Re: [ *** ] Job anacron.service/stop running (15min 49s / no limit)

[Eduard Bloch]

reopen 915379
thanks

Hallo,
* David Wright [Sun, Feb 11 2024, 10:20:16PM]:
> On Sun 11 Feb 2024 at 20:41:51 (+), Darac Marjal wrote:
> > On 11/02/2024 11:21, Rainer Dorsch wrote:
>
> > > - How do I set a timeout/limit for anacron, that it cannot block forever
> > > during a reboot?
> >
> > It may be germane to point out that anacron.service already explicitly
> > sets "TimeoutStopSec=Infinity". So, in the opinion of the developers,
> > the service shouldn't be prematurely killed. Of course you, as the
> > system administrator, always have the right to countermand that sort
> > of decision, but it would be curious to find out why the developers
> > thought they needed to override the systemd default in the first
> > place?
>
> Bug #915379 explains all: long-running cron jobs, like backups, can
> get killed, and there was also an issue with exim.

Yes, and? The opposite is: you have some stupid (and UNKNOWN) task which
hangs forever because of some programming bug. And then your whole system
locks up, unable to reboot, and no way to recover it because the reboot
is stuck because of this. I am observing this on my old hacking laptop
right now, the system took many minutes (5? 7?) to continue, but even
that was pure luck.

Sorry, no, that cannot be the proper way. I am reopening 915379 now, the
maintainer should maybe come up with some sane solution.

Best regards,
Eduard.

Re: Is perl still the No.1 language for sysadmin?

[Eduard Bloch]

Hallo,
* Emanuel Berg [Mon, Apr 03 2023, 02:15:10AM]:

> > The reason Perl gives you more than one way to do anything
> > is this: I truly believe computer programmers want to be
> > creative, and they may have many different reasons for
> > wanting to write code a particular way. What you choose to
> > optimize for is your concern, not mine. I just supply the
> > paint—you paint the picture.
>
> I agree but I think maybe the success of Python, and its
> development speed, is actually because of some of that
> rigidness, yes, including the whitespace lack of freedom.

I don't think so, Sir! Python has certain advantages but the "meaningful
whitespace" is IMHO not one of them.

That said, I have been an active Perl user ~20y ago and for the last
couple of years slowly converting to Python for scripting purposes,
still using Perl here and there.

Therefore, my recent impressions and reflections:

Perl:
PRO:
- still excels on creating quick solutions in a one-liner for many
  purposes where string/text input/output and some generic algorithms
  count
- great cross-platform availability as long as you don't need CPAN, even
  part of Git-Bash and therefore an "obligatory" component
- IMHO clean lifecycle of variables. Means, you can set "strict" and
  then be sure that you manage your variables correctly, without much
  risk of strange runtime effects
- flexibility in statement writing (although making them sometimes
  looking strange and requiring more brackets than you initially wanted)
- a "friendly" API for users who just came from Bash or AWK scripting,
  many things would look familiar and need only minor adaption,
  especially when one knows "computer science" and understands what is
  going on underneath anyway
NEUTRAL:
- true threads are possible... the last time I tried that was great but
  unstable, but I think they have fixed it in a rewrite a while ago (not
  tried again for years, cannot tell for sure)
CONTRA:
- OOP is awkward, has always been, it just sucks
- still sucky when it comes to complex data structures
- ECO system unfortunately slowly fading away
- error handling (exceptions?)
- sometimes too rigid ways of method calling

Python3:
PRO:
- HUGE and modern ECO system, and mostly good&useful documentation
- Developers listening to user's wishes, recognizing and completing
  missing features and seeing modern developments
  (things like string interpolation with f-literals)
- flexible ways of method calling with default/optional/... parameters
- a certain level of rigidness keeps your code understandable even if
  you touch it a year later, and also for the code from your colleagues
- the uniform usage of certain infrastructure components (like argparse)
  became universal in the last few years
NEUTRAL:
- OOP is okay. Good or bad depending on whom you ask. Good because it
  allows for "wild west" OO design which still allows the interpreter to
  convert it into something useful in the end. Bad because this happens
  and runtime and because there are life-saving type checks and no real
  interfaces possible with Python.
CONTRA:
- OOP: ... one MUST keep writing "self." prefix all the time. What the
  heck, one of the reasons for refactoring to OO style is to have this
  shortened because the context prefers local objects.
- "meaningful whitespace" - whoever came up with that concept should maybe
  rethink it, and give people alternatives
- no (easy user-controllable) concept of references etc. which always
  feels like something causing memory waste
- performance SUCKS when it comes to actual computation (yes, there is
  Cython coming as alternative and replacement/completion but it has its
  own issues)
- awkward lifetime of variables (yes, they are not variables,
  just "assignments" AND YOU GONNA KNOW IT WHEN YOU UNDERSTAND REAL PYTHON,
  haha, I know, but still, it's a very unusual model, and unnatural /
  problematic to handle for developers who are used to scope driven languages)
- performance (again) of multi-threading... GIL sucks, making regular
  Python basically a single-threaded language (considering that
  performance gain is the main performance of MT)

So that were my few cents regarding the scripting languages. For a real
n00b nowadays, I would still recommend to learn Bash first. But always
keep in mind that "Bash programming" is a mix of executing commands in
the own language but mixing the result with the external forking of
other programs.

However, the lines between "scripting" and "real programming" languages
are blurry. If one is asking for the later, nowadays I would suggest
learning C# or Rust.

Eduard.

Re: No space left when: update-initramfs: Generating /boot/initrd.img-5.10.0-5-amd64

[Eduard Bloch]

Hallo,
* Marco Ippolito [Wed, Apr 07 2021, 09:20:46AM]:

> dpkg: error processing package initramfs-tools (--configure):
>  installed initramfs-tools package post-installation script subprocess 
> returned
>  error exit status 1
>  Errors were encountered while processing:
>   initramfs-tools
>
> # df -h /boot
> Filesystem  Size  Used Avail Use% Mounted on
> /dev/nvme0n1p1  236M  233M 0 100% /boot

That's not much if you have more than a few kernels installed. They are
simply too fat nowadays, too many drivers included. And with the "most"
strategy (which you normally want to have a bootable system), they are
just too fat nowadays. I decided to let MY initramfs images go on diet
and added a little script which removes a few drivers that I certainly
don't need (checked with lsmod) and which contained lots of firmwares
and similar stuff.

So, one _might_ try that but only on your own risk, it could render your
system unbootable. I selected those three because they contributed most
to the overall usage (unpacked initramfs image with unmkinitramfs and
checking with "du|sort -n").

$ cat /etc/initramfs-tools/hooks/zz_drop_stuff
#!/bin/sh

if ! test "$DESTDIR"; then
. /usr/share/initramfs-tools/hook-functions
fi

set -x

#experiments: radeon infiniband lpfc qla2xxx qla4xxx cxgb4 drbd nfs bfa f2fs 
xfs btrfs aic7xxx gma500

for nam in i915 amdgpu ethernet
do
find ${DESTDIR} -name $nam | xargs --no-run-if-empty rm -rv
done

#USEDMODPAT=$(lsmod | grep -v " 0 " | sed -e "s, .*,,;s,_,.,g;s,$,.ko,")
#for pat in $USEDMODPAT
#do
#find ${DESTDIR} | grep "/${pat}$" || echo "WARNING: $pat not found, 
essential module missing!"

That script probably should be set executable.

Best regards,
Eduard.

Re: Archivemail

[Eduard Bloch]

Hallo,
* Mark Fletcher [Fri, Jan 22 2021, 11:22:19PM]:
> Anyone know what happened to archivemail in Debian?
>
> packages.debian.org shows it was in Jessie, Stretch and Buster, but it
> is in neither sid nor Bullseye which presumably means it has been
> dropped. Anyone know why?
>
> bugs.debian.org doesn't show any bugs against it and google is coming up
> short too.

It's gone along with the rest of the Python2 eco system. In the meantime
I have extended my own alternative (mail-expire package) to do something
similar, at least for the usecases I care about. Maybe it's sufficient
for you too, feel free to report bugs or wishes.

Best regards,
Eduard.

Re: apt-cacher-ng's expiry job failing

[Eduard Bloch]

Hallo,
* David Wright [Sat, Apr 27 2019, 09:31:50AM]:
> > /var/cache/apt-cacher-ng/_xstore/rsnap/debrep/dists/unstable/45961554550630227606591
> >
> > I removed the file and it started complaining about other similar files.
> > After deleting a couple, the logs became even more unhelpful:
> >
> > Maintenance task Expiration, apt-cacher-ng version: 2 (Cancel)

Old. And had some issues in that area. Try
https://packages.debian.org/stretch-backports/apt-cacher-ng

Best regards,
Eduard.

stracing login process with systemd?

[Eduard Bloch]

Dear Lazynet,

I have some trouble with one desktop system running Sid. After some
updates a couple of weeks ago, user programs started to fail, like
pulseaudio no longer running or "systemctl suspend" not working.

After little investigation I have found out that "systemd --user"
process is not working which is supposed to run user-specific services.

Which is probably somethign that libpam-systemd should have invoked -
that was the best guess I could extract from the documentation (the
manpage of systemd is AFAICS NOT telling you who/what is supposed to
start the user systemd).

I had some exchange with the maintainer in the BTS but so far that's not
really fruitful. I would like to just strace the whole login process
(console or xdm) and check what's happening there around systemd
invocation, and what's failing.

Does anyone know how to insert strace or any similar tracing utility?

I can imagine that one could try to use kernel tracing here but that
would be a huge hammer.

Thanks,
Eduard.

-- 
Every great idea is worthless without someone to do the work. --Neil Williams

Re: apt-cacher-ng related program and package suggestion

[Eduard Bloch]

Hallo,
* tuxderlinuxfuch...@gmail.com [Wed, Jun 14 2017, 09:57:50PM]:
>Hello,
> 
>I have a suggestion for a client program/package in combination with
>apt-cacher-ng
> 
>Consider following situation:
> 
>User U_A has a laptop which he uses at home and at university
> 
>He has a poor internet connection in both places.
> 
>So he set up apt-cacher-ng services in both places but they are on
>different networks (he does not have permission to change the network
>settings at university to make the IP addresses of the apt-cacher-ng
>servers math)
> 
>Lets say at home the server HOST_1 is [1]http://192.168.1.100:3142
>At university: server: HOST_2: [2]http://192.168.76.56:3142
> 
>He uses Ubuntu (so this should work for Debian in a similar way).
> 
>He makes a backup "/etc/apt/sources.list.bak" of /etc/apt/sources.list
> 
>At university he sets up the cache repo as follows (after overwriting the
>/etc/apt/sources.list with a copy of the backup if it was changed to use
>the home apt-cacher-ng)
> 
>sed -i 's/de.archive.ubuntu.com/192.168.76.56:3142/g'
>/etc/apt/sources.list
> 
>The same procedure at home.
> 
>The shown way is not very comfortable.

You can automate this easily with a one-liner in /etc/network/ip-up.d/ ,
for example.

Actually, apt-cacher-ng provides a similar kind of automation but it can
only decide between "use a proxy" and "don't use a proxy" and the need
to switch by checking whether that proxy server is broken/unreachable by
timeout or a user hook script. This can be used well in a proxy cascade,
i.e. one apt-cacher-ng running on the mobile device itself in
non-storing mode, and local applications use that proxy on localhost.

There is, however, no support for multiple proxies that would be
switchable upon the mentioned conditions check. This could be added,
maybe in future. Please open a feature requeston Alioth (apt-cacher-ng
space) or in the BTS.

Regards,
Eduard.

-- 
Every great idea is worthless without someone to do the work. --Neil Williams

Re: Jessie for Udoo X86?

[Eduard Bloch]

Hallo,
* Larry Dighera [Sun, Apr 16 2017, 09:27:46PM]:
> 
> The new Udoo X86 boards have just begun to ship: . 
> 
> Is anyone able to provide a link to the 64-bit Debian Jessie USB/SD
> installation ISO/img?

Did you try the regular installer from USB stick already?

Data sheet indicates that it supports "All Linux Flavors for x86". Which
means that it's probably usual Intel hardware inside. It might lack a
few drivers for recent hardware revisions but you could install a
kernel from jessie-backports in that case.

Best regards,
Eduard.

Re: Quiet gaming capable gfx card for Debian Sid

[Eduard Bloch]

Hallo,
* Floris [Sun, Aug 14 2016, 11:08:49AM]:
> Op Sat, 13 Aug 2016 20:54:51 +0200 schreef Eduard Bloch :
> 
> > Hallo,
> > 
> > I would like to spend some hours on a normal game without having to
> > reboot into Windows. I have a Steam account and some Linux games are
> > available (at least some from Valve).
> > 
> > But the video card I have is not ok. It's a Nvidia Geforce 750Ti
> > (passive cooling), it works quite well with the Nouveau drivers but
> > there are too many glitches in 3D games, some are basically not
> > playable. I remember that Nvidia drivers (non-free) have been a lot
> > better but I cannot make them work with current Debian Unstable, even if
> > I install the latest packages from Debian Experimental, the drivers
> > simply don't compile with latest kernels.
> > 
> 
> Don't use the experimental branch, but use the latest version from svn:
> https://wiki.debian.org/NvidiaGraphicsDrivers#Building_newer_releases_from_SVN
> 
> I don't have any troubles with the 367.35 version (kernel 4.6)

Thanks for the hint, I will give it a try, either with kernel 4.7 or at
least 4.6 which is in Sid.

Looks like maintainers need some help. It shouldn't take three months to
bring an important update into Experimental.

Regards,
Eduard.

Re: Quiet gaming capable gfx card for Debian Sid

[Eduard Bloch]

Hallo,
* Stefan Monnier [Sat, Aug 13 2016, 03:26:27PM]:
> > drivers basically broken, or you can buy some 3 year old hardware for
> > insane prices, or what else can a normal gamer do?
> 
> Hmmm... my experience with (2nd hand) 3 year old hardware is that it's
> rather cheap, yet I don't think that's what you meant by "insane
> prices".

"insane prices" is what some dealers in Europe expect for the few
remaining ones in their stock, i.e. about 300eur.

And, what's confusing me most: people want over 200eur on eBay for a
used R9-290 which makes much more heat and noise than a NEW RX470
(according to reviews), which is offered for just 10-20% more - however,
it didn't become available yet.

Regards,
Eduard.

Quiet gaming capable gfx card for Debian Sid

[Eduard Bloch]

Hallo,

I would like to spend some hours on a normal game without having to
reboot into Windows. I have a Steam account and some Linux games are
available (at least some from Valve).

