Re: Replace Grub with rEFInd [WAS Possibly broken Grub or initrd after updates on Testing]

[Joel Roth]

On Thu, Jan 04, 2024 at 06:19:01PM +0100, Richard Rosner wrote:
> In theory, it should
> be as simple as refind-install. So the only reason I could guess to be the
> reason would be that rEFInd might not be capable of handling LUKS, which
> would be quite disappointing. 

My experiences are with vanilla filesystems only. LUKS
obviously has specific requirements. Perhaps you could try
having a root partition that is unencrypted?

-- 
Joel Roth

Re: Replace Grub with rEFInd [WAS Possibly broken Grub or initrd after updates on Testing]

[Joel Roth]

On Wed, Jan 03, 2024 at 08:23:29PM +0100, Richard Rosner wrote:
> So, since for whatever reason Grub seems to be broken beyond repair, I today
> tried to just replace it with rEFInd. Installation succeeded without any
> trouble. But when I start my system, rEFInd just asks me if I want to boot
> with fwupd or with the still very broken Grub. Am I missing something? Is
> rEFInd really just something to select between different OSs (and not just
> different distributions like Grub can very well do) and then gives the rest
> over to their bootloaders or am I missing something so rEFInd will take over
> all of Grubs jobs?

I boot my debian-based system with rEFInd.  Grub is not
present. A couple big icons show on the boot screen. The
small print at the bottom mentions hit F2 for more options.
On my system, F2 offers a selection among all kernels
present. 

rEFInd installs into  EFI/refind/ in the EFI partition.
I originally encountered it looking for something to
boot debian on a Intel Mac. It's been trouble-free.




> On 01.01.24 21:45, Richard Rosner wrote:
> > 
> > 
> > On 01.01.24 21:20, Richard Rosner wrote:
> > > 
> > > On 01.01.24 20:30, David Wright wrote:
> > > > On Mon 01 Jan 2024 at 19:04:20 (+0100), Richard Rosner wrote:
> > > > > On 01.01.24 18:13, David Wright wrote:
> > > > > I can boot by hand, but since this is all archived anyways and it's
> > > > > uneccessarily difficult to find some sort of guide how to even do
> > > > > this, it might as well be a documentation for users having such
> > > > > troubles in the future.
> > > > > 
> > > > > Also, besides the way that I have no clue how it would have to look
> > > > > like to set up a paragraph in the grub.cfg, I simply don't see
> > > > > anything wrong with it anyways. So I can't even look at the grub
> > > > > settings files grub.cfg is being generated from to check where the
> > > > > error lies.
> > > > You append the commands that you used to boot manually with into
> > > > /etc/grub.d/40_custom, observing the comments there, and also into
> > > > grub.cfg itself at the appropriate place (near the bottom). The
> > > > former is so that Grub includes it in any new grub.cfg that you
> > > > create.
> > > Good to know.
> > Edit:, never mind. Tried that, it still booted straight to the UEFI BIOS
> > menu after entering my password. At this point, I'm seriously
> > considering slapping rEFInd on it and pray that it picks up on
> > everything automatically and fix the situation. But so should Grub have,
> > besides the fact that I can't even be entirely sure Grub is to blame and
> > not something else.

-- 
Joel Roth

Re: Alpine/Gmail/Imap expert needed.

[Joel Roth]

On Tue, Nov 28, 2023 at 02:06:13AM +0100, hw wrote:
> On Mon, 2023-11-27 at 19:22 +0100, to...@tuxteam.de wrote:
> > On Mon, Nov 27, 2023 at 06:57:51PM +0100, hw wrote:
> > > On Sun, 2023-11-26 at 20:54 -0500, Karen Lewellen wrote:
> > > > Your surprise is surprising given you are living no one's life but your 
> > > > own.
> > > > as they say walk a mile in another person's shoes before you decide you 
> > > > know what solutions are possible for them.
> > > 
> > > In that case, I suggest noone try to give any advice.
> > 
> > Hey, hw: no need to get grumpy. The OP has been polite all the time, she
> > just has special need. If you feel like that, better not answer.
> 
> She has been the opposite of polite.

I'm not offended at all by her manner, but at times find it
hard to parse out what she wants/needs from her description.
I don't blame her; certainly she writes as clearly as she
knows how. Usually someone (in this case Gareth) gets
what she wants. At least it's never dull ;-)


> > Cheers
> 
> 

-- 
Joel Roth

Re: connect two hosts over wifi without router?

[Joel Roth]

On Mon, Nov 27, 2023 at 02:54:44PM +0100, Hans wrote:
> Hi folks,
> 
> just before I am trying forever:
> 
> Is it possible, to connect two hosts directly over wlan without using a 
> router?
> 
> The background: I want to stream video from my drone using RTMP to my 
> notebook. 
> 
> This is already possible, when i am using a router. But in the fields, I got 
> no 
> router available (I have a portable router, yes, but I want to mimize and 
> ease 
> as possible). 
> 
> The goal of my project I am working of, shall be a bootable live-usb or live-
> cd, which is preconfigured with the network address, a listening nginx for 
> RTMP, automatically started X with automatically started VLC.
> 
> The user just has to start his drone software (or whatever) on a tablet or 
> mobile, input an IP in the streaming software and is ready.
> 
> Everything shall be done without any router (because of avoiding as much 
> latencies as possible).
> 
> At the moment I am stuck with the directly connection.
> 
> If someone has running this already, I would be happy for any configurations, 
> or also with the help of some priciples. 
> 
> With an ethernet cable, this is easy (using a crossover ethernet cable), but 
> how do this with wireless? Is this technically possible at all???
> 
> Thanks for any help.
> 
> Best regards
> 
> Hans 
 
Probably there are a lot more knowledgeable people here than
myself, but a naive search for "wifi adapter client host
mode linux" brings up this page.

https://woshub.com/create-wi-fi-access-point-hotspot-linux/

Most wifi adapters can operate as an access point, which is
indicated by AP and AP/VLAN appearing in the output of `iw list`.

Hope this helps


-- 
Joel Roth

Re: FOSS tool to do general stats from text indata

[Joel Roth]

On Fri, Jun 23, 2023 at 10:20:50PM +0200, Emanuel Berg wrote:
> Is there a CLI and FOSS tool that creates stats from text
> indata - e.g.,
> 
>   $ txt2stats path/to/indata/*.txt
> 
> I mean a general tool, but with options to tweak the report
> included, of course.
> 
> To produce neat stats, maybe even figures, and generate fun
> facts of the kind
> 
>The longest word that occurs more frequently than 0.01 ...
> 
>The most common words to start a sentence ...
> 
>Average paragraph length ...
> 
>And even more crazy facts and stuff that you never think
>about until the stats tell you!
> 
> What do we have on that area?

A basic search finds this web tool: 

https://www.usingenglish.com/resources/text-statistics/

Otherwise, I think you'll have to write your own -- or
hire someone (like me :^) to write one for you. 


-- 
Joel Roth

Re: fetchamil / procmail as non root : unable to call script

[Joel Roth]

On Tue, Jun 20, 2023 at 07:52:24AM +0200, BASSAGET Cédric wrote:
> Hello
> I'm using fetchamil / procmail to fetch mails from an POP server and parse
> it then launch a script or system call :
> 
> 
> # cat .fetchmailrc
> set logfile fetchmail.log
> poll imaps.dom.tld proto POP3
> user "u...@dom.tld" pass "xx" preconnect "date >> ~/fetchmail.log"
> ssl
> fetchall
> keep
> no rewrite
> mda "/usr/bin/procmail ~/.procmailrc";
> 
> # cat .procmailrc
> LOGFILE=procmail.log
> VERBOSE=yes
> :0
> * ^Message-ID: \/.*
> #| /usr/bin/curl http://mail.dom.tld/script.php?messageid=$MATCH
> | echo "whoami" > test.txt
> 
> 
> This work fine when calling fetchamil as root with "fetchamil -f
> .fetchmailrc". But when calling fetchmail from a dedicated user, the
> external script in procmail is not called. It's written in the logfile that
> :
> 
> procmail: [25332] Mon Jun 19 16:20:28 2023
> procmail: Assigning "MATCH="
> procmail: Matched "<9088600d-446a-96b4-4043-29ecd0d5a...@dom.tld>"
> procmail: Match on "^Message-ID: \/.*"
> procmail: Assigning "LASTFOLDER= echo "whoami" > test.txt"
>  Subject: test
>   Folder:  echo "whoami" > test.txt
>  1824
> procmail: Executing " echo "whoami" > test.txt"
> 
> but nothung happens.
> Am I missing something ?
> Regards
> Cédric

Are you sure that procmail is running the command in a shell? 
Could it be trying to execute a program named 'echo'. 
You might try creating a shell script you can call.


-- 
Joel Roth

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

[Joel Roth]

Can't resist adding my 2c

On Mon, Apr 03, 2023 at 11:20:26PM +0200, Eduard Bloch wrote:
> 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.

Google adopting python early on has something to do with this, too.  

> 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

I'm surprised to hear this. What is not cross-platform about CPAN?

> - 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

I'm not sure about the connection with managing variable lifecycles,
perhaps you can explain.

This pragma *does* catch a class of typos that python is vulnerable to.
Perhaps python IDEs take care of this. 

> - 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

For writing any code it helps if you understand the underlying concepts. 

> 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

Well, it doesn't make much point to discuss it here, but
I'll note that there are legitimate criticisms and solutions
for those who want them. 

