 Do you have any idea how can I list those installed packages
 that are not required by any other?

You can use sysutils/pkg_cutleaves to determine those.


-- 
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: Restricting Periodic Scripts

2013-02-06 Thread Polytropon

On Wed, 6 Feb 2013 09:26:17 -0800, Tim Gustafson wrote:
 I have a FreeBSD ZFS file server with tens of millions of files stored on it.
 
 But, the daily periodic scripts like
 /etc/periodic/security/110.neggrpperm and
 /etc/periodic/weekly/310.locate take hours iterating through those
 folders, and I just don't need them to be scanned.
 
 I see that I can edit /etc/locate.rc to fix the behavior for
 /etc/periodic/weekly/310.locate but I don't see a way to exclude
 folders from other scripts like /etc/periodic/security/110.neggrpperm
 from scanning them.  Is there any way to prune out folders that I
 don't want scanned, or should I just disable those jobs?

You can disable them per /etc/periodic.conf (see examples in
/etc/defaults/periodic.conf). To keep the functionality, but
restrict it to a smaller amount of files, you could use the
system's scripts as templates, make your own derivates
(wich inclusion or exclusion rules) and place them in
/usr/local/etc/periodic for the system to call them (which
it will if they are present). You can add your custom
configuration flags to /etc/periodic.conf and have your
scripts source that file (like the system's scripts do).




-- 
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 add unused space to an existing install

2013-02-06 Thread Polytropon

On Wed, 06 Feb 2013 11:58:56 -0600, Paul Schmehl wrote:
 When I try to create a new slice using fdisk, it doesn't seem to work.  If 
 I move to the label editor, I get this:
 
  FreeBSD Disklabel Editor
 
 Disk: mfid0 Partition name: mfid0s1 Free: 0 blocks (0MB)
 
 Part  Mount  Size Newfs   Part  Mount  Size Newfs
   -   -     -   -
 mfid0s1a  none   2000MB *
 mfid0s1d  none  65536MB *
 mfid0s1e  none   4096MB *
 mfid0s1b  swap65536MB SWAP
 mfid0s1f  none  10240MB *
 mfid0s1g  none601GB *
 
 As you can see mfid0s1g is 601GB, and according to fstab that's /var.
 
 Yet df -h shows:
 
 # df -h
 Filesystem   SizeUsed   Avail Capacity  Mounted on
 /dev/mfid0s1a1.9G726M1.0G41%/
 devfs1.0k1.0k  0B   100%/dev
 /dev/mfid0s1e3.9G 38M3.5G 1%/home
 /dev/mfid0s1d 62G6.6M 57G 0%/tmp
 /dev/mfid0s1f9.7G7.5G1.4G84%/usr
 /dev/mfid0s1g582G 39G496G 7%/var
 
 So apparently I'm not creating this new slice?  It should be /dev/mfid0s1h, 
 correct?

If you're creating a new slice, that would be mfid0s2, because
mfid0s1 is the 1st slice (DOS primary partition) carrying the
partitions a, swap, d, e, f, g. If I remember correctly, h is
the last partition letter that can be assigned, so this one
should be available.

Problem: The 1st slice mfid0s1 is already of fixed size, so
you cannot add a new partition here without growint that
slice first. Attention, that step isn't free of danger and
should be done with a backup at hand, just in case, and
because you _always_ need a backup. :-)

This problem is not a problem if you create a 2nd slice
mfid0s2 to use it separately.

If you can _really_ create mfid0s2 as a slice, you only need
to format it, e. g. newfs -U /dev/mfid0s2 which creates
mfid0s2c which in turn is called mfid0s2). You can then assign
that new partition (the one covering the whole slice) to
a new mountpoint, e. g. /data, /stuff or whatever you want.
There are also means to merge this partition into some
mountpoint that is already occupied, e. g. /home, maybe via
mount -o union including all possibly negative consequences).



 How to I recapture the remaining 2+TB of space that's not being used?

Without wiping the whole disk(s)?



-- 
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: backspace shows ^? in serial communications

2013-02-05 Thread Polytropon

On Wed, 6 Feb 2013 09:44:37 +0330, s m wrote:
 hi all
 
 i have a problem with backspace in serial communications. i have a
 freebsd8.2 box with a serial card on it. when i connect to other freebsd
 box via serial port backspace does not act as i expected. backspace shows
 ^? on screen. i searched alot and find out that stty has two parameters
 -erase and erase2- to identify erase characters in terminal and they should
 be set correctly. i set erase and erase2 to ^? by stty erase \^? and
 stty erase2 \^? commands but nothing happened.
 please let me know how i can fix it. i know it is simple issue but i really
 do not know how to do that.

If I remember correctly, ^? is delete, ^H is backspace. You
should check your terminal emulator if it outputs ^? instead
of ^H when you press the backspace key.

FreeBSD's default configuration handles keys correctly (if
you have the proper terminal emulation set, e. g. vt100 or
vt220 for your serial line), so there's probably something
wrong with the settings of the terminal program you're using.

For comparison:

% echo $TERM
xterm
% stty -a
speed 9600 baud; 24 rows; 80 columns;
lflags: icanon isig iexten echo echoe echok echoke -echonl echoctl
-echoprt -altwerase -noflsh -tostop -flusho -pendin -nokerninfo
-extproc
iflags: -istrip icrnl -inlcr -igncr ixon -ixoff -ixany -imaxbel -ignbrk
-brkint -inpck -ignpar -parmrk
oflags: opost onlcr -ocrnl tab3 -onocr -onlret
cflags: cread cs8 parenb -parodd hupcl -clocal -cstopb -crtscts -dsrflow
-dtrflow -mdmbuf
cchars: discard = ^O; dsusp = ^Y; eof = ^D; eol = undef;
eol2 = undef; erase = ^H; erase2 = ^H; intr = ^C; kill = ^U;
lnext = ^V; min = 1; quit = ^\; reprint = ^R; start = ^Q;
status = ^T; stop = ^S; susp = ^Z; time = 0; werase = ^W;

And:

% echo $TERM
cons25l1
% stty -a
speed 9600 baud; 25 rows; 80 columns;
lflags: icanon isig iexten echo echoe -echok echoke -echonl echoctl
-echoprt -altwerase -noflsh -tostop -flusho -pendin -nokerninfo
-extproc
iflags: -istrip icrnl -inlcr -igncr ixon -ixoff ixany imaxbel -ignbrk
brkint -inpck -ignpar -parmrk
oflags: opost onlcr -ocrnl tab0 -onocr -onlret
cflags: cread cs8 -parenb -parodd hupcl -clocal -cstopb -crtscts -dsrflow
-dtrflow -mdmbuf
cchars: discard = ^O; dsusp = ^Y; eof = ^D; eol = undef;
eol2 = undef; erase = ^H; erase2 = ^H; intr = ^C; kill = ^U;
lnext = ^V; min = 1; quit = ^\; reprint = ^R; start = ^Q;
status = ^T; stop = ^S; susp = ^Z; time = 0; werase = ^W;

If you want the system's C shell to treat ^? (delete) as
it should be treated (perform delete instead of backspace
or nothing), add those to your .cshrc:

bindkey ^? delete-char  # for console
bindkey ^[[3~ delete-char   # for xterm

Note that this only affects the C shell.



-- 
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: which pkg repository with 9.1

2013-02-04 Thread Polytropon

On Mon, 04 Feb 2013 21:04:40 -0600, Joseph A. Nagy, Jr wrote:
 I've almost always built from source since I switched to FreeBSD (I 
 sometimes, during the initial installation, used pkgs), it's longer but 
 more reliable.

I'm also doing this since I have sufficient CPU and RAM. :-)

However, for systems that are low on capacity, using precompiled
packages is a really comfortable way to initially install software.
The traditional pkg_add -r stuff was possible for most of the
available software with two exceptions:

1. software that needed compile-time options to make them work as
   intended (e. g. mplayer with mencoder and all codecs),

2. software that had no packages (e. g. german OpenOffice which had
   a pkg_add -r de-openoffice way in the past).

With pkgng and the new pkg command set, not just installing would
be possible (as known from pkg_add), but also updating (like with
freebsd-update, but for ports). At the moment, this functionality
is not provided, but it should become possible in the future,
obsoleting the traditional pkg_* tools, while the use of ports,
either with the bare make framework (make update, relying on
SVN instead of CVS, make install, make deinstall and so on) or
by the use of a port management tool (like portmaster) will of
course still be possible. I know even pkgng can't deal with the
two exceptions mentioned above, but it will add the binary updating
and therefor make system _and_ software updates easier, especially
when you're low on resources. It's also a welcome means if you
need to perform an offline installation, i. e. you don't have
Internet connection to obtain binary packages or sources, but
you can install from optical media instead.

The only problem I see (or which I hope not to see) is the
upcoming Linuxism of repositories. Plural: many of them. By
the use of the traditional pkg_* tools and the make framework
for ports, you don't have to deal with selecting repositories.
The correct files will be served. I hope there won't be a situation
in the future where arbitrary or contradicting repositories
free and non-free, vendor-provided, private, development,
different in priority and content, will be required to be chosen
by the user just to make basic things work (again).

For those who have ever tried to explain repositories to a
novice user in regards of a Linux distributions: You probably
know what I'm talking about. :-)




-- 
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: which X driver for NVIDIA Quadro FX 570M?

On Sun, 03 Feb 2013 08:53:49 -0700, Gary Aitken wrote:
 On 02/03/13 03:33, Anton Shterenlikht wrote:
  I'm considering buying this Thinkpad T61p laptop:
  http://www.ebay.co.uk/itm/321063105251
  
  It lists this graphics card:
  
  NVIDIA Quadro FX 570M
  
  I'm not sure what driver, if any, will support it.
  
  There seems to be the official Nvidia FreeBSD driver
  provided for it:
  
  http://www.nvidia.com/object/freebsd_100.14.09.html
  
  but this apparently has to be installed outside of
  the ports tree.
  
  xf86-video-nv drivers list several other Quadro cards,
  but not specifically 570M.
  
  Maybe somebody has used this card, so can give a
  definite reply?
 
 I can't specifically address that issue, but according to this page:
   http://www.freebsd.org/doc/en/articles/compiz-fusion/nvidia-setup.html

This is how I got my nVidia card working. It's worth mentioning
that the alternatives -- the nv driver (from X.org) and the
nouveau driver -- did work, but had poor 3D performance. The
proprietary driver by nVidia can be easily installed according
to the article you've mentioned.



 recent nvidia cards should be supported by the 
   x11/nvidia-driver port

This is currently the most advanced driver for that brand.
However, I had trouble with it when installing the 64 bit OS
and I went back to 32 bit, so I can't tell if this has been
improved.




-- 
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: which X driver for NVIDIA Quadro FX 570M?

On Sun, 03 Feb 2013 09:38:07 -0700, Gary Aitken wrote:
 On 02/03/13 08:53, Gary Aitken wrote:
  On 02/03/13 03:33, Anton Shterenlikht wrote:
  I'm considering buying this Thinkpad T61p laptop:
  http://www.ebay.co.uk/itm/321063105251
 
  It lists this graphics card:
 
  NVIDIA Quadro FX 570M
 
  I'm not sure what driver, if any, will support it.
 