But the video card I have is not ok. It's a Nvidia Geforce 750Ti
(passive cooling), it works quite well with the Nouveau drivers but
there are too many glitches in 3D games, some are basically not
playable. I remember that Nvidia drivers (non-free) have been a lot
better but I cannot make them work with current Debian Unstable, even if
I install the latest packages from Debian Experimental, the drivers
simply don't compile with latest kernels.

This makes me think, is AMD a better alternative? I could by one, but
only if the driver support situation is acceptable. And I am not sure
WRT recent developments. There is the amdgpu driver in Sid but the list
of supported models includes only R9-280 or 290 stuff which is two years
old. And no such card on the market seems to support semi-passive
cooling. And then, there are "AMDGPU PRO" drivers on the AMD page, which
seem to be the successor of the non-free fglrx driver. But the packages
there are only made for Ubuntu.

I really don't want to troll but this makes me wonder... am I missing
something? Or does AMD give a s...t about users unless they are running
the latest Ubuntu LTS and nothing else?

And I don't like this situation. NVidia drivers basically broken, AMD
drivers basically broken, or you can buy some 3 year old hardware for
insane prices, or what else can a normal gamer do?

Regards,
Eduard.

Re: Putting It All On a Stick

[Eduard Bloch]

Hallo,
* Thomas Schmidt [Thu, Dec 31 2015, 06:11:39PM]:
> Hi,
> 
> Steve Matzura wrote:
> > The one thing I forgot is cp and dd are not the same.
> 
> I was surprised that cp onto block devices really works as
> needed for this purpose.
> You could also do
>   cat debian-8.2.0-amd64-BD-1.iso >/dev/sdc
> 
> But for old unixers no other tool than dd comes to the mind.

It's like duck typing: if it has syscalls like cp and uses them like cp
then it must be cp... at least something like that.

In the end, dd gives you more control (buffer sizes, IO synchronization
behavior for data/metadata) and it can print the status with SIGUSR1.

But for regular "push data around from FD to FD" job the only crucial
difference I remember is the nasty behavior of dd when a short read
occurres - aborting instead of retrying as cp does. Not sure whether
they changed the default mode to blocking mode nowadays.


Regards,
Eduard.

-- 
 Du solltest wissen: Du bist hier im Internet. Da wird jede Deiner
Aussagen so verdreht, dass man Dich in Grund und Boden flamen kann.

Re: Changing default XDB browser w/o GNOME etc.

[Eduard Bloch]

Hallo,
* Thomas Schmidt [Sun, Dec 13 2015, 10:32:01AM]:

> > Feels like things that have been established in a sane way (alternatives
> > system and sensible-browser) are now reverted to the chaos we used to
> > manage in the nineties :-(
> 
> Well, the kernel has also reverted to the CD drive quirks
> of the 90s. History is chaotic in a cyclical way.

Don't get me started... some things start rotting quickly when the user base
is shrinking.

> > Message-ID: 
> 
> Some things stay constant, though.

Or maybe not. Thanks for reminding, I should use the holidays to finally
overhaul my mail setup.

Regards,
Eduard.

Re: Changing default XDB browser w/o GNOME etc.

[Eduard Bloch]

Hallo,
* Martin Read [Sun, Dec 13 2015, 11:18:24AM]:
> On 13/12/15 09:19, Eduard Bloch wrote:
> >Observation: various programs use the XDG mechanism (freedesktop
> >configuration method) to start a specific browser to visit a hyperlink.
> >Easy to verify with the xdg-open tool.
> >
> >But how can I change this setting in a persistent way especially when I
> >don't use Gnome or Kde or Xfce? I tried LxQt once and apparently the
> >stupid thing has reset some configuration and now xdg-open starts
> >Qupzilla which I don't like. Before, some other stupid program has reset
> >it to Chromium which I don't like either.
> 
> xdg-open examines your environment to determine what desktop environment (if
> any) you are invoking it under, and follows a DE-specific strategy if you're
> using one of the DEs it knows about (KDE, LXDE, XFCE, Gnome, Darwin, MATE).
> This is an entirely reasonable design decision, but also one that reasonable
> people can disagree with.

I didn't realize that it's plain shell script, thanks for mentioning
this below. I haven't even checked because I assumed that it's a program
using the common glib/kde framework functions to run url/mime handlers.

> If you are not using one of the desktop environments it knows about, then it
> uses its "generic" approach. For HTTP URLs, the generic approach is "check
> the BROWSER environment variable, then try the web browsers in a specified
> order if BROWSER isn't set". The first entry in that ordering is www-browser
> if DISPLAY is not set, and x-www-browser if DISPLAY *is* set.

What you describe is exactly what the sensible-browser script does. And
sensible-browser works fine for me, xdg-open does not.

What happens for me is that sensible-browser runs x-www-browser
(pointing at iceweasel) but xdg-open seems to try different fallbacks
first and finds some mime information about for http in
/usr/share//applications/mimeinfo.cache (not ok).

But anyhow, thanks for mentioning BROWSER variable. It seems to
influence glib functions too, so it's just what I need.

> xdg-open is written in POSIX shell, so you could easily copy it into a
> higher-priority location in your particular user account's PATH and modify
> it.

I will try to communicate this with the xdg-utils maintainer.
IMHO the script should respect the specifics of a system and on Debian
consider the configuration of x-www-browser alternative entry rather if
it couldn't detect a specific desktop environment. Or maybe do so only
if the alternative-link is a non-default setting.

Best regards,
Eduard.

Changing default XDB browser w/o GNOME etc.

[Eduard Bloch]

Hello, dear lazynet,

I have a problem and I am not sure which is the best approach to tackle
it.

Observation: various programs use the XDG mechanism (freedesktop
configuration method) to start a specific browser to visit a hyperlink.
Easy to verify with the xdg-open tool.

But how can I change this setting in a persistent way especially when I
don't use Gnome or Kde or Xfce? I tried LxQt once and apparently the
stupid thing has reset some configuration and now xdg-open starts
Qupzilla which I don't like. Before, some other stupid program has reset
it to Chromium which I don't like either.

How can I configure this, and do that in a way that is set persisently?
i.e. no program or package changes the setting to its own yard.

NOTE: I will not install the monster just to change this simple setting.

Feels like things that have been established in a sane way (alternatives
system and sensible-browser) are now reverted to the chaos we used to
manage in the nineties :-(

Regards,
Eduard.

Re: RealTek RTL8192EU drivers

[Eduard Bloch]

Hallo,
* Jose Martinez [Thu, Oct 29 2015, 02:38:04PM]:
> Yea, I had checked that site first, they don't have anything on the
> 8192EU.  In fact, their support for the 8192E is pretty spotty.  They
> don't have anything for the 8192EE either, which is the PCI version of
> the same chip.
> 
> I saw drivers for an 8192E and an 8192U both for windows, but nothing
> for linux.  There are a couple of other versions of the 8192 up there
> too, but nothing for the E series.

RTL8192EU/RTL8188EUS is supported by recent kernels with a driver from
the staging area. The driver has been in the mainline for a while, the
normal Debian kernel should support it. You also need to install the a
recent firmware-realtek package.

Regards,
Eduard.

-- 
 Das wurde vom User Joey angelegt, das ist keine
   Applikation, der darf das.

Re: Broadcom 43121 wlan issue

[Eduard Bloch]

Hallo,
* Himanshu Shekhar [Sun, Oct 04 2015, 04:09:55PM]:
> thanks dude ! i upgraded my driver. But can you tell why some packages
> break after updating to newer version and still debian stable version is
> named so.
> infact after updating java.. libreoffice4 failed and i had to install
> libreoffice5 from jessie backports.
> Is this normal or and issue?

This is not the whole truth. You DELIBERATELY installed a kernel from
backports, now you have to deal with consequences.

Regards,
Eduard.

-- 
Fällt der Pfarrer in den Mist, lacht der Bauer, bis er pißt.

Re: Broadcom 43121 wlan issue

[Eduard Bloch]

Hallo,
* Himanshu Shekhar [Sun, Oct 04 2015, 03:06:34PM]:
> I have no idea of what has happened to my system. My sources.list is
> attached along with make.log. If I need to downgrade my kernel, please
> provide a fix.
> Thanks for replying... I was waiting!

Downgrade the kernel or upgrade the driver, your choice.

https://packages.debian.org/sid/broadcom-sta-dkms
https://packages.debian.org/sid/broadcom-sta-common
https://packages.debian.org/sid/broadcom-sta-source

Regards,
Eduard.

Re: Extracting directories from an ISO image, command line tool?

[Eduard Bloch]

Hallo,
* Richard Owlett [Thu, Sep 10 2015, 09:18:18AM]:

> >if you need to automate it as non-root it looks like you can do that
> >with orrisox (part of the xorriso package):
> >
> >orrisox -indev /path/to/file.iso -extract . -subdir
> >/path/to/output/directory
> >
> 
> Browsing man page suggests that "-extract iso_rr_path disk_path" is
> explicitly what I'm looking for. Don't have time for a detailed read right
> now. Thank you.

You could also try isodump for single files.

Regards,
Eduard.

Re: apt-cacher-ng not expiring any more

[Eduard Bloch]

Hallo,
* David Wright [Tue, Aug 25 2015, 11:00:41AM]:
> Running apt-cacher-ng on wheezy (with wheezy-backports), some missing
> files in the repository are making the expiration step fail. Here are
> the relevant lines from the log:
> 
> Checking/Updating 
> debrep/dists/jessie-backports/main/binary-i386/Packages.bz2...
> 404 Not Found
...
> As I don't understand why there are Packages.bz2, Packages.xz and
> Packages.gz files variously scattered throughout debrep, I'm not sure
> whether this problem is a failure to maintain debrep or a bug in
> apt-cacher-ng (version 0.8.0-3~bpo70+2) whose fix hasn't been backported.

There is a newer BPO version where this problem should be fixed.

> Anyone else noticed this? If you're running under jessie, does it
> fail, does it skip those .bz2 files without complaining, or does it
> just use one of its companion files (.xz or .gz as available)?

They were removed from mirrors because, well, they are not necessary and
Jessie can work with .xz versions. However, apt-cacher-ng fails to
accomodate to this fact and will try to get the zombie versions. As
said, there is a workaround in a newer version.

Regards,
Eduard.

-- 
Das Positive am Skeptiker ist, daß er alles für möglich hält.
-- Thomas Mann

Re: how to debug this fuse problem

[Eduard Bloch]

Hallo,
* Gene Heskett [Thu, Mar 19 2015, 12:08:25PM]:

> Mode rant (with some profanity):

TLDR. Is this about fuse?

Eduard.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/20150319162912.ga13...@rotes76.wohnheim.uni-kl.de

Re: how to debug this fuse problem

[Eduard Bloch]

Hallo,
* Harry Putnam [Wed, Mar 18 2015, 07:41:21PM]:

> > Something is wrong with your modules installation. Run "dmesg" to see
> > what happened there.
> > Run "depmod -a" once, it might help if some stupid tool damaged the
> > module database (I had that "fun" with old dkms before).
> > Check the installed files (debsums linux-image-3.16.0-4-586)
> > Reinstall if needed (apt-get --reinstall install linux-image-3.16.0-4-586)
> 
> dmesg shows what looks like quite a mess.  I' don't know what it might
> mean, perhaps someone here has seen it before or understands what it
> means?
> 
> ,
> | dmesg|grep fuse
> | 
> | [7.166080] fuse: Unknown symbol pagecache_get_page_fixed (err 0)
> | [7.166143] fuse: disagrees about version of symbol iov_iter_get_pages
> | [7.166188] fuse: Unknown symbol iov_iter_get_pages (err -22)

Please proceed as I suggested before (debsums, depmod, etc.). If the
package state is OK then feel free to open a bug report (with
reportbug utility).

> Apparently it is a little more going on in those lines above.  I don't
> see anything on the page you cite with names like: 3.16.7-ckt4-3
> 3.16.7-ckt7-1
> 
> searching with `aptitudue search ckt7' (or ckt4)  finds nothing at all.

It's just an arbitrary version string. It can be (almost) anything, even
1.2.3.myCuteVersion-10.9.8.

> I suspect I could/should be running i686 with pae.  Not sure why the
> installer choose a 486 kernel originally ... (it changed to 586 during a
> full-upgrade a mnth or two ago)
> 
> The jessie OS here is running as vbox guest on a 2x Xeon on HP xw8600

I don't know why the installer chose that kernel, your Xeon should be
fully i686 compatible. If the installation is damaged somehow (see
above) then switching to another kernel might make the problem
disappear.

Regards,
Eduard.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/20150319160603.ga12...@rotes76.wohnheim.uni-kl.de

Re: how to debug this fuse problem

[Eduard Bloch]

Hallo,
* Harry Putnam [Wed, Mar 18 2015, 07:48:54AM]:
> Running `jessie'
> 
> I've been using encfs, drawn from wheeze for months.  But following a
> `full upgrade' I can no longer mount an encfs file system.
> 
> The error message from mount attempt:
> 
> ,
> | fuse: device not found, try 'modprobe fuse' first
> | fuse failed.  Common problems:
> |  - fuse kernel module not installed (modprobe fuse)
> |  - invalid options -- see usage message
> `
> 
> Attempting to modprobe fuse gets this error:
> 
> ,
> | root # modprobe fuse
> | modprobe: ERROR: could not insert 'fuse': Invalid argument

Something is wrong with your modules installation. Run "dmesg" to see
what happened there.
Run "depmod -a" once, it might help if some stupid tool damaged the
module database (I had that "fun" with old dkms before).
Check the installed files (debsums linux-image-3.16.0-4-586)
Reinstall if needed (apt-get --reinstall install linux-image-3.16.0-4-586)

> The upgrade log shows no upgrade for fuse at the last `full-upgrade'
> but does show something happening with the kernel. However, I don't
> understand what it means.
> 
> ,
> | [UPGRADE] linux-compiler-gcc-4.8-x86:i386 3.16.7-ckt4-3 -> 3.16.7-ckt7-1
> | [UPGRADE] linux-headers-3.16.0-4-586:i386 3.16.7-ckt4-3 -> 3.16.7-ckt7-1
> | [UPGRADE] linux-headers-3.16.0-4-common:i386 3.16.7-ckt4-3 -> 3.16.7-ckt7-1
> | [UPGRADE] linux-image-3.16.0-4-586:i386 3.16.7-ckt4-3 -> 3.16.7-ckt7-1
> | [UPGRADE] linux-libc-dev:i386 3.16.7-ckt4-3 -> 3.16.7-ckt7-1

It's just the kernel from Jessie. See
https://packages.debian.org/jessie/linux-image-3.16.0-4-586

Regards,
Eduard.