Also, I seem to recall perl borrowed its OO design 
from python. Checking... yes.

"I don’t really know much about Python. I only stole its
object system for Perl 5. I have since repented."(*1)

There is a new object system being cooked up, based on
decades of experience with OO in perl and other languages. 

There is already more than enough OO goodness for me to
get my work done :-)

> - still sucky when it comes to complex data structures

Well, maybe this is not the place to discuss it, but I'm
curious if it's the syntax or the implementation that
bothers you. 

> - ECO system unfortunately slowly fading away

I would say the most highly depended-on distributions in
CPAN are well maintained and new contributions continue.
Sometimes I encounter a project (like Prima, a cross
platform GUI) that has been around for years and only now
come to my attention.  As part of developing new perl
releases, the entire of CPAN is checked to make sure new
features do not break existing code. 

> - error handling (exceptions?)

Of course this can be improved. The basic behaviors
are sufficient, tho. 

> - sometimes too rigid ways of method calling
 
Perl is rarely accused of this. Using it's the other way around--
too many ways.  Do you have a specific example?

Otherwise I don't really belong comparing the two languages
as haven't done significant work with python.

Have fun with whatever language :-)

Joel

> Python3:
> PRO:
> - HUGE and modern ECO system, and mostly good 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 ar

Re: cpan oddity

[Joel Roth]

f...@dnsbed.com wrote:
Jude DaShiell wrote:
> > I ran cpan and did quick configuration and chose sudo to elevate
> > privileges when necessary.  Unfortunately I don't have write access on
> > /usr/local/bin so cpan is crippled.

> try cpanminus?
> $ sudo apt install cpanminus

Please note that when installing via cpanminus to a 
unprivileged directory, such as $HOME/perl5,
you must also include it in your environment.
see 'man local::lib'. 


-- 
Joel Roth

Re: Issues with audio after updating

[Joel Roth]

On Sun, Oct 09, 2022 at 10:37:18AM +0200, Maximiliano Estudies wrote:
> After an update yesterday my audio stopped working. I'm currently on
> Linux version 5.19.0-2-amd64 (debian-ker...@lists.debian.org) (gcc-11
> (Debian 11.3.0-6) 11.3.0, GNU ld (GNU Binutils for Debian)
> 2.38.90.20220713) #1 SMP PREEMPT_DYNAMIC Debian 5.19.11-1 (2022-09-24)
> 
> The issue seems to be with pulseaudio, but I'm not really sure.
> pavucontrol shows "no available cards for configuration" on the
> configuration tab, and ```pactl list cards``` returns nothing but
> ```pacmd list-cards``` does returns all the available cards. There is
> a similar behavior when listing the sinks, pactl returns only the
> dummy sink and pacmd returns the alsa sink, but with state: IDLE
> 
> I have no idea how to debug this further so any tips are much appreciated!

Hiya Maximiliano,

That the two pulseaudio utilities disagree on what devices
are present, it seems plausible that some part of pa is to blame.

You may like to verify that your ALSA subsystem is working properly,
and include this in your bug report. 

For that, disable pulseaudio[1], 

systemctl --user stop pulseaudio.socket
systemctl --user stop pulseaudio.service

test play an audio file with aplay, mpv, etc.
to a device listed by aplay -l, or cat /proc/asound/cards

To start pulse audio again:

systemctl --user start pulseaudio.socket
systemctl --user start pulseaudio.service

1. 
https://askubuntu.com/questions/8425/how-to-temporarily-disable-pulseaudio-while-running-a-game-under-wine

I've not tested this myself.

cheers,

-- 
Joel Roth

Re: No HDMI Audio

[Joel Roth]

On Sun, Jul 24, 2022 at 01:37:34PM -0400, Timothy M Butterworth wrote:
> All,
> 
> I am not getting any sound through HDMI. I have an AMD Ryzen 7 4700U CPU
> with integrated Renior Graphics. I checked the device and it is not muted.
> Does anyone have any ideas?
> 
> inxi -F
> Audio:Device-1: Advanced Micro Devices [AMD/ATI] driver: snd_hda_intel
>   Device-2: Advanced Micro Devices [AMD]
> Raven/Raven2/FireFlight/Renoir Audio Processor driver: snd_rn_pci_acp3x
>   Device-3: Advanced Micro Devices [AMD] Family 17h HD Audio driver:
> snd_hda_intel
>   Sound Server: ALSA v: k5.10.0-16-amd64
> 
> *aplay -l*
> card 0: Generic [HD-Audio Generic], device 3: HDMI 0 [HDMI 0]
>  Subdevices: 1/1
>  Subdevice #0: subdevice #0
> card 0: Generic [HD-Audio Generic], device 7: HDMI 1 [HDMI 1]
>  Subdevices: 1/1
>  Subdevice #0: subdevice #0
> card 1: Generic_1 [HD-Audio Generic], device 0: ALC287 Analog [ALC287
> Analog]
>  Subdevices: 1/1
>  Subdevice #0: subdevice #0
> 
> *lspci*
> 04:00.1 Audio device: Advanced Micro Devices, Inc. [AMD/ATI] Device 1637
> 04:00.5 Multimedia controller: Advanced Micro Devices, Inc. [AMD]
> Raven/Raven2/FireFlight/Renoir Audio Processor (rev 01)
> 04:00.6 Audio device: Advanced Micro Devices, Inc. [AMD] Family 17h (Models
> 10h-1fh) HD Audio Controller

I would stop pulseaudio and test `aplay sample.wav`
using the device option to target various alsa devices.
Once you've found the HDMI device that works, you
can set it as the default device. 

-- 
Joel Roth

Re: Converting an old Chromebook to pure Debian, was: OT, Recommendation for low cost laptop

[Joel Roth]

On Mon, Jul 11, 2022 at 11:55:02AM +0100, Ottavio Caruso wrote:
> Hi, inspired by:
> 
> On 11/07/2022 08:32, john doe wrote:
> 
> > I'm looking for something cheap (max would be around 300 bucks), do you
> > have any suggestions/ideas?
> 
> 
> My local Cash-Converter/Generator(s) have plenty of old-ish Chromebooks for
> £50 or less.
> 
> I know it's possible to run some sort of Linux with trickeries like Crouton
> or similar. I wonder if it's just possible to nuke all Google related junk
> and install native Debian or is it not technically possible?

Crouton lets you install a native debian. You have to put
the Chromebook into developer mode first.

One problem I had (ten years back or so) was that every
subsequent boot brings up a dialog to *leave* developer mode. 
Saying yes clobbers the system you painstakingly installed.

cheers

> 
> -- 
> Ottavio Caruso
> 
> A: Because it messes up the order in which people normally read text.
> Q: Why is top-posting such a bad thing?
> A: Top-posting.
> Q: What is the most annoying thing in e-mail?
> 

-- 
Joel Roth

Re: Suggestions for rm [WAS: Re: Feature request: install package by passing URL to apt-get]

[Joel Roth]

On Sat, Jun 25, 2022 at 09:45:58AM -0400, Greg Wooledge wrote:
> On Fri, Jun 24, 2022 at 07:42:25PM -1000, Joel Roth wrote:
> > I list the files first:
> > 
> > ls some-pattern
> > 
> > then add a pipe to rm:
> > 
> > ls some-pattern | rm 
> > 
> > or
> > 
> > ls some-pattern | rm -rf
> 
> Those commands do not work.  rm does not read a list of files from stdin.
> 
> Even if you were to add xargs, those commands still would not work in
> all cases.  They would only work in the simplest cases, where none of
> the filenames contain whitespace, single-quote characters, or double-quote
> characters.
> 
> That's because xargs does not split its input on newlines.  It splits
> its input on "quoted words".  Like this:
> 
> unicorn:~$ echo 'a list of "quoted words"' | xargs printf '<%s>\n'
> 
> 
> 
> 
> 
> Any unbalanced single or double quote will cause an error:
> 
> unicorn:~$ echo "I don't know.mp3" | xargs printf '<%s>\n'
> xargs: unmatched single quote; by default quotes are special to xargs unless 
> you use the -0 option
> 
> 
> And even if xargs *did* have some option to split its input only on
> newlines, a filename that *contains* a newline would still break it.
> 
> Finally, ls does not always reproduce filenames exactly.  There are
> some systems (I'm not sure about all versions of Debian, but definitely
> some Unix systems) where certain characters will be printed as
> question marks by ls.  That means any file containing one of those
> characters would make the ls|xargs rm construct fail.
> 
> If you want to use ls as a preview for what will be removed, that's
> perfectly fine.  You just can't use "up arrow | xargs rm" as your
> follow-up command.  Instead, use "up arrow" and then use command editing
> to replace the ls with rm.

I was actually using 'xargs rm' in an alias. Thanks for
pointing out the limitations. 

-- 
Joel Roth

Re: Suggestions for rm [WAS: Re: Feature request: install package by passing URL to apt-get]

[Joel Roth]

On Fri, Jun 24, 2022 at 07:02:35AM +, Andrew M.A. Cater wrote:
> There are a couple of useful habits to get into when removing things:
> 
> There's an 
> 
>  rm -i
 
> Use the pwd  command to check where you are in the
> filesystem. (It may be short for "print working
> directory").
 