  There seems to be the official Nvidia FreeBSD driver
  provided for it:
 
  http://www.nvidia.com/object/freebsd_100.14.09.html
 
  but this apparently has to be installed outside of
  the ports tree.
 
  xf86-video-nv drivers list several other Quadro cards,
  but not specifically 570M.
 
  Maybe somebody has used this card, so can give a
  definite reply?
  
  I can't specifically address that issue, but according to this page:
http://www.freebsd.org/doc/en/articles/compiz-fusion/nvidia-setup.html
  recent nvidia cards should be supported by the 
x11/nvidia-driver port
 
 Ugh, I just tried to build that and it is marked as IGNORE

Is the reason still requires fairly recent FreeBSD-STABLE, or
FreeBSD-CURRENT? See /usr/ports/x11/nvidia-driver/Makefile for
details. I've been successfully installing 270.41.19 in 8-STABLE
some time ago.



  The nvidia legacy driver page does not list that card, so it may be
  supported with the current x11/nvidia-driver port
http://www.nvidia.com/object/IO_32667.html
  
  I have a GT-610 and it is also not listed.
  I am in the process of upgrading to 9.1 as well, 
  but am fighting a bad SSD and have to reconfigure.
 
 The most recent nvidia port I can find is  
   NVIDIA-FreeBSD-x86_64-304.64.tar.gz 
 It builds fine, but on installing it craps out with the following message:
   /usr/X11R6/lib/modules/drivers  No such file or directory

Very strange.



 On 9.0, where I am using the 304.60 version, X11R6 is a pointer to /usr/local.

That's the default for some time now. Even FreeBSD v8 already
has that symlink, maybe it has already been part of v7 (not sure).

lrwxr-xr-x  1 root  wheel  10 2011-08-21 22:32:05 /usr/X11R6@ - /usr/local



 But my install of 9.1 has an actual /usr/X11R6, and that means things are
 scattered all over, some in /usr/local/lib/xorg and some in /usr/X11R6.

Where did you get the real X11R6/ directory entry from? Is this
some legacy of continuous updating the system beginning at a
version that actually had X11R6/ as a directory?



 This may be a glitch on my end; not sure yet. 

I think it is. The port expects something that actually is in
/usr/local, which would have been the same as in /usr/X11R6,
but in your case, it's not.



 I had to restart my ports
 build because I screwed up and didn't do a fetch before building x11/xorg,
 and that may have created some bad directories.

In worst case, maybe you can remove your ports and recreate the
reqired directory structures using the mtree templates
BSD.local.dist (and maybe BSD.x11.dist or BSD.x11-4.dist),
or maybe the xorg port will install the symlink automatically?
If you're not sure, create the symlink yourself and continue
as if nothing had happened. :-)


 Can anyone verify if /usr/X11R6 under 9.1 is supposed to be a symlink to
 /usr/local?

I stronly assume it is (no v9 OS at hand to check).



-- 
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: [kde-freebsd] tmux and konsole characters

On Sun, 03 Feb 2013 11:54:16 -0500, Andre Goree wrote:
 On a related note, I guess I can set LANG=en_US.UTF-8 in my .bashrc and
 have that as my default, no?

As far as I remember, login.conf is the file to set this, but
you can basically set environmental variables wherever you want.
For example, I have a system where I set them globally in the
C-Shell configuration, so all shells (even non-csh-shells like
bash) inherit the settings. Example from /etc/csh.cshrc:

setenv LC_ALL  en_US.UTF-8
setenv LC_MESSAGES en_US.UTF-8
setenv LC_COLLATE  de_DE.UTF-8
setenv LC_CTYPEde_DE.UTF-8
setenv LC_MONETARY de_DE.UTF-8
setenv LC_NUMERIC  de_DE.UTF-8
setenv LC_TIME de_DE.UTF-8
setenv LANGde_DE.UTF-8

Note that this creates a settings conglomerate from english
and german settings which looks stupid, but works (and is therefor
intended, or at least accepted as being established). :-)

Note the correct language prefix: en_US (as there's also en_GB).



 Any nuances on en_US.UTF-8 vs. ISO8859?

As soon as you have non-standard characters (like german umlauts)
encoded in ISO8859-1 (the default for this region) they won't show
up properly in UTF-8 (just as UTF-8 encoded umlauts will not show
up properly in a ISO8859-1 terminal session).



 I've never really ever needed to deal with locales before, but I believe
 UTF-8 offers more characters, no?

It offers many more, and especially if you're dealing with inter-
national documents, it seems to be the best way to go at the
moment. For example, I had to work on a document containing
german umlauts and chinese characters, so UTF-8 was the solution
to have all of them properly displayed and editable.





-- 
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: which X driver for NVIDIA Quadro FX 570M?

On Sun, 3 Feb 2013 11:29:30 -0700 (MST), Warren Block wrote:
 On Sun, 3 Feb 2013, Gary Aitken wrote:
 
  I suspect it is from building the nvidia driver,
  as those Makefiles use
   X11BASE=   /usr/X11R6
 
 Last I heard from the ports guys, that is obsolete.

If I remember correctly, X11BASE equals LOCALBASE, which
is /usr/local, so any Makefile which uses X11BASE will
be directed to /usr/local. It's not clear to me why a
Makefile defines a location that has been obsolated...





-- 
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: svn for 9.1

On Sun, 03 Feb 2013 17:26:52 -0500, Fbsd8 wrote:
 What is the category / port name to install svn?

It's in devel/subversion.



-- 
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: Opera

2013-02-02 Thread Polytropon

On Fri, 01 Feb 2013 19:42:19 -0600, ajtiM wrote:
 Hi!
 
 I use Opera 12.12. I use KDE 4 and Fluxbox.
 In operaprefs.ini I have
 [File Selector]
 Dialog Toolkit=4
 
 which help me that Opera works othervise I get:
 libpng error: incorrect data check
 libpng error: incorrect data check
 Segmentation fault (core dumped)
 
 This happened on KDE but without above lines in operaprefs.ini, Opera works 
 without problem on Fluxbox.
 
 What is different, please?

I don't see an obvious difference except that something
seems to be wrong with your PNG library. Is everything
in sync? Or is some mechanism relying on libpng malfunctioning
somewhere else?

Regarding Opera: For better integration with desktop
environments newer versions can emulate the DE's native
dialogs instead of using its own one (which was superior
in functionality).

If you enter about:config in your address bar and search
for toolkit, you will find the corresponding setting with
the following explanation:

File dialog toolkit 
  0 = Autodetect toolkit to use for file selector
  1 = Use Qt for file selector (deprecated, will fall back to KDE)
  2 = Use GTK for file selector
  3 = Use KDE for file selector
  4 = Use X11 for file selector

Maybe this causes trouble in your specific non-mainstream
setting? However I can't imagine how...



-- 
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: Opera

2013-02-02 Thread Polytropon

On Sat, 02 Feb 2013 05:35:46 -0600, ajtiM wrote:
 Looks like Opera has problem autodetect on KDE but not in Fluxbox.

This is because Fluxbox is neither KDE or Gnome. :-)

But simply coredumping is a bad default value for how to act
when not using KDE or Gnome. I'm using Opera on WindowMaker
here for many years without that specific error (11.50 at the
moment), so maybe another sign for disimproved software... :-(


-- 
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 fix a broken owner for files from world build from ports?

2013-01-29 Thread Polytropon

On Tue, 29 Jan 2013 03:54:55 +0100, Ralf Mardorf wrote:
 On Tue, 29 Jan 2013 03:41:34 +0100, Joshua Isom jri...@gmail.com wrote:
  On 1/28/2013 7:56 PM, Ralf Mardorf wrote:
  Still not perfect, I guess I need something similar to ls -RAl for some
  directories :S and I didn't test what awk will do with names including a
  space.
 
  Try `find /dir -ls`.  You can pipe it into sed like this `find /dir -ls|  
  sed -e 's%/dir%%g'` and then get something easily comparable.
 
 Cool, it does display the path, but there's still the other issue:
 
 $ touch test\ test
 $ find * -ls| sed -e 's%/dir%%g'| awk '{print $5 $11}'
 rocketmouse test
 
 Perhaps awk isn't that important, but it e.g. will filter different file  
 sizes, for e.g. configurations I edited in the meantime.

A thing regarding awk: For extended formatting, use the
printf() command which works the same as in sh and C, os
if you need, you can do things like

printf %s '%s', $1, $2;

Also note that you can have a custom delimiter for parsing
the input, e. g. -F : (if you would generate input lists
in :-separated CSV format).

Additionally, it seems you're running into the fun of spaces
in file names. Even though you can put them there, it doesn't
imply it's good to do it. Spaces are separators (for commands
and options), and everytime they're _not_ (e. g. when they
appear in file names), you need to care for this fact, by
escaping or quoting them.

Maybe those articles by David A. Wheeler are interesting
to you to learn about this annoyance for people writing
short shell scripts to automate tasks:

Filenames and Pathnames in Shell: How to do it correctly

http://www.dwheeler.com/essays/filenames-in-shell.html

Fixing Unix/Linux/POSIX Filenames:
Control Characters (such as Newline), Leading Dashes, and Other Problems

http://www.dwheeler.com/essays/fixing-unix-linux-filenames.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: How to fix a broken owner for files from world build from ports?

2013-01-29 Thread Polytropon

On Tue, 29 Jan 2013 12:23:09 +0100, Ralf Mardorf wrote:
 On Tue, 2013-01-29 at 10:08 +0100, Matthias Apitz wrote:
  This is a mayor damage and can only be repaired by a new installation.
 
 Perhaps true, but if such a simple mistake can't be fixed, [...]

Excuse me, it's not a _simple_ mistake. It may have initially
been even a typo, but anything executed with root privileges
is not simple; root has the power to do anything, even to
completely destroy the system, and that can also be as simple
as calling rm or dd with carefully carelessly crafted options,
and there is no simple fix for this.



 [...] what happens
 when somebody makes a big mistake?

The size of the mistake doesn't even matter. :-)



 Perhaps more people stay with Linux
 than other *NIX, regarding to the policy, that issues should be fixed
 instead of always starting from the beginning. ;)?

The fix to your issue is, in pseudocode:

for part in ( OS , ports ) do:
determine owner rocketmouse:* for all files
compare with list with correct owner
for each deviating file do:
if owner != correct owner then:
chown file to correct user
fi
od
od

Of course OS and ports have to be treated seperately.

As you have mentioned to own a backup where the permissions
(owners) are correct, obtaining the required reference data
from that backup would be the easiest part.

The alternative: reinstall world, reinstall ports. To avoid
this task, you need to activate your admin skills. :-)



 Of course, if I simply would restore from a dump, it will be less time
 consuming and it wouldn't annoy you, but I would have the bad feeling,
 that if ever needed, thinks can't be fixed, I always would have to
 restore from backups. And what happens, if for what reason ever a backup
 shouldn't be available?

In that case, you would need other references to get the correct
file owners. Files are usually installed to the system by the
install command, and it is employed in the Makefiles for the
OS and also for ports. As you correctly recognized, not simply
all files belong to root, so everything non-standard could
be derived from such control files.

Of course, the more files you have to treat (see wc -l of your
result list), the harder the task can become, and maybe installing
the port again is faster than finding out where permissions are
set for the install program call. If you only have 10 files or
so, do it manually, but if there are 100 and more files, coming
from several different ports, reinstalling them sounds easier,
and it's not a big deal to do that with portmaster.