-- 
Every great idea is worthless without someone to do the work. --Neil Williams


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/20150318123238.ga9...@rotes76.wohnheim.uni-kl.de

Re: how to remove libsystemd0 from a live-running debian desktop system

[Eduard Bloch]

Hallo,
* Luke Kenneth Casson Leighton [Tue, Feb 17 2015, 04:28:04PM]:
> so to summarise:
> 
> * the use of libselinux1 is dormant (i.e. whilst you can't remove it
> without inconvenience, its use is entirely optional, right from the
> kernel level)
> * its development and documentation is rational and well-researched
> * the timeline behind its introduction was done in a respectful and
> reasonable way

> now let's compare that to the situation that we find ourselves in with
> libsystemd0:

Talk only about you, please. The "yourselves" of yours seems to lack
basic research skills.

> * the use of libsystemd0 is MANDATORY and EXCLUSIONARY (everywhere
> except slackware and FreeBSD)

Wrong. It is also dormant, except for a little function that
applications call.

> * its development is a moving target and the documentation of the
> roadmap is informal and sparse.

And that little piece of code had just six commits according to:
git log ./src/libsystemd/sd-daemon/sd-daemon.c | grep commit
and most of them are minor changes.

That's a HELL OF MOVING TARGET, yeah.  Now, enjoy the feel of being my
personal hero, take a cookie, STFU and go away. Thanks.

Deleted the rest of unfunded crap.

Regards,
Eduard.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/20150217170822.ga7...@rotes76.wohnheim.uni-kl.de

Re: how to remove libsystemd0 from a live-running debian desktop system

[Eduard Bloch]

Hallo,
* Lisi Reisz [Mon, Feb 16 2015, 11:42:14AM]:
> On Monday 16 February 2015 08:09:19 Marco d'Itri wrote:
> > > to debian-users: you don't have complete choice (yet), but i have
> > > demonstrated with a few hours work that there is a way to run
> > > (certain) desktop environments without requiring libsystemd0 or any of
> > > its dependencies, and after a little investigation there do appear to
> > > be people working hard to give you your right to choose what software
> > > to run *without* having to abandon debian.
> >
> > I strongly recommend that the people who cannot live with libsystemd0
> > installed on their systems leave Debian,
> 
> Oh, yes, please!  Please.  Please.  You have lots of choice.  There are lots 
> of distros and you can fork (devuan).  But leave us in peace!  It has been 
> lovely on this list without you lot.

+1, please!

(And if you still want to start a crusade against system libraries,
please try practicing first with more famous targets like libselinux1 or
even libpthread... nobody needs this dependency hell, right?)

Regards,
Eduard.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/20150216185543.ga18...@rotes76.wohnheim.uni-kl.de

Re: Debian fork: 'Devuan', Debian without Systemd

[Eduard Bloch]

Hallo,
* Erwan David [Wed, Dec 03 2014, 04:13:14PM]:
> > As explained several times on this ML, depending against libsystemd0
> > package doesn't mean anything about requiring systemd to be used as
> > PID1 or not. Even Ian's GR was not taking the "I don't want any systemd
> > package on my machine" use case into account you know.
> 
> Why focus on PID1 ? As I said, systemd-resolved proved to be

Because that set of "systemd depending packages" is BS, most of the
ones listed there are only linked with libsystemd0 and might attempt to
send a dbus message here and there, which only has some effect if and
only if systemd is PID1.

Bitching about not used features is like demonizing libselinux1.

Regards,
Eduard.

-- 
Erst wenn der letzte Programmierer eingesperrt und die letzte 
Idee patentiert ist, werdet ihr merken, daß Anwälte nicht programmieren können


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/20141205231359.gb25...@rotes76.wohnheim.uni-kl.de

Re: apt-get source download has files not in git repository

[Eduard Bloch]

Hallo,
* Joel Roth [Thu, Nov 20 2014, 10:05:08PM]:
> > >   pkgcache.apt
> > >   pkgcache.bin
> > >   restore
> > >   sources.list
> > >   sources.list.destdir
> > >   srcpkgcache.bin
> > 
> Hi Andrei,
> 
> > Are you sure these files are from dbus? I'd rather guess they are from 
> > the 'apt' source package. Anyway:
> 
> Yes, they are. I'm curious at which step they get generated.

Have a look at the timestamps which might provide a clue. I cannot see
any integration tests or similar activity config in the packaging so
it's hard to explain.

Also make sure you have checked out the right branch, see
debian/gbp.conf for details.

Regards,
Eduard.

-- 

 hallo ich möchte eine datei mit unrar entpacken, ich verwende
dazu den befehl "unrar -e *.rar", es kommt jedoch immer nur die liste
der optionen?
 wurst-griller: dann kannst du wohl nicht lesen
 haeufige fehlerursache...


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/20141123173344.ga27...@rotes76.wohnheim.uni-kl.de

Re: init scripts [was: If Not Systemd, then What?]

[Eduard Bloch]

Hallo,
* Miles Fidelman [Sun, Nov 16 2014, 02:41:14PM]:
> Andrei POPESCU wrote:
> >On Du, 16 nov 14, 11:50:25, Miles Fidelman wrote:
> >>So... with systemd, one has to:
> >>- rely on packagers to generate systemd service files, and/or,
> >>- rely on systemd's support for sysvinit scripts, which
> >>
> >>In the later case, one just has to read:
> >>http://www.freedesktop.org/wiki/Software/systemd/Incompatibilities/
> >>to get very, very scared
> >I don't see any item that would matter on a Debian system, care to
> >elaborate?
> >
> 
> It's very simple.  I have a bunch of stuff running on my system that I
> compiled from upstream code, including init scripts that work just fine.
> Now I have to worry that some of those scripts are incompatible with
> systemd.

You have already opened the Pandora box when you started relying on
self-compiled software which is installed aside of any knowlege of dpkg.

Now you hope to reduce your own efforts on upgrading by holding back any
progress. Is this fair? I don't think so.

If you don't see what I mean, think about all the other packages where
Debian does not promise backwards compatibility (Perl/Python/...
modules, etc, i.e. all that beasts that make a transition in big
clusters). You already have to read the release notes now, why should we
make an exception here for startup related parts?

Apart from that, have ever read those init scripts from upstream
packages or is this all this aversion just lazyness and fear / panic to
"not touch a running system" (but do a system upgrade, ha, ha, ha)?

Regards,
Eduard.


signature.asc
Description: Digital signature

Re: Joey Hess is out?

[Eduard Bloch]

Hallo,
* Nate Bargmann [Mon, Nov 17 2014, 07:46:48AM]:
> > Since systemd hatters usually fail on the second task, the rhetorical
> > arsenal is chosen accordingly (ad-hominem, trolling, misquoting, ...).
> Here is the crux of the problem in these discussions and that is the use
> of "haters" as an ad hominem as part of the shaming language directed at
> those with concerns.  It matters not who started it, but when one of the

There is no offensive of discriminatory meaning implied; you have read
my definition and that's how I apply it. Not more, not less.

> primary developers of systemd also engages in using this language to
> describe those who do have legitimate concerns, the discussion descends
> quickly into a non-productive direction especially when others follow
> his lead as being somehow acceptable.

If the final outcome is simply pidgeonholing into "those with us" vs.
"those against us" then I agree. But for somebody not participating in
this stupid holly wars about systemd this argument sounds somehow
far-fetched.

Regards,
Eduard.

-- 
 alpha welche distro hast du eigentlich???
-- gefragt im Channel #debian.de


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/20141117185804.ga9...@rotes76.wohnheim.uni-kl.de

Re: Joey Hess is out?

[Eduard Bloch]

Hallo,
* Slavko [Sat, Nov 08 2014, 04:13:12PM]:
> > Why don't the anti-systemd people do what they've been threatening the
> > whole time and fuck off to another distro or to FreeBSD?
> 
> This is exact example why i stopped all my contribution to Debian, and
> i will not start it again, despite if i stay with Debian in future or
> not! I orphan my packages in near future.
> 
> If the community consider people which have another opinion as bad, it
> is time do not contribute to it more. And whole debate is about one
> idea: If you don't like systemd, you are stupid.

That's not my impression. It's more like: if you don't like systemd,
don't use it. If you don't like it that much that you want to rant
against it, make sure your arguments have some proof/backup.

Since systemd hatters usually fail on the second task, the rhetorical
arsenal is chosen accordingly (ad-hominem, trolling, misquoting, ...).

Regards,
Eduard.


signature.asc
Description: Digital signature

Re: systemd - so much energy wasted in quarreling

[Eduard Bloch]

Hallo,
* st [Tue, Nov 11 2014, 02:01:46AM]:
> Hans wrote:
> 
> >And at the beginning things never work perfect
> 
> That's why they shouldn't make it into Stable as defaults,
> now should they?

We shouldn't and we are not. Because it's not "the beginning", systemd
already has multiple years of history behind.

I am wick of people whining about problems that systemd allegedly has but
that turn out to be a bug in something else when you dig deeper, or are
actually solved long ago. Sometimes this solutions are badly documented,
even so much that it hits my nerve, but apart from that systemd is
mature enough for inclusion as primary init system.

Regards,
Eduard.


signature.asc
Description: Digital signature

Re: OT - Convert output of byte count to GB count?

[Eduard Bloch]

Hallo,
* Bob Proulx [Sat, Feb 16 2013, 01:54:44PM]:
> Eduard Bloch wrote:
> > Nah, not having much spare time to post doesn't mean I have to drop
> > all the good habits.
> 
> I had nothing in the mailbox for the last year of recent memory.  I
> will call that good enough to be called "a while".
> 
> > Bob Proulx wrote:
> > > $ perl -le 'print 5605687296 / (1024*1024*1024)'
> > > ...
> > > For anyone who gets upset by the topic I will only offer this
> > > following treatise as highly recommended reading.
> > > 
> > >   On Holy Wars and a Plea for Peace
> > >   http://www.ietf.org/rfc/ien/ien137.txt
> > 
> > And there is it again :-( Please don't justify pure misuse of
> > terminology with false analogies like this document about Holy Wars at
> > absolutely equivalent things (equivalent WRT their application).
> 
> Please quote my exact words where I misused terminology.  I do not see
> it.  I never said GB nor GiB.  I used a number.

Yeah. A number as answer to something about "GB". Have you ever been on
the Jeopardy show?

> Because of the fervor on this topic I believe the Holy Wars article is
> very obviously as relevant now as then.  It is talking about human
> nature.  As was the original Swift.  As proven by this thread human
> nature hasn't changed any in all of these years.

What is so unclear in what I have written before? In Gulliver's travels,
as in most holy wars, the outcome is absolutely identical, so choosing
any method doesn't make a difference. This analogy does NOT apply here.

Following your logics means that it's ok to use "pint" for business
communication because it's a soo well defined unit and anyone
understands how much it is. Now tell that to a globally operating
company...

> > You can prefer whatever you want but the means of reliable
> > communication are unambiguous terms. EOD.
> 
> The number 1024 that I used is not ambiguous.  It is an exact value.

Sure. But please, never use it to refer to any unit name using a SI
prefix.

> > Some dudes in nineteen-seventies didn't get it and another generation
> > of dudes is still trying to protect those "values" even perfectly
> > knowing they are wrong.
> 
> I used a number 1024 as a divisor for binary data and I don't think I

Responding to someone asking for "GB".

> When I am working with binary data I use binary numbers.  When I am
> working with decimal data then I use decimal numbers.  This is neither

And when exactly does the data stop being decimal and becomes binary?
When it's read from the the PHY buffer and written to the 2nd level
cache of your CPU? Sorry, that kind of hairsplitting is just
ridiculous.

> Others are of course free to use their own preferences.  Many people
> still prefer octal representations.  Such as with chmod.  I typically

Yet another false analogy. You keep presenting more and more of them,
always implying that somebody ...

> eschew use of octal values with chmod and prefer the modern purely
> symbolic modes.  But I don't declare others wrong for using octal
> values with chmod.  At times it is also convenient to the data it is
> describing.

... revealing them as not applicable here is "declaring others wrong".

A better analogy would be like telling compiler makers to watch
out for the special word "chmod". If present and the programmer writes
chmod(..., 755) they should not handle it like chmod(..., 0755) because
that's what looks convinient for many people. Oh, and please, the
compiler should guess the meaning based on the count of *.c files in the
home directory because that's a good indication for being used by an
expert or newbie.

> > In the field of arithmetics the effects of such "misnomers" can become
> > fatal. For example, see the US speciality called "billion" which has a
> > custom meaning incompatible to the rest of the world.
> 
> I do not understand your point with regards to "billion".  Please
> explain further.
> 
> I am aware that in the old days (prior to the 1970's?) for some
> countries (primarily the British Commonwealth?) a billion was a
> million million (10^12) and a trillion was a million million million
> (10^18).  Those derived from "bi" and "tri" meaning million^2 and
> million^3.  Which makes sense.  But I believe that now common English
> communication throughout the US, UK, and Commonwealth it is now
> considered obsolete usage.  Today a billion is 10 (10^9) aka a
> thousand million and a trillion is 1 (10^12) aka a
> thousand thousand million.

There is a long story ("short and lo

Re: OT - Convert output of byte count to GB count?

[Eduard Bloch]

Hallo,
* Jerry Stuckle [Sat, Feb 16 2013, 01:36:42PM]:

> >>>>On 2/16/2013 5:43 AM, Eduard Bloch wrote:
> >>>>>In fact, the main reason for becoming a drama is the impression that
> >>>>>numbers have on some people's mind. "I bought a 32 gig stick and Windows
> >>>>>says it's not even 30 gig on it... that's FRAUD!!"
> >>>
> >>>>Sure it has 32GB on it.  But that doesn't mean 32GB is available for
> >>>>your files.  There is OS overhead, also.
> >>>
> >>>Bullshit. Overhead does not consume 7% of the disk capacity. This is
> >>>just the difference between 1 GiB and 1 GB.

Funny. I made an example to demonstrate the irrational
effects that numbers have on people, and you both immediately start a
holy war based on it.

> >>Overhead can easily take up 7% of the disk, depending on the file system
> >>being used.
> >
> >Let me rephrase it. Can you provide an example of a freshly created
> >filesystem of a common size with an overhead of at least 7%, i.e. with
> >the free space being less than 93 % of the raw size ?
> >
> >
> 
> How about answering my question.  Exactly which disk do you have

I am with him. There is no real question from you that I could discover
so far, just some attempts to distract attention after not understanding
the point (or even the content) of the original example.

Regards,
Eduard.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20130216204019.gb15...@rotes76.wohnheim.uni-kl.de

Re: OT - Convert output of byte count to GB count?