> If you are deleting one file - change to the directory it is in, check that
> it exists there first with the
> 
>  ls -al [filename]
> 
> command. Since the file is in the current directory,you can use the
> 
>  rm ./[filename]
> 
> [That's a period and a forward slash - limiting you to a file in the current
> directory]
> 
> Try and avoid using rm -rf and forced removal. The one exception is that
> you have to remove a non-empty directory with -rf. If you first try -r
> and it fails, that's a clue that you are actually about to delete a 
> directory.
> 
> Again, if it's a single directory, change directories to the directory
> that it is in and use the ./ The last suggestion, and it's the simplest:
> 
>  rm [filename] -rf
... 
> If you need to be doubly sure rm [filename] -irf will put in the interactive
> prompt again.
> 
> Some of this is learnt the hard way from administering one machine that
> other people relied on :)

All good suggestions, along with making backups. 

I list the files first:

ls some-pattern

then add a pipe to rm:

ls some-pattern | rm 

or

ls some-pattern | rm -rf

I know the OP said they weren't asking advice, but I can't
help putting in my two bits :-)

cheers,






> With every good wish, as ever,
> 

-- 
Joel Roth

Re: Printing the old way

[Joel Roth]

On Tue, Jun 14, 2022 at 04:59:36PM -0400, pa...@quillandmouse.com wrote:
> Folks:
> 
> Back in the dark days of early Linux, before CUPS, we printed with
> printers all the time. There was an infrastructure for doing this. Does
> anyone remember how that worked? As in, what packages were needed, etc.?

Using magicfilter with lprng lets you print a variety of 
file types. 

-- 
Joel Roth

Re: Recommendations for a home server running Debian Bullseye (11)?

[Joel Roth]

On Mon, Apr 25, 2022 at 11:27:24AM -0600, Charles Curley wrote:
> On Mon, 25 Apr 2022 16:07:28 +
> "Andrew M.A. Cater"  wrote:
> 
> > Alternatively, it might be worth looking at something small and silent
> > from a US? equivalent of QuietPC / Overclockers UK.
> 
> I've had good results from silentpc.com, in Washington state.

And on the cheaper side, not necessarily "commercial grade", is this
$600 fanless PC.

https://www.amazon.com/gp/product/B0912ZKKYM/ref=ppx_yo_dt_b_asin_title_o04_s00?ie=UTF8=1
> 
> -- 
> Does anybody read signatures any more?
> 
> https://charlescurley.com
> https://charlescurley.com/blog/
> 

-- 
Joel Roth

Review of Refind boot manager (was Re: USB UEFI recovery stick)

[Joel Roth]

On Sun, Jan 23, 2022 at 09:47:10PM -0500, Polyna-Maude Racicot-Summerside wrote:

> rEFInd is a fork of the rEFIt boot manager. Like rEFIt, rEFInd can
> auto-detect your installed EFI boot loaders and it presents a pretty GUI
> menu of boot options. rEFInd goes beyond rEFIt in that rEFInd better
> handles systems with many boot loaders, gives better control over the
> boot loader search process, and provides the ability for users to define
> their own boot loader entries.
> 
> https://www.rodsbooks.com/refind/installing.html

I just migrated my system to new hardware with a GPT
formatted disk. (Well, almost migrated; I still have to
figure out the driver for the wifi chip.)

There was no reason not to reformat the disk as MBR (it is
only 500GB) but anyway decided to try to use the GPT/EFI
format. 

I repartitioned, leaving only the EFI system partition.
For the bootloader I've been avoiding Grub due to its
complicated config files. (Maybe I'm just contrarian ;-)

lilo seemed to barf. I didn't get beyond the warnings that
there was some partition table error and that I was attempting
something dangerous. 

My next try was Refind, a friendly OS-agnostic bootloader
that handles EFI. (I'd used it once previously when
installing devuan on a Intel Mac.)

The main limitation compared to Grub is that Refind doesn't
let you edit the command line prior to booting. 

Despite lots of detail on the website, the actual
configuration is simple, even compared to lilo.conf.  There
are just three lines in /boot/refind_linux.conf containing
options for normal, single-user mode and minimal boot. The
boot partition(s) are identified by UUID. Refind generates
menu choices for all kernels found.  

Refind comes with several utility scripts. I used
refind-install to generate refind_linux.conf, but needed to
tweak the result for it to correctly find the initrd images
matching the kernel versions.  

So in conclusion, Refind is easy to install and configure,
and suits my purposes just fine. 

> -- 
> Polyna-Maude R.-Summerside
> -Be smart, Be wise, Support opensource development

-- 
Joel Roth

Re: Debian-friendly laptop

[Joel Roth]

George Shuklin wrote:
> On 5/18/21 10:49 PM, George Shuklin wrote:
> > I'm trying to choose between Purism and System76, and, as far as I
> > understand they both supports linux very well, but..

> That leave System76. But they have only models with nvidia available, and we
> all knows how open are nvidia blobs.
> 

I wonder what support you need. Many others have mentioned
that Thinkpads are quite linux friendly. A great resource,
if you're considering that direction is
https://www.thinkwiki.org/wiki/ThinkWiki .

I looked at these commercially supported laptops years ago,
found that they didn't offer me much over community
supported Thinkpads. 

-- 
Joel Roth

Re: Putting small web site online

[Joel Roth]

Hi john,

john doe wrote:
> Debians,
> 
> I need to have a small web site online but I don't have a commercial
> link nor a server at home that can be publickly available.
> 
> I'm planning to test/build the web site locally then have it published
> where it is publickly available.
> 
> I'm thinking of using Gitlab to host my web site do you have a better
> solution when you can't host your web site yourself?

freeshell.de is free with registration. They run debian.
You get a full shell environment, which probably means more
flexibility than gitlab. You can use PHP, python or perl
with a mysql database if you like. And the admin answers
emails. 

Speaking as a satisfied (mostly dormant) user. 


-- 
Joel Roth

Re: Multichannel audio playback

[Joel Roth]

On Wed, Apr 28, 2021 at 06:21:27PM +, Rodolfo Medina wrote:
> Joel Roth  writes:
> 
> > I don't think there is a problem with ecasound. 
> > For comparison, you can try
> >
> > aplay -D  test.wav
> >
> > You get the device name from aplay -L.
> > Please also show the output from 
> >
> > cat /proc/asound/cards
> >
> > With your ecasound examples, the -a argument is not necessary, and
> > selecting channels is not its purpose.
> 
> 
> Now it works: 
> 
> 
> $ ecasound -a:1,2,3 -i:sndfile,3canali.wav -o alsaplugin,2,0
> **
> *ecasound v2.9.1 (C) 1997-2014 Kai Vehmanen and others
> **
> (eca-chainsetup) Chainsetup "untitled-chainsetup"
> (eca-chainsetup) NOTE: Real-time configuration, but insufficient privileges
> ... to utilize real-time scheduling (SCHED_FIFO). With small buffersizes,
> ... this may cause audible glitches during processing.
> (eca-chainsetup) "rt" buffering mode selected.
> (audioio_sndfile) Using libsndfile to open file "3canali.wav" for reading.
> (eca-chainsetup) NOTE: using existing audio parameters -f:s16_le,3,48000
> ... for object 'sndfile' (tried to open with -f:s16_le,2,44100).
> (eca-chainsetup) Opened input "sndfile", mode "read". Format: s16_le,
> ... channels 3, srate 48000, interleaved (locked params).
> (eca-chainsetup) Opened output "alsaplugin", mode "write". Format: s16_le,
> ... channels 2, srate 48000, interleaved.
> - [ Connected chainsetup: "untitled-chainsetup" ] 
> - [ Controller/Starting batch processing ] ---
> - [ Engine - Driver start ] --
> 
> - [ Engine - Processing finished ] ---
> - [ Controller/Batch processing finished (0) ] ---
> - [ Controller/Processing stopped (cond) ] ---
> - [ Engine exiting ] -
> (eca-control-objects) Disconnecting chainsetup:  "untitled-chainsetup".
> 
> 
> So we could consider problem 1) solved.
> 
> Let's try to go to problem 2).  The above command properly sends channel 1 to
> loudspeaker 1 and channel 2 to loudspeaker 2, but it doesn't manage to send
> channel 3 into loudspeaker 3.  Actually, the Behringer seems to behave in
> stereo mode.  But the strange thing is that outputs `B' (i.e. 3 and 4) do not
> work at all and seems to be dead.  So please help: how can I - if ever 
> possible
> - send channel 3 to loudspeaker 3?  And, more in general: is it possible, even
> with some other device, to achive what I want?

The soundcard capabilities that linux sees are documented
in the /proc/asound filesystem. That's worth investigating.