-- 
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: Viewing processes hierarchically

2013-01-29 Thread Polytropon

On Tue, 29 Jan 2013 13:52:47 -0800, Patrick wrote:
 Is there any way in FreeBSD to view all running processes hierarchically,
 like Activity Monitor in Mac OS X can do?
 
 e.g.
 http://f.cl.ly/items/37310J17273X3F1E1l0G/Image%202013.01.29%2013:50:36%20.png
 
 I believe I have a masked process spawned from an Apache process, but I'm
 having a hard time tracking it down.

You can do this with htop, then press PF5. Or use pstree.
Both are in 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: Starting with ZFS on fresh install

On Mon, 28 Jan 2013 07:55:06 -0500, Carmel wrote:
 I have a spare amd64 PC that I want to install FreeBSD 9.x on. I want
 it to utilize ZFS right from the start. There are two HD's in the PC.
 One will handle the /var partition and the other everything else. The
 last FBSD installer I used was on the 7.x branch. Does the new
 installer in the 9.x branch handle that automatically?

If you tell it to? :-)

No, honestly: The PC-BSD installer can be used to install
a normal FreeBSD system _and_ take care of ZFS initialisation.



 Would it be advantageous to run something like FreeDOS to remove the
 existing partition information, etcetera and then reformat the dives?
 They are currently formatted in Microsoft's NTFS format.

That's not needed. The installers (of FreeBSD and of PC-BSD)
can remove existing partitions in an early step.



 Also, for the 9.1 branch, I read that I have to run:
 
 # /usr/sbin/pkg
 To bootstrap the system. I also need to place:
 WITH_PKGNG=   yes
 in the /etc/rc.conf file.

The above option doesn't look as if it belongs to /etc/rc.conf.
Instead it would probably go to /etc/make.conf.





-- 
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: Sharing a mail folder between Linux and FreeBSD

On Mon, 28 Jan 2013 10:51:21 +0700, Erich Dollansky wrote:
 Hi,
 
 On Mon, 28 Jan 2013 01:36:36 +0100
 Ralf Mardorf ralf.mard...@rocketmail.com wrote:
 
  On Sun, 2013-01-27 at 13:58 +0100, Polytropon wrote:
   #  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)
  
  
  What source tree? I only checked out the kernel source using svn and
 
 yeah, what source tree? It seems that you do not have one.

It's rather untypical to check out _only_ kernel sources without
the top level content. The tree will probably be complete, but
if it hasn't been used yet to create world (and kernel), the
/usr/obj cache will be empty, so no quick re-installation
for modified binaries (modified in permissions, not in context).



  # freebsd-update -r 9.1-RELEASE upgrade
 
 This is very much a binary upgrade. You might have a source tree for
 8.3 which is not very helpful now.

Except freebsd-update also updates the src/ subtree, usually if
the default Components src world kernel is kept. So it will
probably be the corresponding RELEASE tree.



 I do not know if this program is able to fix your problem.

Probably not, except by a binary update to the same version
which can be considered a re-install.



  I wanted to run it tonight, but since I don't know where my source
  tree is, I can't continue.
 
 I think that you simply do not have one. At least not a current one.
 Read the handbook how you can get the source tree and then download and
 compile it.
 
 I believe that all other options will end in a re-installation.

If this is also to be considered for installed ports, that will
probably be the easiest solution. On the other hand, if the
amount of work is justified, tracking down the individual
defective permissions and manually fixing them could be an
option not to go that way, as it will definitely take some
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: Usage of restore

On Mon, 28 Jan 2013 15:53:25 +0100, Ralf Mardorf wrote:
 Hi :)
 
 I don't understand how to use the restore command.

The answer is in man restore. :-)



 root@freebsd:/mnt/dump # restore -v -t  
 dump-9.1-RELEASE-20130123_193142-usr_f.dump
 Verify tape and initialize maps
 /dev/sa0: No such file or directory

Correct. The program defaults to a tape drive /dev/sa0 to read
dumps from. You need to specify -f file if you want to read
from a file instead.

Also: Note that restore will usually restore the files in the
current (!) directory, so you need to cd to where you want to
extract the dump to.



 root@freebsd:/mnt/dump # restore -v -t -f  
 dump-9.1-RELEASE-20130123_193142-usr_f.dump
 Verify tape and initialize maps
 Tape block size is 32
 Tape is not a dump tape

Did you create the dump in some non-default format?



 FWIW the dump files are on a ext3 fs. I know that I don't need to backup  
 /tmp.

It shouldn't matter what filesystem the dump files are stored
on. It's just important they are in the correct format.



 The backups were done by this script, perhaps I've done something wrong:
 
 root@freebsd:/mnt/dump # cat /root/dump.sh
 #! /bin/bash

Ern... two things: Do you _really_ have /bin/bash on FreeBSD?
I know this is possible.

And do you use any bash-specific features in your script? If
not, why not use /bin/sh, the universally accepted standard? :-)



 # bash dump.sh
 
 dumpstart=$(date +%Y%m%d_%H%M%S)
 dump_path=/mnt/dump/dump-$(uname -r)-$dumpstart
 rootdir_a=/dev/ad4s1a
 usr_dir_f=/dev/ad4s1f
 var_dir_d=/dev/ad4s1d
 tmp_dir_e=/dev/ad4s1e
 
 #mkdir $dump_path
 dump -0Launf - $rootdir_a | bzip2  $dump_path-roota.dump
 dump -0Launf - $usr_dir_f | bzip2  $dump_path-usr_f.dump
 dump -0Launf - $var_dir_d | bzip2  $dump_path-var_d.dump
 dump -0Launf - $tmp_dir_e | bzip2  $dump_path-tmp_e.dump
 echo Started: $dumpstart
 echoDone: $(date +%Y%m%d_%H%M%S)
 
 exit 0

Looks correct, but read carefully:

 dump -0Launf - $usr_dir_f | bzip2  $dump_path-usr_f.dump

Now compare again:

restore -v -t -f dump-9.1-RELEASE-20130123_193142-usr_f.dump

Do you notice bzip is involved in the creation step, but not
in the reading step?

You need to apply uncompression first so that restore will
have data in .dump format (instead of .dump.bz2 which it
obviously cannot understand). Verify:

file dump-9.1-RELEASE-20130123_193142-usr_f.dump

So using proper file extensions can prevent confusion. :-)



-- 
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: Sharing a mail folder between Linux and FreeBSD

On Mon, 28 Jan 2013 15:58:05 +0100, Ralf Mardorf wrote:
 On Mon, 28 Jan 2013 15:37:25 +0100, Polytropon free...@edvax.de wrote:
  It's rather untypical to check out _only_ kernel sources without
  the top level content.
 
 For the update from 8.x to 9.1 I even didn't check out the kernel source,  
 this is something I did much later ;). I simply followed the FreeBSD  
 instructions. Perhaps it were binaries, but anyway, IIRC a kernel was  
 compiled?!?

That would require kernel sources (and some of the top level
files). Probably only a binary kernel update has been performed
because that's the purpose of freebsd-update (versus installing
from source).


-- 
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: Usage of restore

On Mon, 28 Jan 2013 17:28:10 +0100, Ralf Mardorf wrote:
 On Mon, 28 Jan 2013 16:54:29 +0100, Polytropon free...@edvax.de wrote:
  The answer is in man restore. :-)
 
 No it isn't ;). I did read it.

Eáu còntrair! :-)

 -f file
 Write the backup to file; file may be a special device file like
 /dev/sa0 (a tape drive), /dev/fd1 (a floppy disk drive), an ordi-
 nary file, or `-' (the standard output).  Multiple file names may
 be given as a single argument separated by commas.  Each file
 will be used for one dump volume in the order listed; if the dump
 requires more volumes than the number of names given, the last
 file name will used for all remaining volumes after prompting for
 media changes.  If the name of the file is of the form
 ``host:file'', or ``user@host:file'', dump writes to the named
 file on the remote host using rmt(8).  The default path name of
 the remote rmt(8) program is /etc/rmt; this can be overridden by
 the environment variable RMT.

You could even set $TAPE to the file name and omit -f. :-)



 This was a Wald'n'Bäume situation.

That's possible.



 Even if I would have add a .bz2, I
 would have missed it, since on Linux I .tar.bz backups and it's more
 automated to extract a .tar.foo.

But .tar.bz != .dump and != dump.bz (different programs: tar and
dump are working differently).



 However, I should add .bz2 in the future.

At least this is a comfortable way to avoid confusion and know
file content by simply looking at its name.



  Ern... two things: Do you _really_ have /bin/bash on FreeBSD?
  I know this is possible.
 
  And do you use any bash-specific features in your script? If
  not, why not use /bin/sh, the universally accepted standard? :-)
 
 No /bin/bash,
 
 # ls /usr/local/bin/bash
 /usr/local/bin/bash
 
 I run bash file instead of sh file, IIRC I already had an issue when  
 writing a script and running sh file.

Okay, that is a fully valid solution, in that case #!/bin/bash
is ignored.

But just think about a typical worst case scenario: You
have a script that requires bash, and you need to apply it
it SUM with /usr unmounted, and you accidentally do not have
a statically linked /bin/bash. Wouldn't it be much more
convenient to rely on the /bin/sh default scripting shell?

Maybe if you review your script, you can find out what
bash-ism is employed and turn it into a valid sh equivalent
while keeping the intended functionality. At a first glance,
I don't see anything sh can't do.



 I use it, since I use it on Linux too, another shell might cause issues,  
 if I continue writing this script or if I should write another script.  

Interoperability for scripting is hard in regards of Linux,
where sh isn't bash, and sometimes bash isn't bash either. :-)



 I thought /bin/sh is a link to another shell.

Sometimes it is. Sometimes not.

-r-xr-xr-x  1 root  wheel  115348 2011-08-21 20:23:20 /bin/sh*

On FreeBSD, /bin/sh is technically ash (This code is derived
from software contributed to Berkeley by Kenneth Almquist.,
see /usr/src/bin/sh/main.c), a replacement for the traditional
Bourne shell whose name it inherits. It's a real binary, not
a link.




-- 
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 fix a broken owner for files from world build from ports?

On Mon, 28 Jan 2013 18:18:05 +0100, Ralf Mardorf wrote:
 I still have no idea how to check this for the files build from ports.

Are there _many_ on the list (rocketmouse:* in /usr/local)?

If not: A simple reinstallation of that port would be sufficient,
except you can easily spot the installation permissions from
the port's Makefile (in that case: do it manually).

The find | grep solution you're already using is sufficient
for checking. For correcting... it depends. In ultra-worst case,
re-install _all_ ports (portmaster -af plus options to avoid
undesired interactivity).


-- 
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: sh export

On Mon, 28 Jan 2013 18:55:02 -0500, Fbsd8 wrote:
 I'm reading a script and i see a lot of exports.
 
 Is there some command to display the exported environment?

Yes, sh's builtin env does this.



 The env command does not show them. Only see things made by setenv command.

It seems you're mixing shells here. The C Shell uses setenv
to set variables, printenv to list them. The systems's sh
uses export to set variables, and env to print them.