[Eduard Bloch]

Hallo,
* Jerry Stuckle [Sat, Feb 16 2013, 09:09:42AM]:
> >And there is it again :-( Please don't justify pure misuse of
> >terminology with false analogies like this document about Holy Wars at
> >absolutely equivalent things (equivalent WRT their application).
> >
> >In the field of arithmetics the effects of such "misnomers" can become
> >fatal. For example, see the US speciality called "billion" which has a
> >custom meaning incompatible to the rest of the world.

> I agree, I wish people WOULD stop applying terms in the incorrect
> context.  That's what confuses matters.

Which part do you agree with? I cannot see any. You keep insisting on
being right with the incorrect use of other people's words, telling
them that THEY have to rethink and adapt the meaning every time they
need to communicate with you in your little sandbox.

Reminds me on doublespeak and doublethink (google it if you don't know
what I mean).

Regards,
Eduard.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20130216195405.ga15...@rotes76.wohnheim.uni-kl.de

Re: OT - Convert output of byte count to GB count?

[Eduard Bloch]

Hallo,
* Lisi Reisz [Sat, Feb 16 2013, 08:20:45AM]:
> On Saturday 16 February 2013 03:20:54 Jerry Stuckle wrote:
> > When dealing with computers, it's powers of 2.  When
> > dealing with distances, it's powers of 10.
> 
> Not so.  Manufacturers of hard drives normally (frequently?) give the size in 
> decimal, though they obviously don't say so, to make them look bigger.

Actually, they do say so. Most harddisks I have seen so far have the
explanation printed on them or an explicite reference to the
documentation where they tell you that.

In fact, the main reason for becoming a drama is the impression that
numbers have on some people's mind. "I bought a 32 gig stick and Windows
says it's not even 30 gig on it... that's FRAUD!!"

Regards,
Eduard.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20130216104351.gb17...@rotes76.wohnheim.uni-kl.de

Re: OT - Convert output of byte count to GB count?

[Eduard Bloch]

Hallo,
* Bob Proulx [Sun, Feb 10 2013, 11:46:30AM]:
> Eduard Bloch wrote:
> > Hallo,
> > * Bob Proulx [Sat, Feb 09 2013, 11:44:03AM]:
> > > You say "tomato", I say "tomahto".  You can make your own divider the
> > > number 34,969[1] if you like.
> > 
> > Wow, someone trying to disguise the own mistaken convention as pure
> > matter of preference?
> > 
> > > For me I prefer binary powers of two when using it in relation to
> > > computer tasks.
> > 
> > You can prefer whatever you want but the means of reliable communication
> > are unambiguous terms. EOD.
> > 
> > Some dudes in nineteen-seventies didn't get it and another generation of
> > dudes is still trying to protect those "values" even perfectly knowing
> > they are wrong.
> 
> Wow.  My response pulled you in to post on debian-user!  It has been a
> while.  Good to see you here.

Nah, not having much spare time to post doesn't mean I have to drop all
the good habits.

> For anyone who gets upset by the topic I will only offer this
> following treatise as highly recommended reading.
> 
>   On Holy Wars and a Plea for Peace
>   http://www.ietf.org/rfc/ien/ien137.txt

And there is it again :-( Please don't justify pure misuse of
terminology with false analogies like this document about Holy Wars at
absolutely equivalent things (equivalent WRT their application).

In the field of arithmetics the effects of such "misnomers" can become
fatal. For example, see the US speciality called "billion" which has a
custom meaning incompatible to the rest of the world.

Regards,
Eduard.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20130216102816.ga17...@rotes76.wohnheim.uni-kl.de

Re: OT - Convert output of byte count to GB count?

[Eduard Bloch]

Hallo,
* Bob Proulx [Sat, Feb 09 2013, 11:44:03AM]:
> Pascal Hambourg wrote:
> > Bob Proulx a écrit :
> > >   $ perl -le 'print 5605687296 / (1024*1024*1024)'
> > 
> > Note that 1 GB (gigabyte) is 10^9 bytes as can be seen in the GNU dd
> > output. 1024^3 is 1 GiB (gibibyte).
> 
> You say "tomato", I say "tomahto".  You can make your own divider the
> number 34,969[1] if you like.

Wow, someone trying to disguise the own mistaken convention as pure
matter of preference?

> For me I prefer binary powers of two when using it in relation to
> computer tasks.

You can prefer whatever you want but the means of reliable communication
are unambiguous terms. EOD.

Some dudes in nineteen-seventies didn't get it and another generation of
dudes is still trying to protect those "values" even perfectly knowing
they are wrong.

Regards,
Eduard.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20130210105658.ga6...@rotes76.wohnheim.uni-kl.de

Re: 2TB file system

[Eduard Bloch]

#include 
* Camaleón [Tue, Aug 16 2011, 05:30:07PM]:
> On Tue, 16 Aug 2011 12:31:53 -0400, Rick Pasotto wrote:
> 
> > I recently acquired a 2TB SATA HD that I have not yet installed. It will
> > be used entirely to store media files. Would there be any problems in
^

> > Any other considerations?
> 
> Hum... Reconsider partitioning or even reconsider the file system.

That really depends on what "media files" means. It can be anything,
from 2MiB music files to 200GiB raw video streams.

For many, many thousands of small files, smaller partitions make sense.
But for other cases... is it really worth it?

> A single volume of 2 TiB is very big (and big file systems are more prone 
> to errors and hard to recover in the event of a corruption... fsck can 
> take... ages? :-P). 

You blatantly ignore the real question... why would fsck take ages?
Because of the file count or because of the overall data size?

> Also, if you plan to host specifically multimedia files you could 
> consider using XFS instead, I've been told is very good for such 
> purpose ;-)

I have been told many things, and some of them aren't something you
should believe if you want to stay sane and healthy.

Regards,
Eduard.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20110816231132.ga28...@rotes76.wohnheim.uni-kl.de

Re: Installing Debian from NFS

[Eduard Bloch]

#include 
* RR [Fri, Feb 04 2011, 10:19:06PM]:

> helps build one a local mirror. It's been going on for almost 20 hrs and
> it's downloaded 28GB so far after I used a LOT of EXCLUDEs but oh well. Was
> a Bad move in a way since Squeeze is going to be out soon and then I'll have
> to build a mirror for that as well. I'm probably stupid or impatient but I
> didn't understand the concept of apt-cacher-ng from a brief glimpse. It's a
> HTTP proxy to a local mirror or some other mirror or does it make an NFS
> partition accessible via HTTP? Since it uses the word "cache" in it, I'm
> assuming, it caches the packages so everytime you need something you get it
> locally than reaching out to the Internet mirror?

Yes, it's a tiny HTTP proxy with some sugar, getting extra knowledge
from package databases or predefined configuration and making more
optimal decissions than ordinary HTTP proxies.

If you decide to use it, you could reuse the data you already
downloaded. There are instructions for integration into the acng cache
in /usr/share/doc/apt-cacher-ng/html/howtos.html . And if you want to
save some disk space and your NFS storage is permanently mounted, then
you could even put symlinks into the import directory (cp -asf
/download/... /var/cache/apt-cacher-ng/import/).

And if you still want to have a local snapshot of the whole Debian
archive in order to avoid future internet connections, there is a new
feature in recent versions called "Guided precaching". That function
actually examines some Packages.bz2/Sources.bz2 files (or whatever you
specify as long as the file is parseable by acng) and downloads
everything listed therein. Usually the resulting tree in the cache
directory is defacto a partial mirror, containing files which apt might
need for a certain architecture.

Regards,
Eduard.
-- 
 Und nun sehen sie das alphascorpii in der freien Wildbahn...
 Es umschleicht den Malzkaffe, sein natürliches Beutetier..


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20110205120952.ga11...@rotes76.wohnheim.uni-kl.de

Re: [SOLVED] Is squeeze compatible with WD20EARS and other 2TB drives?

[Eduard Bloch]

#include 
* Stan Hoeppner [Sun, Jan 16 2011, 01:22:28PM]:

> > of native 4k sectors used
> > with a smaller transfer size. But the whole system programing domain has
> > tons of similar situations.
> 
> Transfer size?  SAS and SATA are both capable of large multi sector transfers.
> This has nothing to do with "transfer size", but SECTOR size.  It seems you 
> lack
> basic understanding of the technology we're discussing.

Do you realize that "transfer size" is an abstract term? Doesn't look so
because you obviously fail to make the connection to the subsequent
example. And sorry for crushing your theory but you are wrong; usually I
wasn't sleeping in related CS lectures.

> Me stating that these drives suck with Linux (fact) is, to some owners of
> said drives, apparently akin to me calling their sister fat or mother ugly.

And if you are looking for things that suck then remember the bad
old days (if you were with us at that time). When you needed to install
emulation software like Ontrack to make your harddrive work with DOS and
then patch the Linux kernel to get along with it. Or had to write
modelines manually because other the monitor might go up in smoke. Or
needed to make experiments with hdparm to find the fastest Multiword-DMA
mode that still worked with your HDD. Etcetera, etcetera.

The "problems" with AF are a joke.

Regards,
Eduard.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20110120001911.ga22...@rotes76.wohnheim.uni-kl.de

Re: [SOLVED] Is squeeze compatible with WD20EARS and other 2TB drives?

[Eduard Bloch]

#include 
* Stan Hoeppner [Wed, Jan 12 2011, 10:28:40AM]:
> Stefan Monnier put forth on 1/11/2011 9:46 PM:
> >>> I have no idea what makes you so angry against "green" drives.
> >> I am against using any drive, at this time, in Linux, with a native
> >> sector size other than 512 bytes.
> > 
> > Again, I fail to see why you're so emotional about it.
> 
> You've got that backwards.  You own a bunch of these drives and feel I am

That's not my impression. The only good argument I could identify in
YOUR answers is the strong (personal) dislike of native 4k sectors used
with a smaller transfer size. But the whole system programing domain has
tons of similar situations.

Why don't you object to 32bit/64bit PCs? They work with small transfer
bytes (bytes) but they native WORD size is a lot bigger nowadays. How
dare they to do a such thing?

Maybe you should consider throwing your PC away right now and buy a
modern 8088 XT system from some garage sale.

Regards,
Eduard.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20110116115239.ga10...@rotes76.wohnheim.uni-kl.de

Re: [SOLVED] Is squeeze compatible with WD20EARS and other 2TB drives?

[Eduard Bloch]

#include 
* Klistvud [Sat, Dec 18 2010, 10:32:10PM]:

> First of all, let me thank all of you who responded. As promised, I
> am giving feedback to the list so that future purchasers of Western
> Digital WD EARS/EADS models and similar "Advanced Format" hard
> drives may benefit.

Err, what? EADS don't use AF, TTBOMK.

> The first thing of notice is that the Load_Cycle_Count of the drive
> heads increases every 8 seconds by default. As seen on the Internet,

That only refers to EARS. And it's wrong. Umount everything on that disk
and wait a while, no loading/unloading should stop.

What really happens is that the disk parks after 8 seconds when it's
IDLE. Which is ok when you either read or write stuff all the time or
don't do anything at all. It is not ok if you use them as system disks
where a few bytes are written every couple of seconds and certain
popular Linux filesystems like to flush (means: write out to disk) that
data every 10..15 seconds (just a bit more than 8 seconds) and so
causing the LCC growing quite quickly over time.

So DO NOT use an EARS drive as SYSTEM DISK.

> this may pose a problem in the long run, since these drives are
> "guaranteed" to sustain a limited number of such head parking
> cycles. The number given varies from 300.000 to 1.000.000, depending
> on where you look. The first thing I did was, therefore, launch a

There is no reason to put the word guaranteed into double-quotes or
refer to weird sites. Just have a look at the official data sheet and
the common definition of MTBF please.

> the WD proprietary utility wdidle3.exe, and the first link obtained
> by googling for "wdidle3.exe" did the trick:
> http://support.wdc.com/product/download.asp?groupid=609&sid=113
...
> thousand ticks overnight. Interestingly enough, the drive loaded and
> unloaded its heads at the amazing rate of twice per second even
...
> "disabled" to "every 300 seconds", which appears to be the maximum
> interval allowed. It would seem that, for the time being at least,
> this made the Load_Cycle_Count stay put at 22413. Whew!

Err, what? You play with a dangerous toy which was not designed for your
drive and you wonder that it's all messed up now?

> Now, the second issue: the hardware/logical sector alignment.
> Since it will affects real-world transfer speeds, let's first check
> out the theoretical speeds of this drive in this particular
> environment -- a 3GHz Pentium-IV motherboard with a humble
> integrated SATA controller (I think it's an early SATA-I
> generation).

My company had a lot of them. The SATA controllers were crap
performance-wise. I remember a colleague who got a shiny new OCZ SSD
drive which was supposed to deliver >>200MB/s but never got beyond
70MB/s on his system. The solution was a 15EUR PCIe controller card
which suddenly make it work as expected. 

> obelix# hdparm -tT /dev/sda

Err, what does this have to do with pro/contra of logical sector sizes?
Counter-example, WD20EARS on an AMD-78xx mainboard:

/dev/sdc:
 Timing cached reads:   8042 MB in  2.00 seconds = 4023.46 MB/sec
 Timing buffered disk reads: 360 MB in  3.02 seconds = 119.38 MB/sec



Eduard.

-- 
Naja, Garbage Collector eben. Holt den Müll sogar vom Himmel.
   (Heise Trollforum über Java in der Flugzeugsteuerung)


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20101219224217.ga5...@rotes76.wohnheim.uni-kl.de

Re: which package can show CPU temperature and fan speed

[Eduard Bloch]

#include 
* Long Wind [Sun, Apr 18 2010, 01:24:09AM]:
> I have etch and HP VL420
> It runs P4 1.6G
> but BIOS can't show CPU temperature and fan speed
> so I download a bios from motherboard maker ASUS
> but can't update bios
> then I install lm-sensor in etch
> but reading data seems incorrect
> Can you help???

If the kernel is fresh enough and ships with all required modules then
you need to run sensors-detect as root. It should detect suitable
drivers and load the modules. It would also ask you to add the list to
/etc/modules which you should confirm unless you want to run the
detection again after reboot.

Regards,
Eduard.

-- 
 TV ist Zeitverschwendung
* micsch muss auch nochmal seine TV-Karte einbauen


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20100418094347.ga13...@rotes76.wohnheim.uni-kl.de

Re: Vfat or NTFS?