Here's one reference to what you can find there:

https://www.ti.com/lit/an/sprac10/sprac10.pdf

Another helpful source for guidance would be the Linux Audio
Users mailing list, with many knowledgeable and helpful
members. 

cheers


> Thanks,
> 
> Rodolfo
> 

-- 
Joel Roth

Re: Multichannel audio playback

[Joel Roth]

On Wed, Apr 28, 2021 at 11:48:33AM +, Rodolfo Medina wrote:
> didier gaumet  writes:
> 
> > Le 28/04/2021 à 09:37, Rodolfo Medina a écrit :
> > [...]
> >> First of all, problem 1): same error now with a 2-channel and a 3-channel
> >> file:
> > [...]
> >> ERROR:  Connecting chainsetup failed: "Enabling chainsetup: AUDIOIO-ALSA:
> >> ... Channel count 2 is out of range!"
> > [...]
> >
> > http://nosignal.fi/ecasound/Documentation/examples.html#multichannel
> >
> > I do not use ecasound but it seems to me that this indicates that by default
> > ecasound is in a one channel (mono, channel #1) mode and that channels have
> > to be specified with -a, perhaps like this:
> >
> > $ ecasound -a:1,2,3 -i:sndfile,3canali.wav -o alsahw,2,0
> >  or perhaps
> > $ ecasound -a:all -i:sndfile,3canali.wav -o alsahw,2,0
> 
> 
> Thanks, Didier:
> 
> it seems to work with 2 channels:
> 
> $ ecasound -a:1,2 -i:sndfile,2canali.wav -o alsahw,2,0
> **
> *ecasound v2.9.1 (C) 1997-2014 Kai Vehmanen and others
> **
> (eca-chainsetup) Chainsetup "untitled-chainsetup"
> (eca-chainsetup) NOTE: Real-time configuration, but insufficient privileges
> ... to utilize real-time scheduling (SCHED_FIFO). With small buffersizes,
> ... this may cause audible glitches during processing.
> (eca-chainsetup) "rt" buffering mode selected.
> (audioio_sndfile) Using libsndfile to open file "2canali.wav" for reading.
> (eca-chainsetup) NOTE: using existing audio parameters -f:s16_le,2,48000
> ... for object 'sndfile' (tried to open with -f:s16_le,2,44100).
> (eca-chainsetup) Opened input "sndfile", mode "read". Format: s16_le,
> ... channels 2, srate 48000, interleaved (locked params).
> (eca-chainsetup) Opened output "alsahw", mode "write". Format: s16_le,
> ... channels 2, srate 48000, interleaved.
> - [ Connected chainsetup: "untitled-chainsetup" ] 
> - [ Controller/Starting batch processing ] ---
> - [ Engine - Driver start ] --
> 
> - [ Engine - Processing finished ] ---
> - [ Controller/Batch processing finished (0) ] ---
> - [ Controller/Processing stopped (cond) ] ---
> - [ Engine exiting ] -
> (eca-control-objects) Disconnecting chainsetup:  "untitled-chainsetup".
> 
> 
> but not with 3:
> 
> $ ecasound -a:1,2,3 -i:sndfile,3canali.wav -o alsahw,2,0
> **
> *ecasound v2.9.1 (C) 1997-2014 Kai Vehmanen and others
> **
> (eca-chainsetup) Chainsetup "untitled-chainsetup"
> (eca-chainsetup) NOTE: Real-time configuration, but insufficient privileges
> ... to utilize real-time scheduling (SCHED_FIFO). With small buffersizes,
> ... this may cause audible glitches during processing.
> (eca-chainsetup) "rt" buffering mode selected.
> (audioio_sndfile) Using libsndfile to open file "3canali.wav" for reading.
> (eca-chainsetup) NOTE: using existing audio parameters -f:s16_le,3,48000
> ... for object 'sndfile' (tried to open with -f:s16_le,2,44100).
> (eca-chainsetup) Opened input "sndfile", mode "read". Format: s16_le,
> ... channels 3, srate 48000, interleaved (locked params).
> ERROR:  Connecting chainsetup failed: "Enabling chainsetup: AUDIOIO-ALSA:
> ... Channel count 2 is out of range!"
> 

Hi Rodolfo,

I don't think there is a problem with ecasound. 
For comparison, you can try

aplay -D  test.wav

You get the device name from aplay -L.
Please also show the output from 

cat /proc/asound/cards

With your ecasound examples, the -a argument is not necessary, and
selecting channels is not its purpose. 

cheers

> Please help...
> 
> Rodolfo
> 

-- 
Joel Roth

Re: Multichannel audio playback

[Joel Roth]

--
--- Begin Message ---
On Wed, Apr 28, 2021 at 07:37:27AM +, Rodolfo Medina wrote:
> Joel Roth  writes:
> 
> > On Tue, Apr 27, 2021 at 10:22:11AM -1000, Joel Roth wrote:
> >> 
> >> ecasound -i:libsndfile,audiofile -o alsa
> >  
> > that should be -i:sndfile 
> 
> 
> Thanks, Joel...
> 
> 
> 2 problems seem to be involved:
> 
> 1) make ecasound read and play the file;
> 
> 2) properly use all those many outputs on the Behringer's back.
> 
> (Up to now, I can't make the `B' Playback Outputs work yet.  Only `A' seem to
> respond.)
> 
> First of all, problem 1): same error now with a 2-channel and a 3-channel 
> file:
> 
> 
> $ ecasound -i:2canali.wav -o alsahw,2,0
> **
> *ecasound v2.9.1 (C) 1997-2014 Kai Vehmanen and others
> **
> (eca-chainsetup) Chainsetup "untitled-chainsetup"
> (eca-chainsetup) NOTE: Real-time configuration, but insufficient privileges
> ... to utilize real-time scheduling (SCHED_FIFO). With small buffersizes,
> ... this may cause audible glitches during processing.
> (eca-chainsetup) "rt" buffering mode selected.
> (eca-chainsetup) NOTE: using existing audio parameters -f:s16_le,2,48000
> ... for object '2canali.wav' (tried to open with -f:s16_le,2,44100).
> (eca-chainsetup) Opened input "2canali.wav", mode "read". Format: s16_le,
> ... channels 2, srate 48000, interleaved (locked params).
> ERROR:  Connecting chainsetup failed: "Enabling chainsetup: AUDIOIO-ALSA:
> ... Channel count 2 is out of range!"
> 
> 
> 
> $ ecasound -i:sndfile,3canali.wav -o alsahw,2,0
> **
> *ecasound v2.9.1 (C) 1997-2014 Kai Vehmanen and others
> **
> (eca-chainsetup) Chainsetup "untitled-chainsetup"
> (eca-chainsetup) NOTE: Real-time configuration, but insufficient privileges
> ... to utilize real-time scheduling (SCHED_FIFO). With small buffersizes,
> ... this may cause audible glitches during processing.
> (eca-chainsetup) "rt" buffering mode selected.
> (audioio_sndfile) Using libsndfile to open file "3canali.wav" for reading.
> (eca-chainsetup) NOTE: using existing audio parameters -f:s16_le,3,48000
> ... for object 'sndfile' (tried to open with -f:s16_le,2,44100).
> (eca-chainsetup) Opened input "sndfile", mode "read". Format: s16_le,
> ... channels 3, srate 48000, interleaved (locked params).
> ERROR:  Connecting chainsetup failed: "Enabling chainsetup: AUDIOIO-ALSA:
> ... Channel count 2 is out of range!"
> 
> 
> Both files were created via `sox -M' command from respectively 2 and 3 mono
> files.
 
Ecasound can correctly identify and open the audiofile in
both of your tests above, with and without libsndfile. You
can see it identifies the sample rate and channel count
correctly both times. 

The problem is with the ALSA output. Something is wrong with
the soundcard device, its driver or its alsa interface via
alsahw,2,0. 

You could try listing the sound devices. 

aplay -L

Choose one you want and use -o alsa,DEVICENAME

You can also cat /proc/asound/cards.

Oh, and make sure you don't have pulseaudio running.

cheers

> Please help if you can.
 
> Thanks
> 
> Rodolfo
> 

-- 
Joel Roth
--- End Message ---

Re: Multichannel audio playback (was: Multichannel audio listening)

[Joel Roth]

On Tue, Apr 27, 2021 at 10:22:11AM -1000, Joel Roth wrote:
> 
> ecasound -i:libsndfile,audiofile -o alsa
 
that should be -i:sndfile 


-- 
Joel Roth

Re: Multichannel audio playback (was: Multichannel audio listening)

[Joel Roth]

On Tue, Apr 27, 2021, Rodolfo Medina wrote:
 
> Four years ago this interesting thread.  Now I've purchased a Behringer
> UMC404HD 4-channels audio interface and finally want to do my experiment.  Now
> I have a 3-channel .wav file and 3 loudspeakers connected to Behringer's
> output: I want to playback each channel to a different speaker.  I get error:
> 
> $ ecasound -i 3canali.wav -f:16,3,44100 -o alsahw,2,0
> **
> *ecasound v2.9.1 (C) 1997-2014 Kai Vehmanen and others
> **
> (eca-chainsetup) Chainsetup "untitled-chainsetup"
> (eca-chainsetup) NOTE: Real-time configuration, but insufficient privileges
> ... to utilize real-time scheduling (SCHED_FIFO). With small buffersizes,
> ... this may cause audible glitches during processing.
> (eca-chainsetup) "rt" buffering mode selected.
> ERROR:  Connecting chainsetup failed: "Enabling chainsetup: AUDIOIO-WAVE:
> ... Only WAVE_FORMAT_PCM and WAVE_FORMAT_IEEE_FLOAT are supported."

Hi Rodolfo,

As didier gaumet writes, 3canali.wav is not one of
ecasound's supported formats. You should find out what it
is. Maybe try 

file 3canali.wav

Also, ecasound can use libsndfile to expand the range of 
compatible audiofiles for reading and writing.

ecasound -i:libsndfile,audiofile -o alsa

If the extension doesn't match the file format, you can
specify the file type, e.g.

ecasound -i:libsndfile,audiofile,aiff -o alsa

cheers

-- 
Joel Roth

Re: Cannot mount DFS shares after upgrade to 10.9

[Joel Davies]

> So this is the case, I was able to reproduce the issue further on

> curren 4.19.184 and with the mentioned commit reverted on top the bug

> is gone.