Example (with exported and non-exported variable:

$ export ASDF=yxcvbnm
$ env | grep ASDF
ASDF=yxcvbnm

$ JKL=qwertzuiop
$ env | grep JKL
$ echo $JKL
qwertzuiop

And compare for the C shell:

% setenv ASDF yxcvbnm
% printenv | grep ASDF
ASDF=yxcvbnm

% set JKL=qwertzuiop
% printenv | grep JKL
% echo $JKL
qwertzuiop

If you omit the | grep step, the full list will be printed.



-- 
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: Sharing a mail folder between Linux and FreeBSD

On Sun, 27 Jan 2013 15:38:38 +0700, Erich Dollansky wrote:
 Hi,
 
 On Sun, 27 Jan 2013 09:15:09 +0100
 Ralf Mardorf ralf.mard...@rocketmail.com wrote:
 
  Good morning,
 
 ood morning? The sun is settling soon!

The sun of the planet of the ood?
Or the former Sun of one of the microsystems? :-)



  if I run 'make deinstall reinstall' for a port, it doesn't ask a
  single question, at least not for dbus.
  
  # ls -l /usr/local/bin/dbus-daemon
  -r-xr-xr-x  1 rocketmouse  wheel  377744 Jan 18 22:44  
  /usr/local/bin/dbus-daemon
  
  # cd /usr/ports/devel/dbus ; make deinstall reinstall
  
  # ls -l /usr/local/bin/dbus-daemon
  -r-xr-xr-x  1 root  wheel  377744 Jan 27
  08:55 /usr/local/bin/dbus-daemon
  
  Is there a command to deinstall and reinstall all ports or an idea
  for a script to do it?
  
 Check portupgrade or one of other utilities to handle ports. There is
 one option to force an upgrade even if it would be a downgrade.

With tools like portmaster, this task can easily be automated.
If you only will have to handle a few ports, using the bare
ports method (make) is probably the easiest way (in case
everything else stays definitely consistent).



  Do I have to reboot into single user mode and then to run make  
  installworld only to reinstall world?
 
 No, you just run it as root. It should work afterword except for
 currently running programs.

The comment header of /usr/src/Makefile suggests installing the
world in single user mode (steps 5 - 11).

#  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)

This should be the safest method.


-- 
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: [OT-ish] Need a Binary for lang/sml-nj

On Sun, 27 Jan 2013 16:40:51 -0600, Joseph A. Nagy, Jr wrote:
 At this point I'm not sure what the problem is, though I do appreciate 
 the help.
 
 root@alex-laptop:/root # pkg add 
 http://people.freebsd.org/~olgeni/smlnj-110.0.7_3.txz
 smlnj-110.0.7_3.txz 100% 2586KB 287.3KB/s 285.9KB/s 
 00:09
 Installing smlnj-110.0.7_3...pkg: wrong architecture: freebsd:9:x86:32 
 instead of freebsd:9:x86:64
 
 Failed to install the following 1 package(s): 
 http://people.freebsd.org/~olgeni/smlnj-110.0.7_3.txz

Did you have any success using the old-fashioned pkg_add method
with the -f option, and using the FreeBSD 9-STABLE precompiled
package (should work for 9.1-RELEASE too)?

# pkg_add -f 
ftp://ftp.freebsd.org/pub/FreeBSD/ports/i386/packages-9-stable/Latest/smlnj.tbz

Again, check /usr/ports/lang/sml-nj/Makefile for runtime dependencies
you might need to add, and in worst case use libmap (the library mapper)
to make them work for this program.

Also see /usr/ports/lang/sml-nj/pkg-descr if you can use a different
ML interpreter as suggested in the description. ;-)





-- 
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: [OT-ish] Need a Binary for lang/sml-nj

On Sun, 27 Jan 2013 16:53:05 -0600, Joseph A. Nagy, Jr wrote:
 On 01/27/13 16:49, Polytropon wrote:
  Did you have any success using the old-fashioned pkg_add method
  with the -f option, and using the FreeBSD 9-STABLE precompiled
  package (should work for 9.1-RELEASE too)?
 snip
 
 No, because I have pkgng since I installed 9.1 release (first update I 
 did) back in December when the 9.1 image was available on the ftp 
 server. pkg_add is no longer a valid command.

Okay, I didn't know that, as I'm still on an older system
here, so I've not advanced enough to use the new pkg command
on a daily basis. :-)

I assume getting the sources for pkg_add from a 9.0 system
and building it, then using it to forcedly install the
available package is going to break something...

However, does any other available (S)ML implementation (nml,
polyml, moscow_ml) fit 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: Sharing a mail folder between Linux and FreeBSD

On Mon, 28 Jan 2013 01:36:36 +0100, Ralf Mardorf wrote:
 On Sun, 2013-01-27 at 13:58 +0100, Polytropon wrote:
  #  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)
 
 
 What source tree? I only checked out the kernel source using svn and
 IIRC it's using the /usr/src directory (I'm booted into Linux at the
 moment), without a subdirectory /kernel. I can delete the kernel source,
 since it's IMO fishy to have headers of another revision, than the
 kernel is, but when I asked, I got a reply, that it should be ok for
 FreeBSD. However, I never used the kernel source.

The content of /usr/src does not only contain the kernel. It's
the whole OS, except of course you have only installed selected
parts of this tree. The file I've mentioned is at the top of
this structure: /usr/src/Makefile contains a short instruction
of how to install kernel and world (and explains other possible
targets).



 When I updated I did it like that (without subversion or cvs):
 
 # cd /usr/ports/misc/mc  make install clean
 # uname -r
 8.3-RELEASE
 # freebsd-update -r 9.1-RELEASE upgrade
 # freebsd-update install
 # shutdown -r now
 
 # freebsd-update install
 # cd /usr/ports/ports-mgmt/portupgrade  make install clean
 # /usr/local/sbin/portupgrade -f ruby
 # rm /var/db/pkg/pkgdb.db
 # /usr/local/sbin/portupgrade -f ruby18-bdb
 # rm /var/db/pkg/pkgdb.db /usr/ports/INDEX-*.db
 # /usr/local/sbin/portupgrade -af
 # freebsd-update install
 # shutdown -r now
 
 # freebsd-update IDS  outfile.ids
 
 I wanted to run it tonight, but since I don't know where my source tree
 is, I can't continue.

That's the binary way of updating. If you'd update from source,
the steps would usually involve first updating /usr/src (by
whatever means, CVS no more, SVN or as part of a binary update
that also keeps the OS sources current). To take this approach,
the sources have to be complete. You can follow a -STABLE and
even -CURRENT (-HEAD) branch if you like.

My suggestion would have been: If you have already used this
method before, and maybe if your current system has been installed
that way, you can do it again; if /usr/obj (the result tree
for building world and kernel) is still present, only the
make installworld steps would have been involved; even better,
if you only have to deal with a few system components, a selective
make install would have been sufficient.

However, it has already been suggested to utilize mtree, because
a real re-installation isn't actually needed (as no files have
been changed, only their permissions, and that can be checked
and corrected using the /etc/mtree reference files).




-- 
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: Sharing a mail folder between Linux and FreeBSD

On Sat, 26 Jan 2013 10:01:18 +0100, Ralf Mardorf wrote:
 On Sat, 2013-01-26 at 08:48 +0100, Polytropon wrote:
  On Fri, 25 Jan 2013 16:15:28 -0600, Joshua Isom wrote:
   As for the listings in /usr/local 
   they'll need fixed.  On my system, almost everything's owned by root. 
  
  There are a few exceptions when files are owned by a daemon.
  As I said, re-installing those parts (or even world) should
  fix this, but maybe it's possible to apply some mtree magic
  to fix the owner to the proper one (root in most cases).
 
 Rebuilding world only shouldn't take that long.

If you still have the /usr/obj subtree where you installed
world from last time, you only need to make installworld
(as explained in /usr/src/Makefile's comment header).

There has also been a very good advice on how to use mtree
to do this (as the files don't need re-installation per se,
because they haven't changed).



   The man directories are owned by man, and 
   /usr/local/libexec/polkit-set-default-helper is set as polkit:polkit.
  
  That's a good example for the non-root exceptions; there might
  be others.
 
 There are others on my system, so I can't simply run chown -R :(.

That's correct. If you can spot those irregularities in
/usr/local, it seems to be the safest way to re-install
the ports those files belong to.





-- 
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: Sharing a mail folder between Linux and FreeBSD

On Sat, 26 Jan 2013 10:11:37 +0100, Ralf Mardorf wrote:
 On Sat, 2013-01-26 at 08:49 +0100, Polytropon wrote:
  On Fri, 25 Jan 2013 23:22:29 +0100, Ralf Mardorf wrote:
   On Fri, 25 Jan 2013 22:51:55 +0100, Joshua Isom jri...@gmail.com wrote:
Sorry if my original command ended up breaking your system
   
   Don't worry, I run dump to backup it, but I'll try to fix it without  
   restoring it from the backup.
  
  Maybe you can read the original owners from that backup and
  just _change them_ accordingly? As the files haven't been
  altered, there would be no need to rewrite them entirely.
 
 I used dump and can't find how to extract something from the dump files.
 If I would restore from a dump I would lose something, since the dump is
 some days old and I worked on my FreeBSD.

Maybe something along restore -t is possible, as you only
would want to extract owner information and nothing more.
That information could be used to chown the files which need
that change.




 If it would be possible to write a script that does rebuild everything,
 with the same configs and _without the need of user interaction_ I would
 rebuild and if needed update (at least Chromium) everything. But last
 time there were still yes/no (should I delete this file) questions all
 the times.

Maybe you can do a separate check for /usr/local (installed
ports) and everything else (and both excluding /home).
The one is cured by the mtree magic (or make installworld
in worst case), the other by reinstalling ports that need
it (or, if only few files are affected, to check the install
directives in the Makefile and do it manually).




-- 
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: [OT-ish] Need a Binary for lang/sml-nj

On Sat, 26 Jan 2013 12:23:42 -0600, Joseph A. Nagy, Jr wrote:
 I've got an amd64 machine (and no spare) and need help getting a binary 
 for lang/sml-nj (it won't compile on amd64, it's marked ignore for 
 whatever reason). I can send someone a 'pciconf -lv' of my machine if it 
 would help.

1. Have you tried compiling anyway? :-)

2. Maybe you can install the i386 precompiled package for
   that software with $PACKAGESITE set accordingly, and
   it will probably run on amd64.

3. Check /usr/ports/lang/sml-nj/pkg-descr for inspiration
   about alternatives. :-)



-- 
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: [OT-ish] Need a Binary for lang/sml-nj

On Sat, 26 Jan 2013 14:50:34 -0600, Joseph A. Nagy, Jr wrote:
 On 01/26/13 13:44, Polytropon wrote:
  On Sat, 26 Jan 2013 12:23:42 -0600, Joseph A. Nagy, Jr wrote:
  I've got an amd64 machine (and no spare) and need help getting a binary
  for lang/sml-nj (it won't compile on amd64, it's marked ignore for
  whatever reason). I can send someone a 'pciconf -lv' of my machine if it
  would help.
 
  1. Have you tried compiling anyway? :-)
 
 no, I've learned my lesson with trying to compile software when it's 
 marked ignore.

You can keep them in a jail. :-)



  2. Maybe you can install the i386 precompiled package for
  that software with $PACKAGESITE set accordingly, and
  it will probably run on amd64.
 