[Eduard Bloch]

#include 
* Roger Leigh [Sun, Mar 07 2010, 07:20:04PM]:

> > I am wondering now whether to convert my Vfat partitions to NTFS as there
> > are some advantages. For example I recently forgot about the 4GB file
> > restriction of Vfat when trying to download a DVD iso - it got to 4GB then
> > gave up so I had to redownload again to the NTFS volume. I suspect NTFS is
> > more reliable at least in Windows. Are there any potential issues in Linux -
> > e.g. reliability / speed.
> > 
> > Should I go NTFS now for my data files? (keeping the main Lenny root
> > filesystem on an ext3 partition).
> 
> Just my experience but:
> 
> 1) VFAT is a terrible filesystem with all sorts of restrictions, not least
>filesize and encoding issues.

The worst problem with VFAT is IMHO the stupid behavior WRT letter case,
i.e. you get some other filename than you originally intended to store
to which confuses tools like rsync. The encoding issues are usually not
that bad.

> 2) NTFS (using ntfs-3g) performs terribly on Linux.  The FUSE driver is

I disagree. I get about half of the raw speed of the drive from a
Windows system partition. Ext3 is not much better on a Linux system
partition. There were, however, combinations of kernel versions and
ntfs-3g versions where performance really sucked.

>slow, and this makes data transfer a very lengthy affair.  No idea

Cannot agree on that either. FUSE driver needs to copy data around and
therefore having some spare CPU power helps a lot and many people have
multi-core CPUs nowadays.

Regards,
Eduard.

-- 
 hat einer von euch schon bind9 installiert?
 das neue root kit? :->


signature.asc
Description: Digital signature

Re: SOLVED aumix not working

[Eduard Bloch]

#include 
* Martin [Sun, Mar 07 2010, 05:01:19AM]:
> On Sat, Mar 06, 2010 at 03:03:30PM -0800, Freeman wrote:
> > On Sat, Mar 06, 2010 at 08:55:52PM +0100, MList wrote:
> > > Why is aumix failing and how can I fix it?
> > > 
> > 
> > Have you read:
> > 
> > Debian specific notes for the aumix package
> > ===
> > 
> 
> 
> After lot of blind guesses I finaly try command
> from amixer man page:
> $ amixer -c 0 -- sset Master playback -20dB
> after this one command aumix can work now as expected.

That shouldn't be required unless snd_mixer_oss was not loaded before
and amixer just triggered its loading somehow (see quoted README for
details). Or maybe it's some other low-level module. You can compare the
module lists before and after running that command (visible with "lsmod"
command or in /proc/modules).

> I guess that amixer command created /dev/mixer device so aumix can
> find it now. Why was /dev/mixer was not created during instalation
> process I do not know!

They are usually created by udev which is triggered by *bingo* module
loading.

Regards,
Eduard.

-- 
 zpiggy: wo hast du den nick her?
 ach dat bin ja ich


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20100307084613.ga5...@rotes76.wohnheim.uni-kl.de

Re: how to find bad blocks

[Eduard Bloch]

#include 
* James Zuelow [Tue, Feb 09 2010, 08:58:15AM]:

> Can you get SMART data from it?
> 
> I don't think you can list the bad blocks that the drive is re-mapping, but 
> you can certainly get the number of times the drive has had to do that.
> 
> The manpage for smartctl describes how to query individual disks behind 
> various RAID controllers.

And if that doesn't work then you can use dd_rescue to copy the data to
another harddisk and store the list of bad blocks on-the-fly.

If you need to know which files are damaged, in many cases you can later
use that bad block list to retrieve the file list with
debug / administrative tools for particular file system.

Regards,
Eduard.

-- 
 Jeder, der mich verwirrt, muss einen oeffentlichen Security-Bug fixen
und einen Absatz fuer's DWN schreiben.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: apt-get files transfering to another machine.

[Eduard Bloch]

#include 
* Sthu Deus [Tue, Jan 19 2010, 02:22:38AM]:
> Good day.
> 
> 
> Which files I should copy from one machine to another after
> 
> apt-get update
> 
> command lunched at one machine so that another machine might have the same 
> info as if I lunched the command in it?

For the same architecture?

rsync -a /var/cache/apt r...@host:/var/cache

This will also include package files. If that's not wanted then only
transfer /var/cache/apt/*.bin .

Regards,
Eduard.
-- 
 jjFux: Ted hieß ja früher auch Walther
 winkiller: hm... es sind 8... die 7 kandidaten und NOTA
 Ist der jetzt eigentlich eine gespaltene Persönlichkeit, bei der aber
  beide Teile bekloppt sind?


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: [ANNOUNCE] apt-offline 0.9.6 released

[Eduard Bloch]

#include 
* Ritesh Raj Sarraf [Tue, Dec 22 2009, 01:24:13PM]:

> I'm pleased to announce apt-offline version 0.9.6.

Wait a second, unless I am mistaken it's an implementation of the
traditional "Using APT Offline" howto from Jason Gunthorpe, this time
with additional requirement of Python.

If yes, maybe you should tell this upfront?

> Please see the git log for full details.

Where is it?

> apt-offline is an Offline APT Package Manager for APT based systems, i.e. 
> mostly 
> Debian and Debian derived distributions.

And what does "managing the packages" mean in this context? Maybe you
mean: download manager for package files?

> apt-offline can fully update/upgrade your disconnected Debian box without the 
> need of connecting it to the network.
> 
> * It works by generating a signature of the disconnected box.

Is that a signature? A signature is something short you write down with
a pen for identification purposes, basically equivalent to a short
checksum in data processing.

AFAICS what you use is a plain flat database, i.e. an URL list generated
by apt.

> * You carry the signature to a net connected box running anything 
> (Windows/Linux/Mac)

And the system requirements on that net connected box are... what?
That's an important detail, IMHO.

Regards,
Eduard.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: IDE CD-DVD Writer won't write

[Eduard Bloch]

#include 
* Merciadri Luca [Sun, Dec 13 2009, 07:48:59PM]:

> pata_marvell4256  0 

I think you have some third-party IDE controller on the mainboard
(with a Marvell chip) which has a faulty driver. The problem is not
uncommon, i.e. if the the driver author only tested with "even" block
sizes like 512 and 2048 and not those used for burning commands then
there may be bugs.

> > dmesg
> see last message please

Where? AFAICS you only posted some console log with wodim calls so far.

So which kernel version are you using? Consider upgrading.

Regards,
Eduard.

-- 
Der Mensch selbst ist die Killer-Application.
-- Hal Faber


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: IDE CD-DVD Writer won't write

[Eduard Bloch]

#include 
* Merciadri Luca [Sun, Dec 13 2009, 03:58:37PM]:

> I thought that I had specified it before, but, according to your
> message, I did not explain this in an explicit way.
> 
> Under Windows (XP), I am able to read, and write, on any CD or DVD
> medium, whatever the writer. I am thus fully able to use the
> Linux-problematic CD-DVD-writer under Windows XP. That proves that
> cables and related stuff are correctly done. That is the main problem:
> Linux cannot make it, while Windows can make it, for once.

Relax. Then I suggest testing a closed-source alternative similar to
what you may know from Windows experience: Nero Linux 4,
http://www.nerolinux.de/. This way you can exclude kernel issues as
potential problem.

Regards,
Eduard.

-- 
 Halloechen, ihr Spinner, so frueh auf?
 nein, wir schlafen alle im kollektiv
 mein alkoven ist kaputt
 alkohol kaputt?


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: Is there an IM Client that Does Not Use X

[Eduard Bloch]

#include 
* Martin McCormick [Fri, May 15 2009, 02:39:31PM]:
> Is there any sort of instant message application that can run in
> a command-line terminal, similar to talkd?

Not very similar but good:

apt-cache show centerim

-- 
Wer den Menschen die Hölle auf Erden bereiten will, braucht ihnen nur
alles zu erlauben.
-- Graham Henry Greene


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: stitching a filesystem

[Eduard Bloch]

#include 
* Mag Gam [Tue, Mar 17 2009, 07:22:13PM]:
> Is it possible to combine 2 filesystems so they would appear as one?
> 
> For example, you have 2 volumes:
> /vol0 (500GB)
> /vol1 (500GB)
> 
> I want to export both of them so it would appear as 1TB volume.

Kind of stitching (the size reporting might confuse but the contents
appear merged):

apt-cache show funionfs (FUSE variant)
apt-cache search unionfs (kernel implementations, aufs and unionfs modules)

"man funionfs" will tell you more.

Regards,
Eduard.

-- 
Einschlafen dürfen, wenn man müde ist, und eine Last fallen lassen 
dürfen, die man sehr lang getragen hat, das ist eine köstliche, eine
wunderbare Sache.
-- Hermann Hesse


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: nrg2iso

[Eduard Bloch]

#include 
* steef [Fri, Dec 19 2008, 12:58:35PM]:

> this is the command i used to convert a (nero)file into a readable iso:

Why don't you just NeroLinux then?

> *sudo nrg2iso /home/steef/Desktop/PF2.nrg /home/steef/Desktop/PF3.iso
> it seemed to work,
> and i burnt the result, PF3.iso, onto a dvd
> with # growisofs -dvd-compat -Z /dev/scd0:/home/steef/Desktop/PF3.iso

Try this:

mount foo.nrg /bla -oloop,offset=307200

If you can read the files in /bla directory then you can try to burn
from the intermediate loop device with a command like

wodim -dev=/dev/scd0 -isosize /dev/loopX

You can identify your loop device via "mount | grep loop". Or RTFM of
the losetup command if you don't like the mount command above.

But, as said, I would just use NeroLinux.

> no readable file on the dvd results (tried it three times on a re-writable)

What does "no readable" mean? Can it be mounted?

Regards,
Eduard.

-- 
[ New Maintainer Prozess ]
 ein jahr ist ein bisschen zu optimistisch,
<_rene_> panthera: kommt auf den NM/AM an.
/* _rene_ ist pantheras AM und lässt sich mit pantheras
   package check schon ein wenig Zeit ;) */


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: .bash_history deleted

[Eduard Bloch]

> Hi Anna.
> 
> Anna Machens, 08.07.2007 15:48:
> >  (i'm using debian unstable )
> >  i am totally confused. suddenly, my .bash_history file for root did not
> >  exist
> >  anymore.
> 
> That’s really odd; I experienced the same thing yesterday. Suddenly
> there was no
> history anymore but since then, the file has been growing normally like
> before.

I have seen this several times in the past, still having no idea how this 
happens. For a while I suspected bash_completion but I think I saw this also 
without bash_completion. Finally I am suspecting the filesystem journaling 
doing weird things, since it it mostly happened after crashes with JFS and XFS, 
but (almost? not sure) never with Ext3.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: wireless - is it possible?

[Eduard Bloch]

#include 
* Default User [Sat, Apr 28 2007, 04:04:41PM]:
> Easy question: has anyone been able to get any currently available wifi
> USB or PCMCIA adapter to work with a 2004 Toshiba Satellite M35X-S109
> laptop (no built-in wifi), using Debian Etch? 
> 
> Very hard question: if so, how were you able to do it? 
> 
> I have tried a Netgear MA111 (v1?) USB adapter. I have only been able to
> get it to work "out of the box" with OpenBSD (any recent distribution),
> and Freespire 1.0.13 (before the switch to an Ubuntu base). 
> 
> I have also tried a Netgear WG111 (v2) adapter.  To date, it won't work
> "out of the box" with any of many non-M$ OSes tried.  

We tried several sticks at my previous university and most of them
sucked. Some of those Netgear have been supported by the free drivers,
other (v1 rev. 2 or so) have not been supported anymore but ndiswrapper
did the trick, but some has been working and some didn't, and IIRC none
of v2 did work.

For ralink based ones, I have a mixed feeling. An old ASUS stick did
work with an old driver, but a more recent driver screwed up something
in the kernel and made it first loose the keyboard control and then
crash, after just replugging the stick two or three times. A newer
stick with Ralink chipset (IIRC from D-Link) was not really supported
either: driver loads, the interface is created but no data is transfered
after that.

Well, and the best experience we had was with the cheapest stick from
MSI having some Zydas chipset (IIRC MSI US54SE). Stick inserted, WEP or
WPA configured, WORKS! They are still not perfect, in a certain
combination (old Cisco AP and WEP mode) the data transfers got stuck
after few minutes and the stick has been restarted (messages about
reseting the SoftMAC in the kernel log). However, there was no such
trouble with in plain or WPA mode, IIRC.

Regards,
Eduard.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: console based cd burning (wodim and burn)

[Eduard Bloch]

#include 
* Joerg Schilling [Thu, Mar 15 2007, 09:03:15PM]:

> Some Debian people did create this fork because they do not like to cooperate
> with the author in a fruitful way.

"fruitful" from our point of view implies fulfilling user requirements.
If you do not listen to them or are not ready to make compromises then
the cooperation based on solely on your terms is not fruitful.

> This is not related to license problems but to personal deficites from these 
> people.

Really? We have been modifying cdrtools for years before. I wonder how
you can overlook the correlation of the license change and the final
fork triggering condition.

> Unfortunately Debian does not seem to be interested in doing the best for 
> their

"does not seem" for whom actually?

> users but to avoid replacing the people in question by people who are 
> intrested
> in a fuitful cooperation.

Grats, well chosen words to present the others as the source of trouble.
To clean up with your assumptions: there is no need to replace people
to put "pure cdrtools" packages into the archive. It just needs a
volunteer to package them and deal with you. If nobody from the hundreds
of maintainers is interested in a "fruitful cooperation" than it may
have some other reason.

Eduard.
-- 
Naja, Garbage Collector eben. Holt den Müll sogar vom Himmel.
   (Heise Trollforum über Java in der Flugzeugsteuerung)


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: console based cd burning (wodim and burn)

[Eduard Bloch]

#include 
* Joerg Schilling [Thu, Mar 15 2007, 05:57:30PM]:

> >trying to see what should I use to burn CD's from a terminal. This is on 
> >an Etch bases system with only icewm installed on it. I don't want to 
> >use nautilus cd burner and neither k3b, if I can help it.
> 
> >How do wodim and burn compare for such a task? Any experiences with 
> >these utilities? I am trying to decide which one to start and keep using 
> >depending on its stability and features. And it would be great to have 
> >the option of burning audio cd's with text information (never been able 
> >to do so with k3b, even with checking the cd-text option).
> 
> Wodim is based on _parts_ of an very outdated version of cdrtools.

Please stop throwing subjective assesment into objective discussion.
I.e. tell your definition of "very" and do it exactly. And please don't
tell weird things like "contains parts that are 2 years old". "Parts" of
the current Linux kernel may be 10 years old, so what...

> Wodim does not include the DVD burning code from cdrecord and it does not

Of course it does not. Is that a reason to be much worse? I doubt.

> include dozen of bugfixes mainly in mkisofs that have been applied during 

Since those are under GPL there should be not a big problem to pull them
from the current cdrtools. Or would you dare to forbid this?