>

> I will report it upstream.



Thank you, Salvatore. It's been a few years since I compiled a kernel so it was 
going to take me quite a while to test.



Joel

Cannot mount DFS shares after upgrade to 10.9

[Joel Davies]

We have some autofs mounts that stopped working after the 10.9 updates were 
applied. Seems that the kernel update is the cause - I applied just the kernel 
update and same thing happens. The mounts that stopped working were targeting 
Windows domain DFS shares. The DFS root can still be mounted 
(//domain.local/DFS) which we may have to end up using but then will have to 
reconfigure stuff. Subfolders of normal shares also can still be mounted 
(//Server.domain.local/Share/Subfolder). It is just the DFS shares that don't 
work currently (//domain.local/DFS/Share).

Here is what happens mounting the shares manually:

Under kernel 4.19.0-14-amd64 (previous to 10.9 updates):

$sudo mount -t cifs -o username=user,vers=3.0 //domain.local/DFS/Share /mnt 
--verbose
Password for user@//domain.local/DFS/Share:  *
mount.cifs kernel mount options: 
ip=10.x.x.x,unc=\\domain.local\DFS,vers=3.0,user=user,prefixpath=Share,pass=

Normally no messages in syslog and share is successfully mounted.

After updating kernel to 4.19.0-16-amd64:

$sudo mount -t cifs -o username=user, //domain.local/DFS/Share /mnt --verbose
Password for user@//domain.local/DFS/Share:  *
mount.cifs kernel mount options: 
ip=10.x.x.x,unc=\\domain.local\DFS,vers=3.0,user=user,prefixpath=Share,pass=
mount error(2): No such file or directory
Refer to the mount.cifs(8) manual page (e.g. man mount.cifs)

Syslog shows:
CIFS VFS: cifs_read_super: get root inode failed

I didn't find much of use online to troubleshoot that message. I did try 
different values of vers option as well as each of serverino and noserverino 
and several combinations of both - no difference. Any suggestions?

(Occasionally syslog also shows "CIFS VFS: error -2 on ioctl to get interface 
list". This happened about one in five mount attempts both before and after the 
kernel update, regardless of whether or not the share mounted. Since it did 
appear before the update when mounting always worked, I am assuming it is not 
relevant.)

Thanks,
Joel Davies

Re: Blind and Low-Vision UNIX Users Group

[Joel Roth]

On Thu, Mar 04, 2021 at 05:34:27PM -0500, Dan Ritter wrote:
> 
> Since I know we have some people on this list who might be
> interested in this --
> 
> https://blvuug.org/
> 
> is the home of the new Blind and Low-Vision UNIX Users Group.
> 
> They're just getting started.
> 
> -dsr-

Another list for this purpose:

Blinux-list mailing list
 
blinux-l...@redhat.com  
 
https://listman.redhat.com/mailman/listinfo/blinux-list



-- 
Joel Roth

Re: ALSA bug? Laptop headphone output depending on speaker channel setting

[Joel Roth]

Brendon Higgins wrote:
> Seasonal greetings, Debian users!
> 
> I was wondering if anyone has pointers for an audio configuration issue. I 
> have 
> Debian Testing running on a Dell laptop (Latitude E6320, audio chipset 
> detected as HDA Intel PCH, 92HD90BXX). The speaker output works fine. When I 
> plug in headphones, the system seems to detect that okay: it mutes the 
> speakers ("Speaker+LO" channel in alsamixer) and unmutes the headphones 
> ("Headphone" channel).
> 
> The problem is that no sound comes out of my headphones.
> 
> I've noticed that I have to also unmute and restore volume of the Speaker+LO 
> channel to hear anything in the headphones (meanwhile, with headphones 
> connected, the laptop speakers are silent as you would expect). The Headphone 
> volume setting doesn't seem to matter at all, although muting that does also 
> cause the headphones to mute.
> 
> It all used to work fine, but probably a couple of years ago it started 
> having 
> this behaviour. I suspect it was a kernel update, or maybe ALSA library - I 
> can't be certain - and put it off hoping the bug might just get fixed...
> 
> Any suggestions? I looked at the hdajackretask app,  but no idea where I 
> would 
> even start there.

Do you have pulseaudio running on your system?

It is a layer above ALSA, and could be related to your
issue.

> Thanks in advance,
> Brendon
> 
> 
> 

-- 
Joel Roth

Re: Most maintainable way to install perl modules on Debian sysetms

[Joel Roth]

Hi Michael,

On Sun, Nov 15, 2020 at 08:50:57PM +, Michael Grant wrote:
> I try to keep my systems as up to date as possible.  I use apt update 
> regularly.  When I can install a perl module from apt, I usually do so 
> because then apt update picks up new versions of it.  When I install 
> something which has a dependency on a perl module in apt, that module gets 
> installed.  Things just work and I am unstressed and happy.
> 
> However, what should I do if I need a perl module that someone hasn’t kindly 
> created a package for?  I know I can install it from cpan.  But once I do 
> that, then I have to keep that module up to date via cpan.  Perhaps I’m not 
> alone and someone has a way to cross maintain these things that I’m unaware 
> of?
> 
> So I yesterday, I decided to install a perl module from cpan.  I ran cpan for 
> the first time in a long time, it asked me some question which I took the 
> default (maybe I shouldn’t have!), then I installed Mail::DMARC.  It 
> downloaded the source, grinded away for several minutes and when it was 
> finished, it successfully installed the module.  Miraculous!
> 
> But it installed it in /usr/share/perl/5.30/Mail/DMARC.
> 
> I thought that’s odd, why would it install it in a folder that has a perl 
> version?  Why not somewhere a bit more general like /usr/share/perl5?

The directory location may not suit you, but for the short
term you should be able to get the modules you want
installed using root and the cpan client. 

Another way is to use a module called local::lib, 
that let's you have a tree of perl modules in 
an unprivileged directory (by default 
$HOME/perl5). 

It requires setting some environment variables.
Then you need a cpan client that will work with
local::lib. That is cpanm.

So apt-get install liblocal-lib-perl libapp-cpanminus-perl

Setup environment according to `man local::lib`

Then install any module you need with

cpanm Some::Module 

cpanm will respect the environment variable and install
it with all dependencies. You will need to have a
compiler present for building modules with C extensions. 
 
> What happens when Perl gets updated and /usr/share/perl/5.30/ is no longer in 
> perl’s search path for its modules?  I’m worried that using this DMARC perl 
> module, updating perl could just break mail someday!  Shouldn’t the default 
> for cpan be something other than this version based directory on Debian?  
> Should I worry about this?  (this may be more a CPAN question than a Debian 
> question to be honest).

This approach, using an unprivileged directory tree that is
merged with debian perl libraries, will work 
with an upgraded system perl if the modules
are pure perl. Modules with compiled C code
(using perl's XS interface language) will *not* be
compatible with an upgraded binary. 

> So my question is, is there a recommended, maintainable way to install perl 
> modules on Debian that are not installed by apt-get such that things get 
> updated properly?
 
If the locally installed modules are pure perl, they should
be compatible with a newer perl. You will have to determine
if and when you upgrade these modules.

It's possible the same module could be installed in your
local::lib as in the system libraries.  In that case,
libraries will be searched in order specfied by the
$PERL5LIB environment variable. 

FWIW I used to use local::lib, long since prefer to compile my 
own perl interpreters and libraries. Perlbrew and plenv
are tools for managing multiple perl versions. 

have fun,

> I suppose this could even be a general question when you consider other 
> things like python, php, nodejs  and others, all of which have their own 
> module systems.  Feels like there’s a need for a sort of meta-module system 
> that works with apt, but I’m not sure which is why I am asking here.
> 
> Suggestions and advice welcome!
> 
> Michael Grant

-- 
Joel Roth

Re: Plus de son sur la sortie casque (son reste sur les HP)

[BERTRAND Joel]

Solution :

options snd-hda-intel model=inv-dmic

J'avoue, ce n'était pas trivial ;-)

En espérant que ça serve.

JKB

Re: Plus de son sur la sortie casque (son reste sur les HP)

[BERTRAND Joel]

Désolé pour la redite, le wifi des Intercités de la SNCF est... erratique.

Plus de son sur la sortie casque (son reste sur les HP)

[BERTRAND Joel]

Bonjour à tous,

	Depuis quelque temps, je m'aperçois que je n'ai plus aucun son dans la 
sortie casque sur un Toshiba Pxxx qui n'est pas très récent. Le casque 
n'est pas en cause et fonctionne très bien sur une autre machine.


	Ce casque est pourtant reconnu puisque pavucontrol voit le casque en 
périphérique de sortie mais ce sont toujours les haut-parleur du 
portable qui restent actifs. Je viens de regarder toutes les options 
d'alsamixer sans trouver de chose aberrante, le son sort toujours des 
haut parleur internes. J'ai aussi essayé la sortie ligne, même motif, 
même punition.


Le portable en question utilise pulseaudio. Il comporte deux cartes son 
:
cauchy:[~] > lspci | grep -i audio
00:1b.0 Audio device: Intel Corporation 82801H (ICH8 Family) HD Audio 
Controller (rev 03)
01:00.1 Audio device: Advanced Micro Devices, Inc. [AMD/ATI] RV630 HDMI 
Audio [Radeon HD 2600 PRO/XT / HD 3610]


	Les firwares des deux cartes sont insérés. Tout se passe comme si le 
pilote refusait de commuter la sortie son sur la prise casque alors 
qu'elle est bien détectée.