 The official pkg repo's don't exist at the moment.

It looks like

ftp://ftp.freebsd.org/pub/FreeBSD/ports/i386/packages-9-stable/Latest

has smlnj.tbz (for use with pkg_add).

But be sure to check /usr/ports/lang/sml-nj/Makefile for
possible runtime dependencies. In worst case, use libmap.





-- 
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: Sharing a mail folder between Linux and FreeBSD

On Fri, 25 Jan 2013 13:05:51 +0100, Ralf Mardorf wrote:
 The user can't become root using Xfce Terminal Emulator or by ttyv1 (Ctrl  
 + Alt + F2). This was possible before I switched the uid.
 
 Before the switch PPPoE was enabled automatically, now I have to do it  
 manually.
 
 $ su
 su: not running setuid
 
 $ ls -l `which su`
 -r-sr-xr-x  1 rocketmouse  wheel  16880 Dec 23 18:38 /usr/bin/su

Erm... that looks horribly wrong.

The permissions indicate that setuid is set, but the file
owner is wrong. For comparison:

-r-sr-xr-x  1 root  wheel  14604 2011-08-21 20:24:28 /usr/bin/su*

This program has to belong to root. It seems that your
attempt to reflect UID changes in the file permissions
exceeded the scope of this task: Programs of the OS
seem to be affected, which is definitely not good.



 $ ls -l /home/ | grep rocketmouse
 drwxr-xr-x  28 rocketmouse  rocketmouse 1536 Jan 25 12:17 rocketmouse

You can use ls -ld to omit the grep step. :-)



 $ id
 uid=1000(rocketmouse) gid=1000(rocketmouse)  
 groups=1000(rocketmouse),0(wheel)

Seems to be okay.



 Ctrl + Alt + F2 
 '# ppp -ddial alice' does work
 '# find / -uid 1001 -exec chown 1000 '{}' \;' no messages
 '# find / -gid 1001 -exec chown :1000 '{}' \;' no messages
  Ctrl + Alt + F9

I think you can now spot a possible mistake for the file owner
change I mentioned above: Only files inside /home should have
been in the initial scope, but somehow -uid 1001 has been
avaluated true for /usr/bin/su, even though I cannot imagine
what should have caused this.

Do you have other files in /usr or even /usr/local that do
belong to rocketmouse (uid == 1000 or 1001) now? That should
not have happened...



 Without success I again read some important messages of this thread in the  
 archive and googled regarding to the suid issue.

Some programs check by whom they are called or who they
belong to; if that's != root when it is _supposed_ to
be root, that can cause problems, especially when it's
not a simple x (execute), but s (setuid) program like
an X display manager.



 Any hints are welcome!

Check for defective permissions. In worst case, update
your system from source or binary to fix permissions.
Maybe there's also an mtree trick to do it.



-- 
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: Sharing a mail folder between Linux and FreeBSD

On Fri, 25 Jan 2013 13:39:07 +0100, Ralf Mardorf wrote:
 Hi Erich :)
 
 On Fri, 25 Jan 2013 13:25:07 +0100, Erich Dollansky  
 erichsfreebsdl...@alogt.com wrote:
  What happens on a normal TTY?
 
  Ctrl + Alt + F2 
 
  So, you can switch to them. Can you try a su here?
 
 Ctrl + Alt + F* will open a ttyv*
 
 I can log in as root, but if I log in as user, I can't run su successfully.

Because as you quoted, the su binary doesn't have the UID 0,
this means it's not owned by root anymore which may have bad
influence on its runtime behaviour. :-)

You have:

-r-sr-xr-x  1 rocketmouse  wheel  16880 Dec 23 18:38 /usr/bin/su

You should have:

-r-sr-xr-x  1 root  wheel  14604 2011-08-21 20:24:28 /usr/bin/su*

As I mentioned in my previous message, somehow the UID change
had some strange side effects.





-- 
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: Sharing a mail folder between Linux and FreeBSD

On Fri, 25 Jan 2013 14:48:19 +0100, Ralf Mardorf wrote:
 On Fri, 25 Jan 2013 13:33:46 +0100, Polytropon free...@edvax.de wrote:
  $ ls -l `which su`
  -r-sr-xr-x  1 rocketmouse  wheel  16880 Dec 23 18:38 /usr/bin/su
 
  Erm... that looks horribly wrong.
 
  The permissions indicate that setuid is set, but the file
  owner is wrong. For comparison:
 
  -r-sr-xr-x  1 root  wheel  14604 2011-08-21 20:24:28 /usr/bin/su*
 
  This program has to belong to root. It seems that your
  attempt to reflect UID changes in the file permissions
  exceeded the scope of this task: Programs of the OS
  seem to be affected, which is definitely not good.
 
 IMO setuid alone already is a security risk.

The su program is part of the operating system, so it can
safely be considered safe. :-)




  $ ls -l /home/ | grep rocketmouse
  drwxr-xr-x  28 rocketmouse  rocketmouse 1536 Jan 25 12:17  
  rocketmouse
 
  You can use ls -ld to omit the grep step. :-)
 
 $ ls -ld /home/rocketmouse
 drwxr-xr-x  28 rocketmouse  rocketmouse  1536 Jan 25 13:19
 /home/rocketmouse
 
 :)
 
 I was sure that using grep is stupid and should have done a 'man ls',
 since 'help' wasn't helpful.

That's why man ls exists. :-)



 This issue and 'cat | grep' instead of grep
 only are common mistakes by many Linux users.

This reminds me to useless use of 'cat' which is often
used because it constructs a convenient and easy to read
chain of commands, but can often be avoided, especially
when files can be redirected from.



  Do you have other files in /usr or even /usr/local that do
  belong to rocketmouse (uid == 1000 or 1001) now? That should
  not have happened...
 
 /usr/binis ok
 /usr/includeis ok
 /usr/include/*  seem to be ok, I just checked some
 folders
 /usr/lib and /usr/lib/* are ok
 /usr/libdata and /usr/libdata/* are ok
 /usr/libexec and /usr/libexec/*/*   are ok
 /usr/ports  is ok
 /usr/ports/*seem to be ok, I just checked some
 folders
 /usr/sbin   is ok
 /usr/share  is ok
 /usr/share/*seem to be ok, I just checked some
 folders
 /usr/srcis ok
 /usr/src/*/*seem to be ok, I just checked some
 folders
 
 /usr/local  is ok
 /usr/local/bin and /usr/local/bin/* are ok
 /usr/local/bootstrap* and [...]/*   are ok
 /usr/local/etc  is ok
 /usr/local/etc/*seem to be ok, at least PolicyKit and
 ConsoleKit are
 /usr/local/include  is ok
 [snip]
 
 All /usr/local/* are ok and all /usr/local/*/* seem to be ok.
 Other directories in /usr and /usr/local are empty.

You can do something like this:

% ls -lR / | grep -v /home | grep rocketmouse

This will probably show some false-positives in /tmp and
maybe in /var, but should show nothing in /usr directly (or
in other top level system directories).



 OT: /usr/lib32 and /usr/lib32/* belong to the empty folders in /usr.

Allow me a polite note regarding terminology:

There are no folders. Those are called directories. This is
the valid technical term. A folder is the name of a typical
GUI representation element _for_ a directory.

Relations: is a vs. represents a or looks like a.

I know it's common to call directories folders, but this is
as wrong as calling a device driver Bob. ;-)



 So
 FreeBSD is multi arch capable?
 (since there's /usr/ports/astro/google-earth for amd64, I suspect it is)

The system shares some stuff across architectures, and it's
possible to run 32 bit applications on a 64 bit system, so
specific fixed bit width libraries are provided. This is
reflected in naming conventions. Even though the installer
might create those directories in advance, it's possible
that they only receive content under specific circumstances.

Ports do usually work on both systems. Those that do _not_
have a checking mechanism in their Makefile that indicates
on which platform they don't build, or if they are designed
for one specific platform only.



  Some programs check by whom they are called or who they
  belong to; if that's != root when it is _supposed_ to
  be root, that can cause problems, especially when it's
  not a simple x (execute), but s (setuid) program like
  an X display manager.
 
 So I guess I only need to correct the owner for /usr/bin/su.

If that's the only occurance, it should be sufficient.



 $ ls -l /usr/bin/su
 -r-sr-xr-x  1 root  wheel  16880 Dec 23 18:38 /usr/bin/su
 
 I wonder if setting suid is needed, while the kit family is installed. For  
 sure it's possible to add a rool to some kit config.

The su program is part of the OS, while things like PolicyKit
are additional software. It sounds doubleplusungood to modify
a standard (!) system setting at file permission level for such
a purpose.



 Restart
 
 PPPoE was enabled

Re: Sharing a mail folder between Linux and FreeBSD

On Fri, 25 Jan 2013 15:26:23 +0100, Ralf Mardorf wrote:
 On Fri, 25 Jan 2013 15:04:14 +0100, Polytropon free...@edvax.de wrote:
  % ls -lR / | grep -v /home | grep rocketmouse
 
 It's better I umount at least Arch Linux.

True. :-)



 There anyway is an issue, it doesn't show the pass, I checked this with
 
 $ ls -lR /home/ | grep -v /home
 
 after running
 
 $ ls -lR / | grep -v /home | grep rocketmouse
 
 IOW I get tons of files, but don't know to which directory they belong.

Sorry, that was something I didn't take into mind, you're right.
Maybe this command is more efficient:

# find / -exec ls -l {} \; | grep -v /home | grep rocketmouse

It may be a good idea to send the output into a temporary file
and check it when the command has finished. As I said, you will
probably see some false positives, but look for anything
strange in /usr.



  PPPoE was enabled automagically :).
 
  You probably have the required magic in /etc/rc.conf. :-)
 
 Yes, but it wasn't started, when the owner for /usr/bin/su wasn't root.

That was to be expected. :-)



On Fri, 25 Jan 2013 15:32:38 +0100, Ralf Mardorf wrote:
 PS:
 
 I guess the output is different for user and root and it does remove the  
 path, but anyway display also contend of /home.

Yes, access permissions matter a lot, so the command should be
run as root.




-- 
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: Sharing a mail folder between Linux and FreeBSD

On Fri, 25 Jan 2013 20:41:24 +0100, Ralf Mardorf wrote:
 On Fri, 25 Jan 2013 16:12:15 +0100, Polytropon free...@edvax.de wrote:
  find / -exec ls -l {} \; | grep -v /home | grep rocketmouse
 
 -r-xr-xr-x   1 rocketmouse  wheel  32736 Dec 23 18:38 ssh-agent
 -r-xr-xr-x  1 rocketmouse  wheel  32736 Dec 23 18:38 /usr/bin/ssh-agent
 ^C

Definitely to be changed.



 A lot of stuff from /tmp is shown without a path, however

That will probably be the false-positives I mentioned.



 root@freebsd:/usr/home/rocketmouse # ls -l /usr/bin/ssh-agent
 -r-xr-xr-x  1 rocketmouse  wheel  32736 Dec 23 18:38 /usr/bin/ssh-agent
 
 but without write permission.

The permissions haven't change (they're correct), just the
owner is wrong.

For comparison:

-r-xr-xr-x  1 root  wheel  23428 2011-08-21 20:24:03 /usr/bin/ssh-agent*

The program is installed without the w attribute by default.