> the past 12 months. As an important issue, wodum does not use the new 
> enhanced 

"enhanced", how? As above, stop throwing subjective assesment into
objective discussion.

> version of libscg that tries to reintegrate the unwanted /dev/hd* interface 
> for 
> SCSI into the rest of the SCSI namespace from Linux.

Funny. I maybe will start believing your words when -scanbus does
actually present the native Linux namespace and not before.

> If you use the Debian surrogate instead of the original, you suffer from the 
> following problems:
>
> - No support for hardlinks in Rock Ridge

Minor bug. They have diverging inode numbers and wrong link counts, but
it is not important in usual use cases because the filesystem is
read-only.

> - Incorrect permissions for many of the ".." directory entries,
>   so "cd .." wo'nt work.

Unreproducible. Maybe a side effect of your sooper-dooper find
extensions not visible with the good old version?

> - Incorrect link counts on directories.

Maybe. Does that matter? Not really.

> - not correctly working deeply nested directories (level > 8)

Maybe. I have not looked at that part yet.

> - No find(1) support in mkiofs

 - News at 11: Fifth wheel not important for stable driving.

> - No corretly working graft points

How "not correctly"?

> - cdrecord/readcd/cdda2wav -scanbus does not find ATAPI drives on Linux

Because "-scanbus" sucks IMHO. We have "--devices" instead.

> - No auto-target feature in  cdrecord/readcd/cdda2wav (dev= parameter is
>   no longer needed with the original software)

Wrong at two points. First, we do not distribute binaries called
cdrecord/readcd/cdda2wav and you know that. Second, with wodim dev=-1
(default in the config) looks for an appropriate device, even deciding
about CD-R or DVD-R type depending on the track size. The next
user-friendly feature on the list is automatic scanning all drives for a
real audio CD in icedax.

> - No correctly working DVD support from the third party DVD enhancement
>   hack used by wodim.

"No correctly" and "hack" are fuzzy/subjective terms. Better tell how
severe the problem is that you claim to have found. If there is one at
all.

> While most changes in the wodim source are a result of replacing a working
> highly portable build system by something that does not work as good as the 
> original, the original software has been enhanced and:

It works as good as the original in the scope that we need. And in some
areas even beyond, remember AIX5l.

> - mkisofs did replace more than 1/3 of the code (mostly for bug-fixes) 
> since 
>   Debian did fetch the code

"mostly" is good. The changes that attracted most of my attention are
gazillions of meta-changes done to make the source look more closer to
the style used in your personal software zoo. Using the word "fix" or
"cleanup" (like Cstyle fixes) to describe them, ROTFL.

> - More than 2/3 of the cdda2wav code has been reworked in order to clean 
>   up the code, make it maintainable again and fixing bugs.

Oh, we also fixed some stupid bugs. I hope you fixed things like
potential buffer overflows in the meantime like we did. Unfortunatelly
no fixes can be pulled from your source because of the incompatible
license.

> - More than 1/3 of the cdrecord code has been replaced or added since 
>   Debian ffethced their code.

Yes, and? You add DVD support, we add DVD support, and the game
continues.

> If you like to use a recent cdrtools, I recommend to use the latest source 
> from:

So, that is what your message is about? Some FUD enabled marketing?

If you wanna beat othe

Re: Best File System for partitions over 600GB

[Eduard Bloch]

#include 
* Roberto C. Sanchez [Mon, Mar 12 2007, 07:06:43PM]:
> On Mon, Mar 12, 2007 at 11:34:48PM +0100, Mathias Brodala wrote:
> > Hi Roberto.
> > 
> > 
> > I see. I was asking since I have a whole drive full of videos and such 
> > which are
> > usually between 100MB and 300MB per file. So I guess XFS would not really 
> > be the
> > best choice for them. I got ext3 everywhere at the moment and wondered if I
> > could get a bit more performance by using another filesystem. And since I 
> > only
> > used ext3 up until now, I don???t really know which other filesystem to 
> > trust.
> > 
> I would certainly trust XFS.  Of course, if you don't have your machine
> on an UPS, it can cause problems on a crash or power outage.  How are

Great, that is the usual propaganda from XFS users with the same lame
excuse written with small letters. It has this bad tendency to shred the
file contents after powerouts or sudden kernel crashes... silently
inserting lots of 0x0s, IIRC sometimes only a 512 byte block, sometimes
filling the rest of a file after a certain position. I cannot prove it
either, it is just the experience which I had every time after I tried
XFS in the last years. And every time I came back to ext3 where I can
not remember such trouble.

Eduard.
-- 
 meebey: Mail kannst du eigentlich nicht verschachteln ... und dann
 kam MIME.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Attracting newbies (Was Booting Debian/testing fails)

[Eduard Bloch]

#include 
* Andrei Popescu [Tue, Feb 06 2007, 10:43:32PM]:
> On Tue, 06 Feb 2007 15:06:11 -0500
> Michael Pobega <[EMAIL PROTECTED]> wrote:
> 
> > Where is the graphical installer, anyway? I'd love to try it out
> > myself, I haven't got the chance (I just used NetInstall's default
> > curses installer).
> 
> Just boot with 'installgui' or 'expertgui'.
> 
> I've had bad experiences with Kubuntu myself. I convinced my brother to
> try it, but it would freeze at loading KDE. He was pretty disappointed
> and gave up. Now I want to give it another try with etch as I am more
> experienced and I know Debian much better anyway.

Such freezing at KDE start may be a typical hardware problem. For
example, my brother has a really broken sound chip on his laptop and
there was similar freezing when the start sound was to be played.

In such cases, changing the distro won't help you much.

Eduard.

-- 
 jstr: in welchem rfc steht, dass du nicht in die hose machen sollst?
 yath: es gehoert sich nicht.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: SoundBlaster AWE64 ISAPNP on etch

[Eduard Bloch]

#include 
* Kevin Mark [Sat, Feb 03 2007, 12:54:13AM]:

> > But when I try to play any sort of sound (audio CD, raw PCM file, etc.) I 
> > just get static from the speakers.
> > 
> > I have also tried:
> > # modprobe sound
> > # modprobe sound-core
> > but cannot determine what effect, if any, these have.
> > 
> > any advice?
> the basic advice is to at least include the output of 'lspci' (assuming
> its a pci card). And telling us if 'alsaconf' has any problems. Did you

You do not need to assume but to read the subject. It is not a PCI card.
Generaly, the snd-sbawe module is used for this card... however, it
seems that all ISA soundcard support has been removed in the current
kernels. Does anyone know what happened to it?

> try running 'alsaconf' as root and then using 'alsamixer' to turn up all
> the volumes and hit 'm' so that the controls DO NOT SHOW 'm'. 'm' means
> mute, but you use 'm' to toggle this.

And install alsa-utils to save the volume on shutdown.

Eduard.

-- 
* Madkiss würgt.
 Da lag gerade so ein Karton im Kühlschrank. "Lust auf
was Feines?" -- Und das war ... Frittenfett *bllüär*


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: kernel modules

[Eduard Bloch]

#include 
* remigio [Mon, Jan 22 2007, 06:30:29AM]:
> boot and dvd-writer problems
> From:
> [EMAIL PROTECTED]
>   To:
> [EMAIL PROTECTED]
>   Date:
> Today 15:23:25
> 
> Hello,
> I've installed Debian testing 2.6.18-3-486 on a PIV 2.8 GHz equiped
> with a
> dvd-rom and a dvd-writer.
> On bios the dvd drives are both present correctly, but when system
> startup and
> kde is running I can't find the writer: only dvd-rom is visible.
> I tried with wodim -scanbus but writer is not present and k3b says
> "Writer not
> found"
> I tried again on appending "hdc=ide-scsi" in menu.list for scsi

Why? Read /usr/share/doc/wodim/README.ATAPI.setup .

Eduard.

-- 
Naja, Garbage Collector eben. Holt den Müll sogar vom Himmel.
   (Heise Trollforum über Java in der Flugzeugsteuerung)


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Cdrecord with a 2.4 kernel

[Eduard Bloch]

#include 
* Sonixxfx [Tue, Jan 23 2007, 03:11:47PM]:
> Hi,
> 
> I am trying to burn multisession cd's with cdrecord but I have a problem.
> After I have written a multisession disc, and mount it, the burned files are
> not shown at /media/cdrom. The files are written to the disc though, because
> I can access/see them when I have rebooted my system and mount the cd.

Reboot? OMG. Just reload the disk.

Eduard.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Announce: zerotools (tools to keep virtual machine disks cleaner)

[Eduard Bloch]

#include 
* Mike Hommey [Wed, Jan 03 2007, 01:38:23PM]:
> On Wed, Jan 03, 2007 at 01:13:27PM +0100, Eduard Bloch <[EMAIL PROTECTED]> 
> wrote:
> > Please add a technical overview to your introduction. As a techie, I
> > would like to read what is so special about zerotools what I cannot do
> > with:
> > rm crapfiles && cat /dev/zero > bigfile && sync && rm bigfile && sync.
> 
> or just shred crapfiles

Not really. You need to add funny parameters like "-vzn0" to shred to
shred them with zero and then you still need to unlink them manually.
And IMO it is less intuitive than my command chain for this particular
purpose.

Eduard.

-- 
OpenBSD fails miserably in this respect, and makes for an example of how NOT
to work with the community on security issues.  Their approach is, roughly,
"we fixed this a while ago but didn't tell anyone, so you're vulnerable and
we're not, ha-ha-ha".


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Announce: zerotools (tools to keep virtual machine disks cleaner)

[Eduard Bloch]

#include 
* Aleksandr Koltsoff [Wed, Jan 03 2007, 05:47:36AM]:
> Hello
> 
> It was suggested that I'd post a short announcement on these two lists
> in order to get some feedback for a small set of utilities I just
> released. The documentation covers integrating the tools into Debian and
> I'd appreciate feedback on that part. Especially if I'm doing something
> which is out of line established Debian policies (hence the debian-devel
> list) [these parts are mainly listings 7 - 9 with the associated text].
> 
> I'll copy & paste the introduction text so that you may decide whether
> the program might be of interest to you. Target users for the tools are
> people who run Linux systems in a virtualized environment.
> 
> "Introduction
> 
> Zerotools are a set of tools to aid keeping virtual disks clean (by
> filling binary zero to those regions which are no longer in "use"). This

Please add a technical overview to your introduction. As a techie, I
would like to read what is so special about zerotools what I cannot do
with:
rm crapfiles && cat /dev/zero > bigfile && sync && rm bigfile && sync.

And you introduction does not tell me about anything new in your tools
and I do not like reading a long novel to learn that simple details.

Something like:

Introduction

Zerotools are a set of tools to aid keeping virtual disks clean (by
filling binary zero to those regions which are no longer in use as
visible files). In technical terms, they contain a library wrapper used
with LD_PRELOAD to make unlink calls zero-out the data before unlinking
the files and helper executables based on it.
...

Eduard.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: k3b + wodim fails to burn

[Eduard Bloch]

#include 
* Giacomo Montagner [Fri, Dec 22 2006, 03:35:12PM]:
> Hi!
> I just upgraded and have these packages installed:
> 
> Package: wodim
> Architecture: i386
> Source: cdrkit
> Version: 9:1.1.0-1
> 
> Package: k3b
> Architecture: i386
> Version: 0.12.17-5
> 
> When I try to burn data-cd using k3b it fails with the following error:
> 
> 
> 
> [BLAH BLAH BLAH...]

The key to your answer is in BLAH BLAH BLAH. I suggest reading it.

> 
> 
> What seems strange to me is that I tried creating only the iso image with
> k3b, then burnt the cd using:
> 
> dd if=/tmp/k3b_image.iso | cdrecord speed=10 dev=/dev/scd0
> driveropts=BURNFREE -v -
> 
> and it worked perfectly... then I discovered (amazingly enough) that
> cdrecord is a symlink to wodim...

What about READING THE MESSAGES? How can we know what is going wrong if
you hide them?

Eduard.

-- 
Sinclair: Are you telling me this is about a flower?
 -- Quotes from Babylon 5 --


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: apt-cacher questions

[Eduard Bloch]

#include 
* Wayne Topa [Mon, Nov 27 2006, 09:53:12AM]:
> Wackojacko([EMAIL PROTECTED]) is reported to have said:
> > Wayne Topa wrote:
> > >I am trying to setup apt-cacher (v1.5.3) and seem to be getting some
> > >strange results.  I wonder if any apt-cacher users might enlighten me.
> > >
> > >(3)
> > >How have you setup your servers sources.list?  I saw a reference to
> > >doing deb http://127.0.0.1/package.site, but that didn't work for the
> > >server or the clients.
> > >
> > >
> > your missing the port number in your package listing.  Mine looks like this
> > deb http://localhost:3142/site.address
> > 3142 is default port.
> > 
> 
> Sorry, my bad, I did have the port in my sources.list but didn't show
> it in the OP.  :-(
> 
> deb http://127.0.0.1:3142/mirrors.kernel.org/debian/ testing main contrib 
> 
> I had to replace it tho because  (this) server would not upgrade
> until I removed the 127.0.0.1:3142/.  Aptitude kept saying there were
> no packages to upgrade until I removed it, after that, it upgraded 50
> packages.
> 
> The clients work sometime for a few packages and them stop.  I find a
> lot of these in the error.log, on the server, when the clients try
> upgrading.
> 
> ALARM!Faulty package in local cache detected! Removing, to be
> replaced with the next download.
> 
> Sorry for not being more explicit in the OP.

Maybe mixed up with Ubuntu packages? Hard to debug. Enable the debug
logging and look for suspicious things. Unfortunately, I cannot work on
apt-cacher's successor now. See
http://www.rootfs.net/jaws/?gadget=StaticPage&id=3 for details.

Eduard.

-- 
 noel, nobse, Ganneff... ist mit allen das selbe: braucht man sie, sind
sie nicht da, braucht man sie nicht, landet man in den fortunes...


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Can't find header files.

[Eduard Bloch]

#include 
* Amit Joshi [Mon, Nov 13 2006, 12:16:12PM]:
> I recently installed gcc and Anjuta for C programming. 
> All dependencies have been installed. 
> 
> So whenever I try to compile a program, it returns an error saying: 
> Error: stdio.h: Can't find file or directory. 

apt-get install build-essential

-- 
Alles in der Welt endet durch Zufall und Ermüdung.
-- Heinrich Heine


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: cdrecord replacement (wodim) fails

[Eduard Bloch]

#include 
* Wayne Topa [Tue, Oct 24 2006, 03:08:08PM]:
>   Just upgraded cdrecord and k3b to burn a Knoppix CD and find that
>   the new replacement for cdrecord, wodim, can't handle it.  It fails
>   saying the file is larger the the CD.