	J'ai bien essayé dans /etc/modprobe.d de rajouter options snd-hda-intel 
model=laptop ou model=toshiba, rien y fait.


Je suis preneur de toute idée...

Bien cordialement,

JKB

Plus de son sur la sortie casque (son reste sur les HP)

[BERTRAND Joel]

Bonjour à tous,

	Depuis quelque temps, je m'aperçois que je n'ai plus aucun son dans la 
sortie casque sur un Toshiba Pxxx qui n'est pas très récent. Le casque 
n'est pas en cause et fonctionne très bien sur une autre machine.


	Ce casque est pourtant reconnu puisque pavucontrol voit le casque en 
périphérique de sortie mais ce sont toujours les haut-parleur du 
portable qui restent actifs. Je viens de regarder toutes les options 
d'alsamixer sans trouver de chose aberrante, le son sort toujours des 
haut parleur internes. J'ai aussi essayé la sortie ligne, même motif, 
même punition.


Le portable en question utilise pulseaudio. Il comporte deux cartes son 
:
cauchy:[~] > lspci | grep -i audio
00:1b.0 Audio device: Intel Corporation 82801H (ICH8 Family) HD Audio 
Controller (rev 03)
01:00.1 Audio device: Advanced Micro Devices, Inc. [AMD/ATI] RV630 HDMI 
Audio [Radeon HD 2600 PRO/XT / HD 3610]


	Les firwares des deux cartes sont insérés. Tout se passe comme si le 
pilote refusait de commuter la sortie son sur la prise casque alors 
qu'elle est bien détectée.


	J'ai bien essayé dans /etc/modprobe.d de rajouter options snd-hda-intel 
model=laptop, rien y fait.


Je suis preneur de toute idée...

Bien cordialement,

JKB

Zoom conferencing

[Joel Rees]

(I hope no one gets upset about double posting debian and ubuntu users
lists.)

Questions about zoom -- www.zoom.us

Anyone using it?

Issues?

Known reasons they don't put it in the general repositories?

Re: systemd et swap sur iscsi

[BERTRAND Joel]

didier gaumet a écrit :

Justement. Il me semble que le mieux, là-dedans, c'est Requires...


(Google Groups  a l'air toujours dans la panade alors je te réponds
directement à partir de l'archive mail trouvée sur le net)

Je peux avoir mal compris la page man de systemd.unit(5) mais ce que
j'en retire c'est que :

- "Requires=" lance le démarrage des dépendances sans attendre leur
bonne complétion (fonctionnement asynchrone et gain de temps par
parallélisation)
- dans le service A, la clause "After=B" inspecte quelque part une liste
de services à démarrer et réordonne la file pour que le démarrage de A
n'intervienne qu'en cas de bonne complétion du démarrage de B et après
celle-ci (fonctionnement synchrone).
  Mais attention: avec "After=B", A peut être démarré sans B du moment
que B ne figure pas dans la file de services à démarrer à ce moment-là.

J'en déduis que pour ton usage il te faudrait les deux clauses (ça
semble orthodoxe dans la page man):
Requires=B
After=B



Je vais essayer cela.

Bien cordialement,

JKB

systemd et swap sur iscsi

[BERTRAND Joel]

Bonsoir à tous,

	Petite question du soir. J'essaie toujours de mettre en place un swap 
sur iSCSI. Au démarrage, forcément, l'ignoble systemd ne monte pas les 
choses correctement.


	Dans l'ordre, il faut d'abord que iscsi (initiator) soit lancé, puis le 
swap activé. J'ai donc rajouté dans /etc/systemd/system/swap.target :


[Unit]
Description=Swap
Requires=open-iscsi.service
Documentation=man:systemd.special(7)

	Visiblement, ce n'est pas ça. Comment faire pour faire comprendre à 
systemd qu'il doit d'abord lancer iscsi ?


Merci de vos lumières,

JKB

[iSCSI] NetBSD 9.0 target, Linux Debian initiator

[BERTRAND Joel]

Bonjour à tous,

Je bute sur un problème surprenant avec un swap sur iSCSI
(workstation disless).

Les logs de la target sont les suivants :

Feb 21 19:00:48 legendre iscsi-target: > iSCSI Normal login  successful 
from iqn.1993-08.org.debian:01:f44d59dfc4a1 on 192.168.10.103 disk 0, 
ISID 9613344773, TSIH 558
Feb 21 19:00:48 legendre iscsi-target: pid 
20853:/usr/src/netbsd-9/src/external/bsd/iscsi/lib/../dist/src/lib/disk.c:1405: 
***ERROR*** bytec > 1310720m
Feb 21 19:00:48 legendre iscsi-target: pid 
20853:/usr/src/netbsd-9/src/external/bsd/iscsi/lib/../dist/src/lib/disk.c:1266: 
***ERROR*** disk_write() failed
Feb 21 19:00:48 legendre iscsi-target: pid 
20853:/usr/src/netbsd-9/src/external/bsd/iscsi/lib/../dist/src/lib/target.c:475: 
***ERROR*** scsi_cmd.bytes_recv
Feb 21 19:00:48 legendre iscsi-target: pid 
20853:/usr/src/netbsd-9/src/external/bsd/iscsi/lib/../dist/src/lib/target.c:1398: 
***ERROR*** scsi_command_t() failed


On y voit une erreur, un coup de pied aux fesses, et 
l'initiateur se reconnecte. Je suis allé voir la ligne 1405 de disk.c. 
L'erreur est remontée à chaque fois qu'on essaye d'écrire plus d'un Mo 
sur le disque en une fois.


D'où ma question. Y a-t-il un moyen de demander à l'initiateur 
de ne as dépasser cette valeur ? Je n'ai rien trouvé...


Bien cordialement,

JKB

Re: [HS] Loi Avia : le contrôle tous azimuts, un danger réel de censures

[BERTRAND Joel]

Yoann LE BARS a écrit :


Salut à tous !

Le 18/02/2020 à 18:56, BERTRAND Joel a écrit :

 Certes. Mais juste en dessous de la juridiction pénale se trouve la
juridiction administrative en terme de respect du droit. Dans les TJ,
c'est beaucoup, mais lors beaucoup plus aléatoire. J'ai des dossiers
pleins dans lesquels les magistrats des TJ violent ouvertement le code
de procédure avant, pendant et après l'audience, naturellement en toute
impunité parce que pour le leur faire remarquer, il faut se constituer
avec... un avocat du barreau local que vous ne trouverez jamais.


Qu’il y ait des juges incompétents, c’est certain. Que la procédure ne
soit pas toujours respectée, ça l’est également. Au demeurant, est-ce
que ça arrangerait quoique ce soit de se passer définitivement d’un juge
et de s’en remettre à des sociétés privées pour faire appliquer la loi ?


	Ben, entre nous, c'est kif kif, hein. Seuls ceux qui n'ont pas affaire 
à la justice tous les jours peuvent garder quelques illusions. Là, pour 
la petite histoire, je suis en train de démerder une situation bien 
plantée par un avocat et une cour d'appel (alors que les pièces sont 
indiscutables). Problème : la décision a la force de la chose jugée avec 
tout ce que ça implique et risque de ruiner plusieurs familles. Au moins 
avec une boîte privée, tu n'as pas ce risque.



Ceci sachant que sur le sujet particulier, à savoir les expressions de
haine, les divers réseaux sociaux ont clairement établis qu’ils ne sont
pas capable de les réguler – voir les analyses de la Quadrature du net :

https://www.laquadrature.net/loihaine/
https://www.laquadrature.net/2019/05/09/une-loi-contre-la-haine-anti-macron/

Donc, le système judiciaire est loin d’être parfait, c’est entendu. La
bonne solution est-elle donc de s’en passer ?


	Le solution est de rendre les magistrats, parquet comme siège, 
responsables devant les électeurs (et directement responsables devant 
leurs pairs parce que seul l'état est responsable de leurs actes, charge 
à lui de lancer une procédure récusatoire, ce qu'il ne fait quasiment 
jamais !).

Re: [HS] Loi Avia : le contrôle tous azimuts, un danger réel de censures

[BERTRAND Joel]

Yoann LE BARS a écrit :


Salut à tous !

Le 18/02/2020 à 14:07, Philippe a écrit :

C'est l'objet de la loi Avia, quand bien même elle repose sur des considérants 
biaisés : difficile pour le juge français d'agir sur
l'éditeur dont le modèle économique repose sur la permissivité (permettre à 
quiconque d'éditer un contenu), puissant et situé
dans un pays étranger.


Certes, cependant c’est bien le travail du juge que de trancher des cas
difficile, leur formation prend tout cela en compte.


Mouarf.

	Ça fait plus de dix ans que je défends des gens devant les tribunaux. 
Le jour où je trouverai un magistrat connaissant le droit et 
l'appliquant, je le ferai empailler. Les seuls, en France, qui le 
respectent à peu près, ce sont les magistrats de la juridiction pénale. 
Les autres... il faut leur tordre le bras pour que ça arrive.


	Pour la petite histoire, je leur balance actuellement dans la figure le 
141-3 du code de l'organisation judiciaire histoire d'avoir une simple 
mise en état des dossiers ! Sinon, ça passe à la trappe, alors que c'est 
le b-a-ba de la procédure.



Tout le problème de la loi Avia, c’est que justement, contre toutes les
recommandations qui ont été émises, elle se passe d’un juge. C’est
extrêmement problématique.