 I now run
 
 root@freebsd:/usr/home/rocketmouse # find / -exec ls -l {} \; | grep -v  
 /home | grep rocketmouse  find_1000.txt
 
 and will take a look at it tomorrow.

That will be an interesting read. :-)




-- 
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: Sharing a mail folder between Linux and FreeBSD

On Fri, 25 Jan 2013 23:22:29 +0100, Ralf Mardorf wrote:
 On Fri, 25 Jan 2013 22:51:55 +0100, Joshua Isom jri...@gmail.com wrote:
  Sorry if my original command ended up breaking your system
 
 Don't worry, I run dump to backup it, but I'll try to fix it without  
 restoring it from the backup.

Maybe you can read the original owners from that backup and
just _change them_ accordingly? As the files haven't been
altered, there would be no need to rewrite them entirely.



-- 
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: Sharing a mail folder between Linux and FreeBSD

On Fri, 25 Jan 2013 16:15:28 -0600, Joshua Isom wrote:
 Ignore /proc, unmount it even.  You don't need it on FreeBSD and 
 shouldn't expect it to be there. 

As far as I know, Gnome (or at least GDM) _requires_ it to
be able to show the available user names. I have no idea
why. :-)



 As for the listings in /usr/local 
 they'll need fixed.  On my system, almost everything's owned by root. 

There are a few exceptions when files are owned by a daemon.
As I said, re-installing those parts (or even world) should
fix this, but maybe it's possible to apply some mtree magic
to fix the owner to the proper one (root in most cases).



 The man directories are owned by man, and 
 /usr/local/libexec/polkit-set-default-helper is set as polkit:polkit.

That's a good example for the non-root exceptions; there might
be others.



 There's a difference between lib directories and libexec directories. 
 Libraries are stored in lib and programs you're not expected to invoke 
 yourself are stored in libexec.

Correct. That's why my printer filters are in /opt/libexec. ;-)



-- 
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: Again: Security updates of individual porst

On Thu, 24 Jan 2013 16:17:34 +0100, Ralf Mardorf wrote:
 So I have to # portsnap fetch update?

Yes.



 If so, wouldn't it cause dependency  
 issues, if I wouldn't update all ports?

If you use portmaster to deal with updating your installation,
it will take care of the dependencies. However, it might lead
to unrelated ports being udated, too.

Example:

foo-1.0 has vulnerabilities.
Updating ports tree.
foo-1.1 is the safe version.
You're running portmaster foo.
foo is going to be be upgraded.
foo-1.1 relies on bar-2.5, whereas foo-1.0 relied on bar-2.2.
The portmaster run will also upgrade bar.

Possible problem:

baz-5.0 is installed and has been linked against bar-2.2.
baz itself doesn't need updating (not vulnerable).
Depending on how baz implements library calling (dependency),
it might have stopped working.

Solution:

Use portmaster -a to check all ports if they need updating.

Possible follow-up problem:

Ports you don't want to be updated (because you're totally happy
with the version you're running) will also be updated by this
command.

Solution:

Be selective in using portmaster and specify exactly the ports
you want to upgrade.

You can also use SVN to checkout only specific ports, but that
leads to an inconsistend ports tree which is not supported to
work (even though it _mostly_ will).






-- 
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: Sharing a mail folder between Linux and FreeBSD

On Fri, 25 Jan 2013 02:11:27 +0100, Ralf Mardorf wrote:
 Hi all, hi Joshua,
 
 On Thu, 2013-01-24 at 16:10 -0600, Joshua Isom wrote:
  find / -uid 1001 -exec chown 1000 '{}' \;
  find / -gid 1001 -exec chown :1000 '{}' \;
 
 I made one mistake, when I run find / -gid 1001 -exec chown :1000 '{}'
 \; for the fist time, I did it without the :. Later I run it without
 the typo.
 
 There's a serious problem now, rocketmouse still is 1001.

You should have been reading my advice about changing the
UID:GID in detail. :-)

What you seem to be missing is a rebuild of the database
that reflects the content of the password files (where you
have properly made the changes 1001 - 1000 in /etc/passwd,
/etc/master.passwd and /etc/group).

The command you're searching for is pwd_mkdb.



 .login_conf was '1000 1001', after I chown 1001 it, to start X as
 user, it became 'rocketmouse 1001', the user rocketmouse still can't run
 a X session anymore.

UIDs and GIDs should match here. All files belonging to rocketmouse
should be 1000:1000 _and_ the name rocketmouse should be
associated to those numerical values (see files mentioned
above).



 After rebooting this is the output I get:

Rebooting is _not_ the way to make a probem magically
go away. :-)



 # id rocketmouse
 uid=1001(rocketmouse) gid=1001 groups=1001,0(wheel)

This means the change of 1001 - 1000 has not been fully done,
in _all_ involved files.



 # ls -hAl /home/ | grep rocketmouse
 drwxr-xr-x  28 1000 rocketmouse   1.5k Jan 24 18:14 rocketmouse

Here, on file system level, the UID has been changed to 1000
properly, but this UID still doesn't have a matching name.



 # grep 100 /etc/group
 rocketmouse:*:1000:
 musicpd:*:1002:
 
 # grep 100 /etc/passwd
 rocketmouse:*:1000:1000:Ralf:/home/rocketmouse:/bin/sh
 musicpd:*:1002:1002:Music Player Daemon:/home/musicpd:/usr/sbin/nologin
 
 # grep 100 /etc/master.passwd
 rocketmouse:$1$3mMkzcfl
 $VuryrlzFZ92LmaC6cUOa/.:1000:1000::0:0:Ralf:/home/rocketmouse:/bin/sh
 musicpd:*LOCKED**:1002:1002:daemon:0:0:Music Player
 Daemon:/home/musicpd:/usr/sbin/nologin

All correct.

But pwd.db and spwd.db (the password databases with encrypted
content) don't reflect those informations!



 I repeated both find-chown several times and rebooted, nothing changed,
 it doesn't list any files anymore.

No, repeating what has already been done properly and then
rebooting is, as I said, not a way to make problems magically
go away. I don't know a setting where this should work... :-)

So here's what you need to do: Read man pwd_mkdb and rebuild
the databases.

If you would have used the vipw command to make the change
to the passwd (plain text) files, it would have called pwd_mkdb
after the change. But don't worry: Knowing those low level hacks
can be helpful in some worst-case scenario. :-)



-- 
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: Sharing a mail folder between Linux and FreeBSD

On Fri, 25 Jan 2013 07:51:30 +0100, Ralf Mardorf wrote:
 Thank you Kevin, thank you Erich,
 
 On Thu, 2013-01-24 at 21:10 -0500, kpn...@pobox.com wrote:
  The correct way to edit the password file is with the vipw
  command. When you are done with your changes it rewrites the password
  file AND rebuilds the password database.
  
  I'm guessing you have a stale password database now. Use 'vipw' to 
  make a trivial change and then save and exit out.
 
 Ok, I used mcedit to make the changes before and will try vipw now,
 resp. ...

That won't make any difference. :-)

If your $EDITOR points to mcedit, _that_ editor will be used;
afterwards pwd_mkdb will be called and the binary database
files will be updated - and your changes will be fine.



 On Fri, 2013-01-25 at 10:06 +0700, Erich Dollansky wrote:
  did you run something like?
  
  /usr/sbin/pwd_mkdb -d/etc /etc/master.passwd
 
 No, I didn't. I assume this is what vipw will do, so I can run this
 instead of using vipw?

Yes, because vipw can be seen as a chain editor - validate -
update database, involving the lower level programs that you
can call yourself any 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: Sharing a mail folder between Linux and FreeBSD

On Fri, 25 Jan 2013 08:32:55 +0100, Ralf Mardorf wrote:
 On Fri, 2013-01-25 at 08:03 +0100, Polytropon wrote:
  UIDs and GIDs should match here. All files belonging to rocketmouse
  should be 1000:1000 _and_ the name rocketmouse should be
  associated to those numerical values (see files mentioned
  above).
 
 Yes, but because I missed to update the database X login asked for 1001.

Correct - several programs query that database instead of
the plain text files.



  But pwd.db and spwd.db (the password databases with encrypted
  content) don't reflect those informations!
 
 So '# /usr/sbin/pwd_mkdb -d/etc /etc/master.passwd' is ok, regarding
 to ...

Yes, use that command.



  man pwd_mkdb and rebuild
  the databases.
 
 -c and -u switches could be used too, but aren't needed, since the
 entries are correct.

It's not needed to make things that complicated (to selectively
deal with entries, for example). The simple thing of

# cd /etc
# pkd_mkdb /etc/master.passwd

should do the trick here.



  If you would have used the vipw command to make the change
  to the passwd (plain text) files, it would have called pwd_mkdb
  after the change. But don't worry: Knowing those low level hacks
  can be helpful in some worst-case scenario. :-)
 
 And then I don't need to use vi, if the default text editor still should
 be vi.

The $EDITOR variable will be honored, and as long as the program
is available (and the terminal capabilities apply), it will work
as expected. :-)




-- 
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: Fresh installation 9.1

2013-01-23 Thread Polytropon

On Wed, 23 Jan 2013 19:49:13 +0530, Hrisikesh sahu wrote:
 Hi All,
 I am facing a strange issue..
 I installed 9.1 release and installation went successful
  After boot up  i got this error -
 
 
 File system had an unexpected inconsistency.
  ufs : /dev/ada0s1d (/var)
 unknown error ; ! Help
 Error - Aborting Boot
 
 Going to single user mode.
 
 #
 
 
 Please help me if i need to do anything else.

It seems that the /var partition has not been unmounted
properly. Perform a manual file system check and retry.

The command

# fsck -yf /dev/ada0s1d

should be applied in single user mode, with /var being
unmounted.

To make sure file systems are clean _prior_ to booting
the system, put

background_fsck=NO

into /etc/rc.conf; this may lead to longer startup times
when something is strange regarding file systems, but
it will prevent you from booting into a possibly damaged
environment (which could cause more trouble and maybe even
data loss, which justifies the longer boot time in my
opinion).




-- 
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: Fresh installation 9.1

2013-01-23 Thread Polytropon

On Wed, 23 Jan 2013 10:24:38 -0600, Derek Ragona wrote:
 Also if you have this in /etc/rc.conf:
 dumpdev=YES
 
 It always does a dump even on reboot, which also marks the filesystems as 
 dirty.

That doesn't seem to conform to what /etc/defaults/rc.conf
says:

dumpdev=NO# Device to crashdump to (device name, AUTO, or NO).
dumpdir=/var/crash# Directory where crash dumps are to be stored
savecore_flags=   # Used if dumpdev is enabled above, and present.
crashinfo_enable=YES  # Automatically generate crash dump summary.
crashinfo_program=/usr/sbin/crashinfo # Script to generate crash dump summary.

So YES looks invalid.

But note that I could be wrong here, I'm checking on a v8
system, not v9, so maybe the configuration has actually
been changed...



-- 
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: Sharing a mail folder between Linux and FreeBSD

2013-01-22 Thread Polytropon

On Tue, 22 Jan 2013 04:16:42 -0600, Joshua Isom wrote:
 On 1/21/2013 5:33 PM, Ralf Mardorf wrote:
  Hi :)
 
  I'm sharing the same directory for Evolution emails, by several Linux
  installs. For e.g. Ubuntu Precise it's Evolution 3.2.3 and for e.g.
  Ubuntu Quantal it's Evolution 3.6.0.
 