Show your media data, "wodim -atip". And the exact output when tryin to
burn, doing "wodim -vv imagefile".

Eduard.

-- 
Der Aberglaube an die automatische Wirkung der Einsicht kommt
außerhalb der schematischen Propaganda nur noch bei alten
Mathematikern vor.
-- Ernst Bloch


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: why "Starting MTA: exim4" take so much time?

[Eduard Bloch]

#include 
* Serena Cantor [Fri, Oct 13 2006, 07:02:26PM]:
> During sarge installation, I select 3rd option (local delivery). Each time it 
> boots, starting MTA
> take too much time, I have to remove it from /etc/rc2.d.

You have a broken (but configured) internet connection? Your DNS makes
problems?

Run "dpkg-reconfigure exim4-config -p low" and say yes in the "Minimize
DNS requests" question.

Eduard.
-- 
Vertrauen Sie denen, die nach der Wahrheit suchen, und mißtrauen Sie
denen, die sie gefunden haben.
-- André Gide


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: mkdir on dvd+rw

[Eduard Bloch]

#include 
* Fred J. [Fri, Sep 15 2006, 11:53:18AM]:
> Hi
> 
> using debian/testing 2.6.15
> cannot I use a dvd+rw as a hard drive? I mean, to 
> make directories,
> move files around,
> change content of a file,
> ...?
> 
> I just copied "or should I say burned" some files using
> growisofs -Z /dev/cdrom1 -R -J /some/files
> 
> how can I place them in a directory on the dvd+rw media?

Read the manpage. Read mkisofs(8). Use the -root option. Also note
that growisofs creates a multisession disk with the -M option, and so
following "copy" operations cost some overhead (wasted space) because a
new iso9660 filesystem is created, plus some additional metadata for
data tracks.

Eduard.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Debian support for Marvell Yukon Gigabit Ethernet NIC?

[Eduard Bloch]

#include 
* Yu,Glen [Ontario] [Mon, Aug 21 2006, 09:52:26AM]:
> Hi everyone,
> 
> Over the weekend I decide to put Debian on my new PC at home, and while
> it was detecting hardware, it said it couldn't find the proper driver
> for my NIC (Marvell Yukon Gigabit) that's onboard my ASUS MB (P4P800??
> I forgot; but I can confirm once I get home).  I quickly checked

I have an Asus board with VIA Athlon64 chipset. I tried the drivers from
their homepage (and from their partners) but it almost never worked for
me, it simply refused to detect the card. However, the sky2 driver in
a recent kernel worked almost well. (it only stoped working after
suspend-to-disk).

Eduard.

-- 
Wer flie?end fiese Fliesen fliest, der fliest fies fiese Fliesen flie?end.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: cdrecord completion questions

[Eduard Bloch]

#include 
* Hank The Tank Winky Poo [Wed, Aug 09 2006, 03:01:37PM]:
> Hello all
> 
> While we're on cdrecord, at the end of the command
> 
>   sudo cdrecord -v -sao -dev=4,0,0 speed=4 driveropts=audiomaster 
> -audio
> *.wav
> 
> I get something like...(after the audio tracks have been written)
> 
>   Fixating...
>   Fixating time: 29.477s
>   cdrecord: fifo had 10560 puts and 10560 gets.
>   cdrecord: fifo was 0 times empty and 6916 times full, min fill was 92%
> 
> Does the fact that #puts = #gets mean a clean burn with no errors?

It simply means that all data chunks has been passed to the drive, not
more, not less. If the drive works correctly and no other software error
happened, than it means a clean burn.

> Who is fifo? Is it file in/file out?

man cdrecord

"First In First Out", a type of data buffer where data is "moving" into
one direction.

Eduard.
-- 
Bildung ist das, was übrigbleibt, wenn wir vergessen, was wir gelernt haben.
-- Edward Frederick Lindley Wood


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: cdrecord wihout SUID

[Eduard Bloch]

#include 
* Dwayne C. Litzenberger [Mon, Aug 07 2006, 03:16:46PM]:
> On Mon, Aug 07, 2006 at 01:39:43PM -0600, Glenn English wrote:
> >If you want to do it without the extra group, just make the burner world
> >write-able. I don't see how a cracker could get much mileage out of that...
> 
> Note, however, that if cdrecord doesn't run SUID root, it can't get 
> realtime scheduling priority, so you could run into buffer underruns on a 
> slow or heavily-loaded system.

It is not only about priorities. In fact most modern systems are
overpowered for writting tasks especially if burn-free feature is used.

There is a problem with not beeing suid which occurs with certain
cdrtools <-> kernel versions. The upstream author does not support
non-(suid)root usage by principle. There is a patch in the BTS to work
around this problem (bailing out on forbidden "rezero_unit").

Eduard.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Sarge - how to write a filesystem backup directory to a DVD ?

[Eduard Bloch]

#include 
* Keith Christian [Mon, Aug 07 2006, 03:23:07PM]:
> Having gotten a new internal DVD burner, I'd like to know the best way to burn
> a directory of files to a DVD for system backup.  (Also, I'd rather burn the
> DVD using a command at the BASH prompt instead of from a GUI.)  The capacity
> of the DVD is 8gb, write speed is 16x, and the size of the directory to be
> burned to the DVD is about 7gb so there should be plenty of space.
> 
> I've installed dvd+rw-tools from the Sarge CD's.
> 
> I'm a little confused - should the input to "growisofs" be the directory that
> is to be burned to the DVD, or should an .iso be created first, and that in
> turn burned to the DVD, as is done with CD's ?

Honestly, RTFM of growisofs helps reducing your confusion. It runs
mkisofs itself and catches the output. So it is not neccessary to make
an iso image, you specify the same things you would do with mkisofs, eg.
a directory, multiple directories, or "graft-pointed" catalogue.

For big directories, the dirsplit utility can be used (from
www.freshmeat.net, the one in Sarge is not uptodate).

> I've seen references to "formatting" or "initializing" DVD's on a few websites
> but don't know if that's really necessary.

Unless you are going to use the UDF filesystem (writting on-the-fly)
formatting should not be necessary.

>  Free Blocks:   2295104*2KB
>  Track Size:2295104*2KB

Looks like a usual 4.3GiB medium.

Eduard.
-- 
 Joey: ich häng schon am Manual.
 amq: Naja, du musst halt nicht nur dranhängen sondern auch lesen.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: How do I install a non packaged font in Debian?

[Eduard Bloch]

#include 
* Angelina Carlton [Thu, Jul 20 2006, 05:15:33PM]:
> Eduard Bloch <[EMAIL PROTECTED]> writes:

> Hi, I have installed the package fttools and removed the directories in

Yeah, read the hints and ignore them, I like that.

> /usr/share/fonts and /usr/local/share/fonts and now am simply working
> from a ~/.fonts to keep things from getting confusing. 

... and from working.

> But xlsfonts still show no fonts from my ~/.fonts folder. 
> I am trying to use the fonts in urxvt and in emacs if that matters. What
> is frustrating is that on the previous Sid install, I had installed
> these fonts wihout a problem and forgotten to document how I did it :)

As said before, xlsfonts does not care about your ~/.fonts folder
because it uses the old font management driven by the X server which
does not care about your personal directory. See xorg.conf for a list of
known font paths.

You probably used the xft-Syntax with urxvt before so it has used
Fontconfig system instead of X11. See urxvt manpage, eg.

  urxvt -fn "xft:Bitstream Vera Sans Mono:pixelsize=15"

Eduard.
-- 
 die user schreien auch nach ms word
 trotzdem tun wirs nicht ins Archiv
 formorer: nicht?
 formorer: was macht dann _rene_ die ganze zeit?


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: How do I install a non packaged font in Debian?

[Eduard Bloch]

#include 
* Angelina Carlton [Thu, Jul 20 2006, 02:25:05PM]:
> "H.S." <[EMAIL PROTECTED]> writes:
> > IIRC, I tried this and it seemed to have worked for ttf fonts for my user:
> >1.   Copy the ttf font file to a .fonts directory in your home
> > directory.
> >2. Give the command fc-cache -fv in a terminal. Once this command
> > is finished, restart your application, e.g. Firefox, and you should be
> > able to see the newly installed font.
> >
> > You don't have to be root for the above to work.
> >
> >
> > If you want system wide fonts, you need to copy the ttf fonts in
> > /usr/loca/share/fonts directory, then give the 'mkfontdir' command
> > within that dir as root, and then give the 'fc-cache -fv' command as
> > root.
> 
> I created a directory in /usr/local/share/fonts called proggy, did a
> mkfontdir and issued fc-cache -fv as root, so the font will be
> systemwide.
> It spat out alot of output including:
> fc-cache: "/usr/local/share/fonts": caching, 0 fonts, 1 dirs
> fc-cache: "/usr/local/share/fonts/proggy": caching, 2 fonts, 0 dirs
> 
> So it appears it added the fonts? But xlsfonts still wont find them, I
> even added:   FontPath "/usr/local/share/fonts/proggy" to my xorg.conf
> but to no avail.

xlsfonts uses the old X11 font system. And mkfontdir creates index files
needed for it but only for the old pcf fonts. You want to use mkttfdir
instead, from the fttools package.

Eduard.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: 2.6 kernel image for AMD-K6-2

[Eduard Bloch]

#include 
* Joop [Wed, Jul 12 2006, 03:47:18AM]:
> Maybe I got it wrong since I am a bit new to the scene.
> But this is what I did.

You got it wrong. That's about the linux-image-... kernel packages which
have only few profiles, not including -k5, -k6 and many others.

Eduard.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: 2.6 kernel image for AMD-K6-2

[Eduard Bloch]

#include 
* Andrew Sackville-West [Mon, Jul 10 2006, 09:34:17AM]:
> On Sun, Jul 09, 2006 at 05:09:41AM -0700, Helen Easthope wrote:
> > Debian users,
> > 
> > Is there a straightforward way to get a 2.6 kernel image for an 
> > AMD-K6 machine?  The 2.4 image works fine but the 2.6 AMD-K7 
> > image fails on this Compaq Presario 5330.
> 
> I think you want the -686 kernels. THere seems to be no k6 in the 2.6
> series.

AFAIK no, IIRC Athlon was the first fully 686 compatible processor. You
can go with the -386 kernel instead, IMHO you won't loose much by using
it instead of -k6.

Eduard.

Re: apt problems

[Eduard Bloch]

#include 
* martin f krafft [Thu, Jun 22 2006, 02:03:55PM]:
> also sprach Martin Lemmen <[EMAIL PROTECTED]> [2006.06.22.1846 +0200]:
> > APT::Cache-Limit "141943904";
> 
> woho. I don't even set this anymore and:
> 
> $ grep -v '^#' /etc/apt/sources.list | wc -l
> 34 /etc/apt/sources.list
> 
> Anyway, I think your bzip2 problems may be related to a caching
> proxy somewhere between you and Debian. Try using ftp2.de and see if
> the problem remains. If it does not, wait for the cache to expire
> the ftp.de entries (5 days maybe) and change back. Or stay with
> ftp2.de.

Not neccesarily. I have seen apt processing spew weird messages WRT broken
bzip2 decompression. Apparently something is buggy there, but the
problem disappears after rerunning update.

Eduard.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: changing user id for all files in a file system

[Eduard Bloch]

#include 
* George Borisov [Thu, Jun 15 2006, 11:21:24AM]:
> Stephen R Laniel wrote:
> > 
> > sudo find / -uid [old UID] |xargs -i sudo chown [new UID] '{}'
> 
> Thanks for introducing me to the xargs command - never heard of it
> before, but I am sure it will come in handy some day. :-)
> 
> The '-i' option is deprecated in xargs from unstable, though.
> Alternative seems to be:
> 
> find / -uid OLDUID | xargs -I [] chown NEWUID []
> 
> (I've removed the 'sudo' bits, as it would not work with the default
> setup.)

Why fiddling around with bracket monsters? GNU find and xargs have means
to protect commands from failing on spaces&co.

find / -uid OLDUID -print0 | xargs -0 chown NEWUID

Eduard.

Re: sshfs?

[Eduard Bloch]

#include 
* jpg [Tue, May 09 2006, 10:44:58AM]:
> Hey List,
> 
> Trying to get sshfs to work on my 'sarge' box.
> 
> Installed sshfs, added root to the fuse group, then attempted to connect to
> remote machine that has sshd, with sftp extension, and get:
> 
> # sshfs remote_host:/  /
> Password: 
> FATAL: Module fuse not found.
> fusermount: fuse device not found, try 'modprobe fuse' first
> 
> # modprobe fuse
> FATAL: Module fuse not found.
> 
> 
> Nothing in man page describing needing to have a fuse module.
> Nothing in /usr/src/linux-2.6.12.2 about fuse

Because 2.6.12 is pretty old and did not include fuse at that time. Try
"module-assistant prepare" and "m-a a-i fuse". Or upgrade the kernel.

Eduard.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: How Do I Start the Debian GUI

[Eduard Bloch]

#include 
* Kent West [Sun, May 07 2006, 08:24:17AM]:
> [EMAIL PROTECTED] wrote:
> > Hi ... i am newbie too...
> >
> > I have recently installed the GUI (I think)... but there seems to be
> > some problem with starting the GUI
> >
> > startx, startkde doesnt seem to work
> >
> > during installation of the GUI, these messages came out
> > dpkg 'ldconfig' not found on PATH
> > dpkg 'start-stop-daemon' not found on PATH
> > dpkg 'install-info' not found on PATH
> > dpkg 'update-rc.d' not found on PATH
> >   
> This indicates to me some serious brokenness. I suspect your
> installation of Debian is not complete somehow.

Serious? He is more likely not root or messed up his $PATH.

Regards,
Eduard.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: amd64: why is sizeof(int) =4? why not =8?

[Eduard Bloch]

#include 
* Miquel van Smoorenburg [Thu, Apr 06 2006, 10:29:00AM]:
> In article <[EMAIL PROTECTED]>,
> Ron Johnson  <[EMAIL PROTECTED]> wrote:

> If you need exactly 64 bits, int64 is your best bet. If you need
> at least 64 bits, long long will probably work fine.

I guess you mean int64_t and uint64_t (in contrary to int32_t/uint32_t,
etc.). And IMO one should use these types nowadays when the value range
is known upfront. Symbolic names are unreliable, int may be anything
larger than 16 bit and long may be anything larger 32 bit, but you can
never know for sure without testing.