	Certes. Mais juste en dessous de la juridiction pénale se trouve la 
juridiction administrative en terme de respect du droit. Dans les TJ, 
c'est beaucoup, mais lors beaucoup plus aléatoire. J'ai des dossiers 
pleins dans lesquels les magistrats des TJ violent ouvertement le code 
de procédure avant, pendant et après l'audience, naturellement en toute 
impunité parce que pour le leur faire remarquer, il faut se constituer 
avec... un avocat du barreau local que vous ne trouverez jamais.


JKB

Re: hplip

[BERTRAND Joel]

Bonjour à tous,

	Ce matin, je décide de creuser le problème. hp-setup -i termine 
correctement et le plugin s'installer tout seul (!)


	Naturellement, je n'ai _jamais_ redémarré la machine en question. Je 
n'ai pas installé ou réinstallé d'autres paquets. Je suppose que les 
unattented-upgrades viennent mettre un peu de bazar là-dedans et je ne 
saurai jamais quel était le truc fautif.


JKB

Re: Backup Times on a Linux desktop

[Joel Roth]

On Mon, Nov 04, 2019, Charles Curley wrote:
> On Mon, 4 Nov 2019 06:01:54 -1000
> Joel Roth  wrote:
> 
> > These days I use rsync with the --link-dest option to make
> > complete Time-Machine(tm) style backups using hardlinks to
> > avoid file duplication in the common case.  In this
> > scenario, the top-level directory is typically named based
> > on date and time, e.g. back-2019.11.04-05:32:06.
> 
> Take a look at rsnapshot. You have pretty well described it.

Looks like a featureful, capable, and thoroughly debugged
front end to rsync with the --link-dest option. 

Thanks, I'll fool around with this. 

Also for the explanations about file integrity issues when
databases are involved. 

--
Joel Roth

Re: Backup Times on a Linux desktop

[Joel Roth]

On Sat, Nov 02, 2019, Konstantin Nebel wrote:
> So now I am thinking. How should I approach backups. On windows it does
> magically backups and remind me when they didnt run for a while. I like that
> attitude.
(...) 
>  I like to turn off
> my computer at night. So a backup running in night is not really an option
> unless I do wake on lan and run backup and then turn off. 
(...)

Someone already recommended setting up a cron job for triggering backups
on a regular schedule. That takes care of the automagic part.

These days I use rsync with the --link-dest option to make
complete Time-Machine(tm) style backups using hardlinks to
avoid file duplication in the common case.  In this
scenario, the top-level directory is typically named based
on date and time, e.g. back-2019.11.04-05:32:06.

I usually make backups while the system is running, although
I'm not sure it's considered kosher. It takes around 10% of
CPU on my i5 system.

> Whoever read till the end Im thankful and ready to hear your opinion.
 
> Cheers
> Konstantin

--
Joel Roth

Re: Programming pointer? -- options to run Python3 from web page (not via Flask/cgi/Django)

[Joel Roth]

On Wed, Jul 24, 2019 at 02:16:09PM -0400, deb wrote:
> 
> I have a large static html/AJAX .js apache2 site.
> 
> If I want to have a server-side script just to
> handle a contact  and push mail out;
> is there a non-(Django/cgi**/Flask) way to
> run a small Python3 script to do this?

What not cgi? Isn't that the original and minimal
way to accomplish this? 

> The python3 mail script already works standalone (tests out fine from CLI,
> on the server).
> 
> I'd rather not Flask out the whole website just to get the ability to run
> one .py script for contact mail.
> 
> I'd rather not add .php to the server just to do this contact form.
> 
> ** /cgi-bin to run the single python3 script is deprecated [Correct?].

Supported by existing protocols and running millions of
websites.

cheers

> It is a straight-up HTML5 form ... that USED to post to .aspx.
> 
> 
> 
> While I would prefer the answer to be python-ish,
> it does to Have to be python.
> 
> 
> Thank you!
> 
> 
> 

-- 
Joel Roth

"Welcome to the World Heat Bank, where we store your waste
energy and return it with interest."

Re: [OT] send all email from certain From: addresses into a spam

[Joel Roth]

On Wed, Jul 03, 2019, Greg Wooledge wrote:
> On Wed, Jul 03, 2019 at 06:30:32PM +0300, Reco wrote:
> > I'm merely curious why you have to write a new program for this.
> > I mean, there are maildrop and procmail for client-side and sieve for
> > the server-side already.
> 
> procmail might have worked, but it's more of a pain to learn procmail
> than it is to write my own filter.  I also get more flexibility this way.
> 
> The write-up of my approach is at
> <https://lists.debian.org/debian-user/2017/02/msg00100.html>.

Cool.

For those who are friendly with perl, Email::Filter or
Email::Filter::Rules provides a less cryptic alternative to
procmail. I use Email::Filter with Net::IMAP::Client so that
I run one filter process per mail check, rather than one
process per mail when triggering the filter from a .forward
file.

-- 

Joel Roth

"Welcome to the World Heat Bank, where we store your waste
energy and return it with interest"

Re: batch process jpegs

[Joel Roth]

mick crane wrote:
> I've got a bunch of jpgs from the camera where the bits I want are a bit
> underexposed. What I'd like is a GUI that will batch process them.
...
> Any suggestions for available GUI that will batch process jpgs ?

Hi Mick,

I would suggest using the command line utility 'convert'
that belongs to imagemagick. 

Batch processing is generally easier to manage at the
command line, even if it takes you an hour to learn 
how to do it.

cheers and HTH

convert -brightness-contrast  input-file output-file

docs from 
https://imagemagick.org/script/command-line-options.php#brightness-contrast

-brightness-contrast brightness
-brightness-contrast brightness{xcontrast}{%}

Adjust the brightness and/or contrast of the image.

Brightness and Contrast values apply changes to the input image. They are not 
absolute settings. A brightness or contrast value of zero means no change. The 
range of values is -100 to +100 on each. Positive values increase the 
brightness or contrast and negative values decrease the brightness or contrast. 
To control only contrast, set the brightness=0. To control only brightness, set 
contrast=0 or just leave it off.

You may also use -channel to control which channels to apply the brightness 
and/or contrast change. The default is to apply the same transformation to all 
channels.

Brightness and Contrast arguments are converted to offset and slope of a linear 
transform and applied using -function polynomial "slope,offset".

The slope varies from 0 at contrast=-100 to almost vertical at contrast=+100. 
For brightness=0 and contrast=-100, the result are totally midgray. For 
brightness=0 and contrast=+100, the result will approach but not quite reach a 
threshold at midgray; that is the linear transformation is a very steep 
vertical line at mid gray.

Negative slopes, i.e. negating the image, are not possible with this function. 
All achievable slopes are zero or positive.

The offset varies from -0.5 at brightness=-100 to 0 at brightness=0 to +0.5 at 
brightness=+100. Thus, when contrast=0 and brightness=100, the result is 
totally white. Similarly, when contrast=0 and brightness=-100, the result is 
totally black.

As the range of values for the arguments are -100 to +100, adding the '%' 
symbol is no different than leaving it off.



> 
> mick
> -- 
> Key ID4BFEBB31
> 

-- 
Joel Roth
  

Re: Setting a USB for multi usages

[Joel Wirāmu Pauling]

If using Uefi installs you only need to have a vfat formated first
partition with a folder called efi and the appropriate efi
binary/substructure. You can use the rest of the disk as you like.

BIOS installers on the same can be achieved with syslinux in addition to
uefi. However I find that just having uefi seems to go better these days
(at least if installing onto recent hardware). Due to varying levels of
vendor write as around CSM (compatibility BIOS/system mode) breaking dual
BIOS/eufi media installers.

-Joel

On Sat., 12 Jan. 2019, 09:48 MENGUAL Jean-Philippe  Hi,
>
> My purpose is having a USB stick splitted in 2 parts:
> 1. MBR + partitions: a Debian installer from an ISO
> 2. A blank partition to install data or whatver
>
> While I know to "burn" an iso on a key via dd, how can I do to have a
> clean installer but using key for other usages?
>
> Thanks very much
>
> regards
>
> --
> [image: Logo Hypra] JEAN-PHILIPPE MENGUAL
> DIRECTEUR TECHNIQUE ET QUALITÉ
> 102, rue des poissonniers, 75018, Paris
> Tel : +331 84 73 06 61 <+33184730661> Mob : +336 76 34 93 37
> <+33676349337>
> jpmeng...@hypra.fr
> www.hypra.fr
> [image: Facebook Hypra] <https://www.facebook.com/hyprasoftware/> [image:
> Twitter Hypra] <https://twitter.com/Hypra_> [image: Linkedin
> Jean-Philippe]
> <https://fr.linkedin.com/in/jean-philippe-mengual-800133135>
>
>

Re: [OT?] home partition vs. home directory

[Joel Roth]

On Fri, Nov 30, 2018 at 08:39:55PM -0600, David Wright wrote:
> On Fri 30 Nov 2018 at 11:23:57 (-1000), Joel Roth wrote:
> > On Fri, Nov 30, 2018 Cindy-Sue Causey wrote:
> > > Cliff's Notes Version Part I: Flaky USB connections are an important
> > > factor! An accidentally disconnected USB connection can cause data to
> > > become *unknowingly* redirected back to the original directory on the
> > > primary partition. That situation can then potentially lead to loss of
> > > e.g. downloaded data if a user is not always 100% aware of where data
> > > is actually residing at all times.
> > 
> > I've had problems with flakey connections with external USB
> > drives for years. These problems have occurred with various
> > drives and on two different laptops. The usual result is a
> > disk operation such as 'ls' fails with the message
> > "input/output error". This usually happens after
> > the drive has been idle for some hours. 
> 
> Hm, with poor connections, I find that it's usually the laptop
> moving about which disturbs the cable and causes the problem.
> After idle periods, I just get the message, when it wakes up:
>  usb 4-5: reset high-speed USB device number 4 using ehci-pci
> which I assume may be because the USB was in some sort of
> powersaving mode.
 