  I'm doing it by a link:
 
 
 It looks like to me you're asking for long term trouble.  You're using 
 multiple versions, so in the future there could be changes that could 
 corrupt your mail.  Why not just use an IMAP server instead?  It's what 
 I do, so my mail's shared between FreeBSD, Windows, and Android.

That might be overhead, but still the approach contains
potential for future trouble, as you correctly pointed
out.

The reason is simple: While you may not have trouble if
all programs use the same mechanism for _storing_ mail
(e. g. in mbox, MH or Maildir format), they might store
other aspects of communication (read / unread, address
books, configuration settings) differently. This should
happen _independently_ of the mail storage. As long as
all involved programs are the same version, it will
probably work without any trouble. But if one program
of a newer version decides to rewrite the configuration
data in a new (and backwards-incompatible) format, the
older versions will definitely run into trouble.

I've been using a similar approach in the past, having
several GUI and TUI mail clients use the same mail
_storage_. Still as you suggest, running a (local) IMAP
server may prevent trouble, at least on the long run,
and it enables you easier testing for mail clients that
do not use the same storage format as your old ones do.
Still you can have any storage backend you like, so
even plain text work (easily done with MH and Maildir)
can be done if required (like grepping through messages
or processing them automatically in whatever manner).




-- 
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: Upgrading 8.2 to 9.1, Gnome issue

2013-01-22 Thread Polytropon

On Tue, 22 Jan 2013 15:52:04 +0530, Hrisikesh sahu wrote:
 On command prompt Mouse and key board functionality is proper, but
 when i switched to GUI using #startx , then mouse and key stopped
 working.

This looks like a typical configuration error, not quite
uncommon today. :-)

Note that you shouldn't run startx from your root account
(which I assume you do because of #startx). For your local
user account, make sure you've configured Gnome 2 to start
automatically via ~/.xinitrc or ~/.xsession respectively.
Refer to the FreeBSD Gnome FAQ for details.

As you're also using GDM (Gnome display manager) for login,
your ~/.xsession should contain exec gnome2-session (which
I think to remember is the name of the Gnome desktop
executable, but check the documentation to be sure).



 I have included HLAD and DBUS to the rc.conf .
 /etc/rc.conf  --
 -
 keymap=us.iso
 gdm_enable=YES
 gnome_enable=YES
 hald_enable=YES
 dbus_enable=YES
 sshd_enable=YES
 moused_port=/dev/psm0

Looks correct.



 Now I installed wireshark , but I am getting a Gtk-warning **: cannot
 open display.

You can only run this program when in X (e. g. calling it from
an X terminal).



 Please help me what are ports used to upgrade mouse and Key Board
 functionality with X terminal.

Check your X configuration: If both input methods work in
text mode, but not within X, that's often the error.



 Please let me know if there is any command used to rebuild all the
 installed ports and update pkgdb.

If I remember correctly, it's portupgrade -arf, but as you're
using portupgrade, see man portupgrade, the EXAMPLES section
should show the proper 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: Upgrading 8.2 to 9.1, Gnome issue

2013-01-22 Thread Polytropon

On Tue, 22 Jan 2013 13:59:44 +0100, Ralf Mardorf wrote:
 On Tue, 22 Jan 2013 11:22:04 +0100, Hrisikesh sahu  
 hrisikeshs...@gmail.com wrote:
  #portupgrade gnome2
 
 Is there a future for GNOME 2 on FreeBSD?

I assume -- on the _long_ run -- Gnome 2 will be dead,
just as Gnome 1, KDE 1 and 2, and XFCE (capitals, so
it obviously means version 3). MATE and Cinnamon
are still maintained, and if there will be FreeBSD
ports, they will probably work for some time, until
eventually the required support libraries at certain
levels of abstraction will not work (or even build)
anymore, or when security concerns grow enough to
abolish the port. Infrastructures are moving on, and
backwards compatibility isn't the biggest strength
of desktop environments. :-(




-- 
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: Editors are broken after update

2013-01-21 Thread Polytropon

On Mon, 21 Jan 2013 09:51:13 +0100, Ralf Mardorf wrote:
 On Mon, 21 Jan 2013 03:53:56 +0100, Polytropon free...@edvax.de wrote:
  /usr/home/rocketmouse, _not_ user/home/rocketmouse
 
 It wasn't possible to copy the message, so I've written it. It's very  
 likely that I made a typo. I'm a dyslexic and since there's no big  
 difference between /usr and /user and it both is for user it's very  
 likely that I haven't notice it all the times when I read it. If I would  
 have written /usgr or something similar it could happen that I don't  
 notice it when reading it 2 or 3 times, but when reading it for the 4th  
 time I will notice it. It oven happens that people type 'unmount' instead  
 of 'umount', so I've seen Linux distros that ship with an alias 'unmount'  
 :D.

Seen in this context, the message was probably beginning
with /usr instead of user which makes sense (even though
it would be considered a bug when the editor is invoked
without a file name and it instead tries to open something
that is not a file to edit). It also fits the tradition that
usr is often pronounced user and therefor carries that
mental image.



 OT:
 
 # grep alias .cshrc
 alias h   history 25
 alias j   jobs -l
 alias la  ls -aF
 alias lf  ls -FA
 alias ll  ls -lAF
 
 I avoid using aliases. FWIW on Linux mailing lists that aren't for a  
 specific distro, it's unwanted to use aliases when posting to the list.

That's a valid advice, especially when the alias name suggests
that it does something it doesn't do in reality. However, a
common alias ll for list long is widely accepted, even though
the implementation (and additional flags and preferences) may
differ from system to system.

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

;-)



-- 
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: Sharing a mail folder between Linux and FreeBSD

2013-01-21 Thread Polytropon

On Tue, 22 Jan 2013 02:31:11 +0100, Ralf Mardorf wrote:
 On Tue, 2013-01-22 at 08:18 +0700, Erich Dollansky wrote:
   I guess it would be possible to change the id for the existing FreeBSD
   user and then to chown /home/user_name to fit to 1000?
  
  Of course, this would work. But then all existing files of the existing
  FreeBSD would be without owner.
 
 The current user is: rocketmouse
 The uid is : 1001
 
 Isn't it possible to change the uid to 1000?
 This would cause that the owner wouldn't be rocketmouse anymore, but
 still 1001. I then could run chown -R for /home/rocketmouse to switch
 from 1001 to back to rocketmouse = new uid 1000.

You would need to do two changes: First in the password database,
with chsh (tidy way) or by editing the /etc/passwd, /etc/master.passwd
and /etc/group files plus rebuilding the database with pwd_mkdb
(untidy way) to assign rocketmouse = 1000 on FreeBSD.

Then you would also have to promote this change to the file
system, as all the files still belong to a user with UID 1001.
Use chown -R with the new numerical value of 1000.

Result: Your user would have the UID 1000 on all systems, so
all the low level functions would behave similarly.



 Or another idea would be to create a new user with the uid 1000 and then
 to add rocketmouse to the group of this user. I guess this is what you
 already recommended.

Yes, that would also work. You only have to make sure that
group permissions are valid, and the access permission is
provided in /etc/group properly.



-- 
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 fan control via FreeBSD

On Sat, 19 Jan 2013 18:19:39 +0100, Xavier wrote:
 On Sat, Jan 19, 2013 at 05:04:36PM +0100, Polytropon wrote:
 
 Hi Polytropon,
 
  On Sat, 19 Jan 2013 16:55:23 +0100, Xavier wrote:
   On Sat, Jan 19, 2013 at 03:13:58PM +0100, Fabian Keil wrote:
  
   Hi Fabian,
  
Xavier xavierfreebsdquesti...@gmail.com wrote:
   
 Is there any way to control the on and off the fan on a laptop using 
 FreeBSD?
   
It depends on the laptop. On mine it works:
   
fk@r500 ~ $sysctl -ad | grep fan
dev.acpi_ibm.0.fan_speed: Fan speed
dev.acpi_ibm.0.fan_level: Fan level
dev.acpi_ibm.0.fan: Fan enable
   
  
   My output of 'sysctl -ad | grep fan' don't show these parameters. Why ?
   Because, my BIOS don't support these aparameters ? Another answer ?
 
  For laptops, you usually load a kernel module for interfacing
  with the specific ACPI functions, like acpi_ibm.ko. There are
  several others, but see man acpi_ibm for some impressions.
 
 
 Yes, but I have an acer Aspire 5634WLMi, and:
 
 % ls /boot/kernel/acpi*
 /boot/kernel/acpi_asus.ko   /boot/kernel/acpi_panasonic.ko
 /boot/kernel/acpi_asus.ko.symbols
 /boot/kernel/acpi_panasonic.ko.symbols
 /boot/kernel/acpi_dock.ko   /boot/kernel/acpi_sony.ko
 /boot/kernel/acpi_dock.ko.symbols
 /boot/kernel/acpi_sony.ko.symbols
 /boot/kernel/acpi_fujitsu.ko/boot/kernel/acpi_toshiba.ko
 /boot/kernel/acpi_fujitsu.ko.symbols
 /boot/kernel/acpi_toshiba.ko.symbols
 /boot/kernel/acpi_hp.ko /boot/kernel/acpi_video.ko
 /boot/kernel/acpi_hp.ko.symbols
 /boot/kernel/acpi_video.ko.symbols
 /boot/kernel/acpi_ibm.ko/boot/kernel/acpi_wmi.ko
 /boot/kernel/acpi_ibm.ko.symbols/boot/kernel/acpi_wmi.ko.symbols
 
 I don't have an acer kernel module for ACPI.

Two options:

If you can derive from the documentation of your Acer laotop
if it is _compatible_ to one of the implementations provided
by the system, use that instead.

You can do trial  error to see if one of the modules works,
even though the name is different. In worst case, load them
all (by using * wildcard) and check with kldstat.




-- 
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: Dependencies after port tree update

On Sat, 19 Jan 2013 20:40:26 +0100, Ralf Mardorf wrote:
 I now run the command with what options ever, don't remember, but it's  
 compiling and every few hours I have to answer n/y to delete a file, but  
 there aren't stops regarding to the configuration.

I assume there is a -d or -D (delete distfiles) option missing
or probably no -y.



 I'm compiling since  
 yesterday 17 or 18 o'clock, perhaps -8 hours where it might stopped  
 regarding to such a question, when I didn't notice it.

I know portmaster has decent options to avoid those interactivity
intermissions to provide a full batch based unattended update
configuration. You can use /usr/local/etc/mergemaster.rc to store
your options mermanently; there is a sample file provided.



 It's an Athlon  
 dual-core 2.1GHz with 4GB RAM, since I didn't edit anything, I suspect  
 just one job is running. 837 ports where found with 420 new versions. I  
 wonder how long it will continue compiling.

That doesn't sound very much. Less than 24 hours of scheduled
happy downtime should be sufficient. :-)




-- 
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: bsdinstall(8) line drawing characters

On Sun, 20 Jan 2013 05:23:22 +0100, David Lazaro Saz wrote:
 Hi,
 
 I've tried to find an answer to this question without success.
 