Eduard.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: GNOME 2.14 released :-)

[Eduard Bloch]

#include 
* Magnus Therning [Thu, Mar 16 2006, 04:16:52PM]:
> I hope we'll see it in Debian Sid soon. Go GNOME team! Go!
> 
> Anyway I can help?

Did you consider to donate money?

Eduard.

Re: apt-proxy-import help

[Eduard Bloch]

#include 
* Johan Daine [Sat, Mar 04 2006, 08:18:47AM]:
> Hi everyone,
> 
> I am making some tests with apt-proxy.
> apt-proxy is working perfectly. I have been purging gpm, removing the
> 2006/03/04 08:16 CET [-] [apt_pkg] No Packages files available for
> security backend
> 2006/03/04 08:16 CET [-] [apt_pkg] No Packages files available for
> debian-non-US backend
> 2006/03/04 08:16 CET [-] [import] libconfigfile-perl_1.2.1_all.deb
> skipped - no suitable backend found
> 
> and so on...
> Any clue to solve this problem?
> Thanks in advance.

apt-proxy is poorly documented WRT error messages. I guess this happens
because it needs Packages files from the backend directories to get the
paths of that files. No packages files -> no path data -> no clue where
to store the files (to have correct url) -> apt-proxy confusion.

Solution: update the Packages files by running "apt-get update" on
clients, so apt-proxy would need to download them at least once.

(self-add...) I am maintainer of apt-cacher and the current version
will let you import anything, but may deliver wrong files later. The
next version will track the files in a similar way that apt-proxy does.

Eduard.
-- 
Emperor Cartagia: [to Londo Mollari] The trouble with being infallible is that
you have to deal with everyone else making mistakes.
 -- Quotes from Babylon 5 --


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Kernel 2.6.15 networking works with irqpoll

[Eduard Bloch]

#include 
* Nil Cire [Thu, Feb 16 2006, 01:01:32AM]:

> I recently had another post about the networking not working. Well,
> I've solved it. However, I still don't know why it works. What the
> heck does irqpoll do? Is there any way I can do it without using
> irqpoll? Refer to my other post (Networking problems specific to
> linux-image-2.6.15-1-686) something like that.

IIRC irqpoll is a workaround for some bizzare kernel bug that they
could not fix otherwise. I have a similar problem since 2.6.15.1, I also
had this problem with Ubuntu kernels since Hoary or so, and with
resuming after suspend-to-ram with all regular kernels since 2.6.6.
Always saying:

IRQ 11: Nobody cared...

and then just behaving odd. IRQ 11 is the one for my onboard periphery
controller, so my USB is broken (mouse jumps when moved, like having
only 1 dpi instead of 800dpi), my NIC is broken (mii-tool says ok but
its loosing all packages) etc.pp.

Eduard.

-- 
Sinclair: You should never hand someone a gun unless you are sure where they
will point it
 -- Quotes from Babylon 5 --


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: AMD 64 X Intel 32

[Eduard Bloch]

#include 
* Iuri Sampaio [Sat, Feb 11 2006, 05:29:53PM]:

> 80 GB Hard Disk
...
> 120GB HD

You asked about performance and do not know more than "X GB"? Please realize
that most cheaper laptop harddisks are rotating with 4200 RPM, better
ones with 5400 and best with 7200, and there is real speed difference.

If the both harddisks have the same rotational speed the last one is the
better choice, of course. It may even be faster because of higher data
density or have more internal disks.

Eduard.

-- 
  "After watching my newly-retired dad spend two weeks learning how to
  make a new folder, it became obvious that "intuitive" mostly means
  "what the writer or speaker of intuitive likes".
- Bruce Ediger, [EMAIL PROTECTED], on X the intuitiveness of a Mac interface


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: AMD 64 X Intel 32

[Eduard Bloch]

#include 
* Ron Johnson [Sat, Feb 11 2006, 07:26:30PM]:

> > Video card: NVIDIA 128mb dedicated
> 
> This is the video card of choice for Linux.  The (closed-source)
> 3D driver is excellent.

Please, on a laptop? When did they start supporting the power management
properly?

The free driver works well enough, though.

Eduard.
-- 
Marcus Cole: That's a lot of ships!  That's a bloody awful lot of ships!
 -- Quotes from Babylon 5 --


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: XTerm and Bitstream Vera

[Eduard Bloch]

#include 
* Vincent Lefevre [Mon, Aug 15 2005, 04:35:01PM]:
> On 2005-08-15 15:53:29 +0200, Eduard Bloch wrote:
> > More useable fonts gets you "fc-list | grep Mono"
> 
> Or "fc-match --sort Mono".

I don't think so. It shows a lot of fonts which are not pure monospace,
if useable at all, eg. Tahoma.

Regards,
Eduard.
-- 
Citizen G'Kar: Oh, why does the universe hate me.
 -- Quotes from Babylon 5 --


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: XTerm and Bitstream Vera

[Eduard Bloch]

#include 
* Kumar Appaiah [Mon, Aug 15 2005, 07:46:35AM]:
> On 8/14/05, Vincent Lefevre <[EMAIL PROTECTED]> wrote:
> > > xterm -fa "Bitstream Vera" -fs 8
> 
> This is a great setting! Thank you very much!

And if it draws to slow for your liking:

urxvt -fn 'xft:Bitstream Vera Sans Mono:size=8'

More useable fonts gets you "fc-list | grep Mono"

Regards,
Eduard.
-- 
 Tolimar: Was lernt man in einem "Kochen für Singles"-Kurs?
 aba_: Ich habe gelernt, dass es einfacher ist,
eine Tiefkühlpizza in den Ofen zu schmeissen ;)


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: XTerm and Bitstream Vera

[Eduard Bloch]

#include 
* Jochen Schulz [Mon, Aug 15 2005, 10:56:37AM]:
> Vincent Lefevre:
> > On 2005-08-14 17:47:12 +0200, Jochen Schulz wrote:
> > > 
> Fiddling with the DPI setting can cause weird font display. The
> "correct" solution would be to find out the real value for your monitor
> and resolution. You can do this by explicitly telling X how large your
> display is. Put something like this in the "Monitor" section of your
> Xfree86/X.org config file:
> 
> Displaysize 283 212 # display dimensions in mm



In theory, that is correct. In practice it (as usual) depends. Some
software authors think that is the best way to specify the font sizes
they like, and fontconfig plus a correct DPI setting (see above) help
to establish the display of exactly this real font size on every
display. However, it always depends on how close the eye is to the
screen and on the personal liking. So for many people fiddling with
Xft.dpi setting does make sense to move the variaty of font sizes into a
range that they like best.

Regards,
Eduard.

-- 
Alles, was im Kühlschrank steht, ist Public Domain.
-- Torsten Spindler


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: which package contains libXxf86dga.so.1

[Eduard Bloch]

#include 
* belahcene [Fri, Jul 08 2005, 05:45:00PM]:
> Hi every body,
> I install mplayer from  Rpm ( for fedora fc3 ) on my sarge, no error is 

That's your problem. Redhat modified their X11 to provide some libraries
in shared versions though they have never been to become shared-libs.
So users of other distros have to suffer from their "creativity" now.

Consider building mplayer from source our using the Marillat version.

Regards,
Eduard.

-- 
Tyrannei und Diktatur sind bestimmt nicht die besten Staatssysteme.
Kann aber nicht auch zuviel Mitbestimmung ihre Schattenseiten haben
-- H. Bauer


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: CDRW problem

[Eduard Bloch]

Moin Manu!
Manu schrieb am Samstag, den 08. Januar 2005:

> cdrecord -scanbus dev=ATAPI
> Cdrecord-Clone 2.01a38 (i686-pc-linux-gnu) Copyright

Not dev=ATAPI (it sucks, no DMA). Use dev=ATA.

Regards,
Eduard.
-- 
So, your "solution" is to ask "Should I break your system now or after
the next reboot?". Debconf is not an alternative to fixing the
problem. Such questions are still unacceptable bugs.
  -- asuffield in debian-devel about crazy linux-2.4 repackaging


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: nv vs. nvidia

[Eduard Bloch]

#include 
* Jon Dowland [Thu, Oct 28 2004, 12:33:47PM]:

> > In case you don't need to run 3d programs, using nv will be fine for you
> 
> ...provided you have a reasonably modern computer or a small
> resolution display. On my AMD K6-2 450mhz, the nv driver is unbearably
> slow at 1280x1024.

I don't think the CPU is your problem. Do you have a TNT-2 or older
Nvidia chip? The XVideo support for them is not complete, which means
taht watching videos becomes _very_ CPU intensive. Although, the nvidia
drivers does support XVideo at TNT-2. With modern Nvidia cards, I did
not notice any big difference except of GLX support. And the NV driver
is less buggy WRT power saving system modes.

Regards,
Eduard.
-- 
 hmm, naja, aber kde 2 ist nicht wirklich prickelnd
 martoss: wenn du was prickelndes willst, trink mineralwasser
 weasel: :-), ich weiss, aber es muss halt auch ein bischen was fürs
auge da sein...
 martoss: dann tu nen strohhalm rein! :)


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: software suspend 2 (swsusp2), patching a Debian source package kernel vs Vanilla

[Eduard Bloch]

#include 
* H. S. [Fri, Oct 22 2004, 02:11:09PM]:

> If it is, I could just do
> $> patch -p1 ../patch-name-here

Look in what the swsusp2 patch tarball contains (untar it!). There are
patches that are to be applied in some order, and there is a patch
script that will do that for you.

Regards,
Eduard.
-- 
"The quiet ones are the ones who change the universe.
 The loud ones only take the credit."
   L.M. ItB


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Small problems with Sarge net-install

[Eduard Bloch]

#include 

> My recommendations are:
> 
> 1. libc6-dev should be part of the standard installation. It includes 
> all of the standard header files (stdio.h, for example).

This depends on _your_ needs! For example, I don't think that the
compiler is seen as part of a user-oriented operating system. If you
want to do "complicated stuff, coding", eg. compile programs, you want
to install build-essential anyways. Same thing if you install the kernel.

> 2. ssh should also be part of any installation.

ssh has the standard priority and should be suggested somewhere during
the base config. At least you would need to select some special things
to avoid installing it.

If you wanted to say that ssh should be part of the basedebs or even an
"essential" package, then no. Simply no. There are enough boxes where
ssh is not needed or should not be installed for various reasons.

Regards,
Eduard.
-- 
Der Aberglaube ist die Poesie des Lebens; deswegen schadet's dem 
Dichter nicht, abergläubisch zu sein,
-- Johann Wolfgang von Goethe (Maximen und Reflexionen)


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

[survey] TERM=rxvt-unicode as package default

[Eduard Bloch]

Hello,

I am the maintainer of rxvt-unicode and wish to get more user comments
on the recent change in the package. Nowadays, it uses "rxvt-unicode" as
the TERM setting which makes curses-based applications use the new
terminfo file which is slightly different than the one from rxvt (more
and better colors, italic fonts, and should fix some rxvt bugs). That is
the setting prefered by the upstream author.

All the new features are consistenly integrated into the Debian system.

This, however, means a drawback for few users. You may remember problems
caused by xterm when it used xterm-debian as terminfo entry and someone
logged into a remote box using ssh (which silently sets the same TERM on
the remote side). This still happens nowadays with rxvt and Eterm (when
logging into older and/or non-Linux systems). This will happen again
with rxvt-unicode because the terminfo is fairly new and is not included
in the terminfo database (of ncurses) on the common Linux distributions
(not on any other Unix).

To make remote applications useable, one can add TERM=rxvt into the
.ssh/environment file on the remote host or set (and export) it on the
command line. Alternatively, one can work around a such "limitation"
easily by setting a shell alias like 

alias ssh='TERM=`test "$DISPLAY" && echo xterm || echo $TERM` set'

Of course rxvt-unicode can work with applications that expect "rxvt" or
"xterm" mode, but you should expect some visibility bugs and key
incompatibilities (the famous home/end/del/backspace problems).

Personaly, I do not feel that the change to "rxvt-unicode" as default is
a real drawback - I would have to deal with such "problems" for most
remote systems I use (even for "rxvt"). But some people may see this
differently and I may change this for sarge.  Therefore I wish to see
more input ;)

Eduard.
-- 
Immerhin meint die Filmförderungsanstalt, im Jahr 2002 seien 59
Millionen CD-Rohlinge von 5,9 Millionen Nutzern mit Filmen bespielt
worden, im Durchschnitt also zwölf Rohlinge pro Anwender.
-- 


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: vmware packaged as .deb

[Eduard Bloch]

#include 
* William Ballard [Tue, Jul 06 2004, 03:20:02AM]:
> The guy who packages ATI fglrx drivers for Debian
> (http://xoomer.virgilio.it/flavio.stanchina/debian/fglrx-installer.html)
> is a godsend.
> 
> Has anybody ever done a similar sort of thing to VMWare?  It would be 
> highly useful.

Invent something. I did already look on how to automate the
module building with an module-assistant control script but the
makefiles are very confusing when it comes to passing custom parameteres
so the project has been moved to my TODO list.

Regards,
Eduard.
-- 
Der Mann erträgt die Ehe aus Liebe zur Frau. Die Frau erträgt den Mann
aus Liebe zur Ehe.
-- Gabriel Laub


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Mono tutorial --> gtk-sharp?

[Eduard Bloch]

#include 
* Tom [Sun, Jul 04 2004, 02:13:32AM]:

> * How could/should I've known that the mentioned introduction is
>   deprecated? Do you perhaps know some tutorial-like introductory
>   material that is more up to date? While the libraries seem to

I do not currently know one, but looking at the monodoc source (for
example) could help.

Regards,
Eduard.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Mono tutorial --> gtk-sharp?

[Eduard Bloch]

#include 
* Tom [Sat, Jul 03 2004, 02:16:11AM]:

> Today, I thought I'd give the Mono/Gtk# tutorial at 
> www.gotmono.com/docs/gnome/bindings/gtk-sharp/getstart.html a shot, but 
> compiling the very first tiny example yields errors (about not finding 
> the assemblies 'gtk-sharp.dll' and 'glib-sharp.dll'. 

Don't follow deprecated guides. This method did work with alpha versions
of Mono when all .dll and .so files have been thrown into /usr/lib but
does not work anymore.

Nowadays, run:

mcs helloworld.cs /pkg:gtk-sharp

> By now, I must have installed anything available in sid. The gtk-sharp 
> package itself refers to both libgtk-cil and libglib-cil. Those are 
> installed just fine.

Yup, better use Sid (and wait till tommorow when Mono 1.0 appears on
mirrors).

Regards,
Eduard.
-- 
Wenn Schlachthöfe Glasfenster hätten, wäre jeder Mensch Vegetarier.
-- Paul McCartney


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]