 Why does bsdinstall(8) use ASCII characters for drawing lines
 instead of line drawing characters as the old sysinstall(8) did?

I assume this has to do with a change which happened long time
ago. In the past, those characters could be used in text mode
(for example by the Midnight Commander) and even in combination
with LC settings like de_DE.ISO8859-1. Probably the character
sets in /usr/share/syscons/fonts don't support them anymore. I've
had success using the fonts directory content from a 4.x system
and transition it though various later versions, but today I
simply accept the inavailability. :-)

Because the line drawing characters in the Midnight Commander
don't work for me anymore, I use the -a option. Still when run
in an X terminal, they are properly displayed.

I think to avoid problems, the use of linedrawing characters
has been dropped.


-- 
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: Dependencies after port tree update

On Sun, 20 Jan 2013 07:35:16 +0100, Ralf Mardorf wrote:
 The last command I run was
 root@freebsd:/usr/ports/ports-mgmt/portmaster # portmaster --no-confirm -y  
 -G `cat ~/installed-port-list`
 
  From time to time I manually had to answer yes, when I was asked if a file  
 should be deleted.

From man portmaster:

 While recursing through the dependencies, if a port is marked IS_INTERAC-
 TIVE this will be flagged.  In the absence of this notification, under
 normal circumstances the only user interaction required after the port
 starts building is to answer questions about the deletion of stale dist-
 files.  This can be eliminated with the -d or -D options.

Maybe this helps to avoid that interaction in the future.

If you want to continuously update your installed ports, you
don't need to rely on the ~/installed-port-list file anymore,
especially as portmaster can automatically determine which
ports need to be worked on (portmaster -a). It's also a nice
feature to put the non-interactivity options into a config
file (/usr/local/etc/portmaster.rc) so you really only need
the -a option if you want to start an update run.



-- 
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: Editors are broken after update

On Sun, 20 Jan 2013 09:46:00 +0100, Ralf Mardorf wrote:
 Wow, if on Linux something is fishy, it usually has to do with Lennart  
 Poettering, does he break FreeBSD too?
 
 $ su -
 Password:
 root@freebsd:/root # mcedit
 
 Error
 /root is not a regular file [ Dismiss ]

Seems to be a problem with the configuration. Temporarily
try the following:

# cd /root
# mv .mc .mc.orig
# mcedit

That should start the editor with the defaults.

Of course, /root is not a regular file, it's a directory. :-)



 root@freebsd:/root # gedit
 
 (gedit:17410): Gtk-WARNING **: cannot open display:

This editor requires X. If you're running the above su command
in an xterm, use su -m root and try again.



 root@freebsd:/root # vi
 
 vi does run

Because vi belongs to the OS, it's not a port.



 root@freebsd:/root # logout
 $ mcedit
 Failed to run:
 Your old settings were migrated from /home/rocketmouse/.mc
 to Freedesktop recommended dirs.
 To get more info, please visit
 http://standards.freedesktop.org/basedir-spec/basedir-spec-latest.html [1]

Seems that the new MC version has migrated its configuration
files somewhere else...



 $ gedit
 
 As user gedit does start, so perhaps I need gksu to run it as root.

That sounds wrong. What would be a reason for an installed editor
that is intended to be run on X _not_ to run from a user account
that currently runs X?



 [1] That's a nightmare :(, I don't want to read something that has to do  
 with Lennart Poettering.

I'm not sure in how far this XDG stuff applies to MC / mcedit...
but the message seems to originate from MC which adopts to the
FreeDesktop recommendations...



 What the hell is broken now, regarding to the insane ideas of this man?

Implied statement (not _my_ words): Every program should store
its configuration data according to the XDG / FreeDesktop
specification. Seems that the MC agreed.



 OT:  
 http://linux-bsd-sharing.blogspot.de/2008/08/howto-musicpd-music-player-daemon-on.html
   
 is not a valid howto for my upgraded FreeBSD ports :(.

From a short look at it, the content looks valid. But I haven't
tried it in order to confirm that it's working.



-- 
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: Editors are broken after update

On Sun, 20 Jan 2013 11:21:17 +0100, Ralf Mardorf wrote:
 On Sun, 20 Jan 2013 10:38:45 +0100, Polytropon free...@edvax.de wrote:
  # cd /root
  # mv .mc .mc.orig
  # mcedit
 
 $ mv .mc .mc.pre.update-01-Jan-2013
 $ su root -c mv /root/.mc /root/.mc.pre.update-01-Jan-2013
 $ mcedit
 
 Error
 user/home/rocketmouse is not a regular file [ Dismiss ]

Also looks wrong, that doesn't seem to be a valid path.
I assume /home/rocketmouse would be your home directory,
so MC (or mcedit) would access a configuration structure
within that directory (~/.mc).



  That should start the editor with the defaults.
 
 It doesn't do it for the user.

Are you able to start the normal Midnight Commander instead?
And if yes, PF4 on a file to invoke the editor?



  Of course, /root is not a regular file, it's a directory. :-)
 
 Yes and in this case it's true for the users home directory, but I only  
 run mcedit, without a file name.

That shouldn't be a problem. If I start mcedit without a filename
here, I _still_ get the editor launched with an empty file, with
PF2 allowing me to enter a file name.

(Version here: 4.7.5_1 on FreeBSD 8.2-STABLE i386).



  This editor requires X. If you're running the above su command
  in an xterm, use su -m root and try again.
 
 On Linux regarding to this, there is a difference between su and su -,  
 I never had to run su -m root.

The -m option makes sure the environment is not touched, so $DISPLAY
will be kept. If you do a full root login (su - and su -l simulate
a full login, and omission of the name assumes root, so su - is
like su -l root, discarding your user's environment).



 $ su -m root
 # mcedit
 
 Error
 user/home/rocketmouse is not a regular file [ Dismiss ]

That was designed for running gedit as root (because of X). :-)

Again, the path specification just looks wrong - there is no
such thing (not absolute, not relative to ~).



 # ls -l .config/mc
 total 8
 -rw-r--r--  1 rocketmouse  rocketmouse  2931 Jan 20 10:55 ini
 drwx--  2 rocketmouse  rocketmouse   512 Jan 20 09:28 mcedit
 -rw-r--r--  1 rocketmouse  rocketmouse 1 Jan 20 10:51 panels.ini

Okay, so this looks like it would be the new configuration
location. For comparison:

% ls .mc
Treebindings_1  filepos ini
bindingscedit/  history panels.ini

And cedit/ is now mcedit/.



 # rm -r /root/.mc* /root/.config/mc /home/rocketmouse/.mc*  
 /home/rocketmouse/.config/.mc
 rm: /home/rocketmouse/.config/.mc: No such file or directory
 # rm -r /home/rocketmouse/.config/mc
 
 # mcedit
 
 Error
 user/home/rocketmouse is not a regular file [ Dismiss ]

Time for portdowngrade? :-)



 # gedit
 GConf Error: Failed to contact configuration server; some possible causes  
 are that you need to enable TCP/IP networking for ORBit, or you have stale  
 NFS locks due to a system crash. See http://projects.gnome.org/gconf/ for  
 information. (Details -  1: Failed to get connection to session: The  
 connection is closed)
 GConf Error: Failed to contact configuration server; some possible causes  
 are that you need to enable TCP/IP
 [snip]
 networking for ORBit, or you have stale NFS locks due to a system crash.  
 See http://projects.gnome.org/gconf/ for information. (Details -  1:  
 Failed to get connection to session: The connection is closed)
 g_dbus_connection_real_closed: Remote peer vanished with error: Underlying  
 GIOStream returned 0 bytes on an async read (g-io-error-quark, 0). Exiting.
 Terminated

What a scary error message. It seems that gedit relies a lot on
Gtk / Gnome services running to access its own configuration, and
that is not accessible from an instance running as root (in
opposite to running as the user who started X and the services
required).



 On Linux it's a common issue for some distros, when using apps from  
 bloated DEs.

So it seems that this nice tradition is also carried with programs
ported to FreeBSD. Excellent.

% gimp
(gimp:3045): GLib-WARNING **: goption.c:2132: ignoring no-arg, optional-arg or 
filename flags (8) on option of type 0



 It usually needs gksu or similar.

Programs being designed to be primarily used within specific
desktop environments heavily rely on the mechanisms provided
by those environments, even though one would consider them
optional (as the program can be used stand-alone). Obviously
it's not true to consider that.



 At the end of the update I  
 got the information, that for K3b I have to set the suid flag for cdrecord  
 and cdrdao.

That doesn't seem to be the default:

% ll /usr/local/bin/cdr*
-r-xr-xr-x  1 root  wheel  564156 2011-08-22 03:01:50 /usr/local/bin/cdrdao*
-r-xr-xr-x  1 root  wheel  399044 2011-08-22 03:12:57 /usr/local/bin/cdrecord*

% ll /dev/cd* /dev/pass* /dev/xpt*
crw-rw-r--  1 root  operator0, 110 2013-01-20 09:18:50 /dev/cd0
crw-rw-r--  1 root  operator0, 111 2013-01-20 09:18:50 /dev/cd1
lrwxr-xr-x  1 root  wheel4 2013-01-20 09:18:58 /dev/cdrom@ - acd0
crw-rw  1

Re: Editors are broken after update

On Sun, 20 Jan 2013 14:20:10 +0100, Ralf Mardorf wrote:
 Since I only run mcedit without a file, the error message is grotesque  
 and btw.

Exactly - it is, and probably misleading as it implies that
mcedit is trying to access something that doesn't even exist,
even with a maximum of imagination.



 $ ls -l /usr/home
 total 2
 drwxr-xr-x  26 rocketmouse  rocketmouse  1536 Jan 20 13:07 rocketmouse
 $ ls -l /home
 lrwxr-xr-x  1 root  wheel  8 Dec 18 19:19 /home - usr/home

That is a valid symlink on a FreeBSD installation (and has been
a default for a long time), so /home/rocketmouse will equal
/usr/home/rocketmouse, _not_ user/home/rocketmouse as shown
in the error message.



 Thank you for the hints. I take the issues with a good portion of humor.

I think this isn't funny anymore - an editor that stops working
without explaining the reason in an understandable way...



 This might explain why you're confused regarding to the /home path:
 
 $ cat /etc/fstab
 # Device  Mountpoint  FStype  Options DumpPass#
 /dev/ad4s1b   noneswapsw  0   0
 /dev/ad4s1a   /   ufs rw  1   1
 /dev/ad4s1e   /tmpufs rw  2   2
 /dev/ad4s1f   /usrufs rw  2   2
 /dev/ad4s1d   /varufs rw  2   2
 /dev/acd0 /cdrom  cd9660  ro,noauto   0   0
 #proc   /proc   procfs  rw  0   0

No, looks perfectly valid. On my other system, for example, I
have /home being /export/home with the proper symlink, and no
problems at all.



 $ mc
 
 does start midnight commander, but if I push 4 Edit I get the same error  
 message.

Okay, so the problem lies deeply within the editor.



 I did dump -0Launf [...] everything and will continue later. Hopefully  
 with some fun, when starting with audio, instead of mcedit.

In worst case, try portdowngrade (or use svn to obtain an older
copy of the mc port where the editor hasn't been disimproved yet).




-- 
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: LaTeX Error: File `utf8x.def' not found