> > In my experience, the only case where writes went into
> > directory of the mount point (and not the intended
> > partition) was when the partition was not mounted. Once an
> > I/O error occurs, at least in my system, no further read or
> > write operations to the mount point will succeed. 
> 
> I've never had the mount points overwritten, perhaps because
> of how I set their permissions:
> drwx-- 2 root root 4096 Apr 11  2018 cdrom0/
> and I
> # touch -r cdrom0   all the mount points ...
> so it's easy to see when they're in use as the differing
> permission/ownership/timestamp sticks out.

Nice. I usually make a fstab entry for device
and do a mount -a whenever I plug in a new device.

> It also sounds as if you have   ,errors=remount-ro,
> set as an option in your /etc/fstab entries so that any
> error immediately write-protects the partition; though
> I didn't know that prevents reading as well. Does it?

I don't think my devices got remounted, as in that case
they would be at least read accessible. I do use remount-ro
option. 

One of these I/O errors ends the party completely.
Seems obvious that I should check kern.log for clues
next time I observe it.

I've never lost data due to these  errors, just
can be frustrating if the devices need unplugging and remounting.

^^


> Cheers,
> David.
> 

-- 
Joel Roth
  

Re: [OT?] home partition vs. home directory

[Joel Roth]

On Fri, Nov 30, 2018 Cindy-Sue Causey wrote:
> Cliff's Notes Version Part I: Flaky USB connections are an important
> factor! An accidentally disconnected USB connection can cause data to
> become *unknowingly* redirected back to the original directory on the
> primary partition. That situation can then potentially lead to loss of
> e.g. downloaded data if a user is not always 100% aware of where data
> is actually residing at all times.

Hi,

I've had problems with flakey connections with external USB
drives for years. These problems have occurred with various
drives and on two different laptops. The usual result is a
disk operation such as 'ls' fails with the message
"input/output error". This usually happens after
the drive has been idle for some hours. 

In my experience, the only case where writes went into
directory of the mount point (and not the intended
partition) was when the partition was not mounted. Once an
I/O error occurs, at least in my system, no further read or
write operations to the mount point will succeed. 

with regards

-- 
Joel Roth
  

Re: swrast alors !

[joel-dareau]

> Le 10/11/2018 à 18:48, ajh-valmer a écrit :
> > Mais avec quel driver choisi ?

On Saturday 10 November 2018 19:43:37 Txo wrote:
> Le propriétaire de debian via dkms.

Humm, plutôt le Libre de Debian ?

Le proprio, c'est Nvidia.

Re: 100Base-FX (SC) card PCI/PCIe

[Joel Wirāmu Pauling]

Am not  sure recommending 100FX ; the 1000 Base optics are around the
same price and likely a better bet; your application isn't looking at
throughput, but you do want some of the newer resiliency features on
1000 Base (EAM etc).


On Thu, 8 Nov 2018 at 00:06, Michelle Konzack
 wrote:
>
> Good day *,
>
> I am running on my Organic Farm and my Forest into problems with the
> lenght (100m) of Ethernet cables and want to use now for this case
> Fiber Multimode 1310nm cables.
>
> So what I have is following:
>
> 1) Server "Bunker" (19" 16U Rack, 2 Mini-ITX Server, 4G-Router,
>Astra-Internet Switch, currently 8-port D-Link GreenSwitch, UPS)
>
> 2) House (30m distance to second D-Link GreenSwitch)
> 3) Workspace (20m distance, will get maybe an additional D-Link 5-port
>GreenSwitch)
> 4) (Winter-)Greenhouse (60m distance for Siemens LOGO!)
> 5) Powerstation (20m distance, Victron Color Control GX, Siemens LOGO!,
>requires 2 cable or also an another D-Link 5-port GreenSwitch)
>
> So this can be done with 12m thick Ethernet Earthcables which I get
> for under 0,50€/m
>
> 6) Panorama IP cam with remote controll installed on my big Windmill
>mast (100Base-TX, 110m distance)
>
> Best option to connect 6)?
>
> 7) 1 IP cam (100Base-TX) with Interphone in 50m distance
>here I can also use an Ethernet cable and then using a Eternet/Fiber
>switch to connect
>
> and from 7)
>
> 8) 2 IP cams (100Base-TX) in 40m distance
> 9) 2 IP cams (100Base-TX) in 80m distance
> 10) 2 IP cams (100Base-TX) in 140m distance
>
> In 2020 we plan a Stall (160m distance) for our animals which should
> get also an IP cam and an IP based Interphone (SIP?)
>
> In 4-5 years we plan a second House which in a distance of arround
> 300-350m
>
> For the IP cams I can get very cheap the TP-Link MC100CM on which I
> can attach a small temperature controlled heating element (Estonia
> can be cold as -40°C on our farm). The LevelOne IEC-4001 would be
> nicer, but the price is GRMPF!
>
> Fiber SC cables I can get up to 400m preconfigured, so this is no
> problem.
>
>
> The question is now, how to connect all best together?
>
>
> The two Mini-ITX Servers have no free PCI/PCEe ports available,
> hence only Ethernet direct or the MX100CM is an option
>
> Now I need a Fiber Singel-Mode SC Switch and (maybe) Network Cards
> for my Workstations where two have PCI and PCIe slots.
>
>
> Which 100Base-FX/SC Switch and Card can you recommend?
>
>
> Note: Since I have NO AC-Power availlable, the Switches MUST HAVE
>   external power supplies (5-24V DC)
>
> I found on eBay the Microsens MS453111 very cheap (I can buy some
> spare too; however, the Microsens Website 
> does not more work since some time) but I would probably need in my
> "Bunker" a Switch which has 12-16 Eternet ports and 3-4 SC ports.
>
> Also the power consumption does matter because Of the solar- and
> windenergy systems
>
> My old 3Com 100Base-FX (PCI) cards have unfortunately only
> Singel-Mode hence I need new stuff for my 4 Workstations.
>
> Thanks in advance
>
> --
> Michelle KonzackMiila ITSystems @ TDnet
> GNU/Linux Developer 00372-54541400
>

Re: Previously Bootable: Stretch using Grub with GPT, LUKS, & BTRFS

[Joel Brunetti]

On Tue, 11 Sep 2018 at 17:22, Igor Cicimov  wrote:

>
>
> On Tue, 11 Sep 2018 11:45 am Joel Brunetti  wrote:
>
>> Hey Team,
>>
>> I'm having trouble booting a previously bootable system.
>> This system has been in use since very shortly before the Stretch release
>> and has always been Stretch.
>> I'm using Grub to boot a fully encrypted system. Each drive is
>> partitioned with GPT and encrypted using LUKS. The drives are then used
>> together with BTRFS.
>>
>> This system has worked with some minor boot problems (Which I thought
>> were fixed by adding the bios_grub flag to my partition and the pmbr_boot
>> flag to my disk) for at least a year.
>> Today I can not boot the system.  I suspect I've made it worse for trying
>> to repair it so I will jump to where I am now.
>>
>> When I boot I get on either device:
>> error: no such device: (UUID of my decrypted luks volume / btrfs pool)
>> error: unknown filesystem
>>
>
> Maybe check that the uuid hasn't changed somehow if mounting by uuid in
> /etc/fstab
>

The uuid in my fstab is correct. Unfortunately at boot it is not present
because grub fails to open the luks devices.

I managed to get my system to work by sending my btrfs sub volume to
another drive that was partitioned with MBR. There was no change to the
filesystem.
I chrooted in and installed grub and with a few other minor changes to
mount points my system is working again.

I have a poor understanding of GPT but I think GPT with grub and luks
encryption is a little unstable.

Thanks for the help.


>
>
>> I've chrooted onto the system using a usb key.
>> I can open my encrypted drives and mount the btrfs filesystem.
>> I suspected a bad kernel or grub update so I:
>> update initramfs -u -k all
>> update-grub
>> grub-install /dev/sda
>> grub-install /dev/sdb
>>
>> This gives the above errors when I boot.
>>
>> When I inspected /boot/grub/grub.cfg I noted it is missing "insmod
>> cryptodisk" and other encryption related modules. This is despite
>> /etc/default/grub containing "GRUB_ENABLE_CRYPTODISK=y".  I tried restoring
>> /boot/grub/grub.cfg from a snapshot that does include those modules and
>> then grub-install to both drives again but to no avail.
>>
>> I'm really at a loss and could really use some help in restoring my
>> system.
>>
>> Thanks,
>> Joel
>>
>>
>>
>>

Previously Bootable: Stretch using Grub with GPT, LUKS, & BTRFS

[Joel Brunetti]

Hey Team,

I'm having trouble booting a previously bootable system.
This system has been in use since very shortly before the Stretch release
and has always been Stretch.
I'm using Grub to boot a fully encrypted system. Each drive is partitioned
with GPT and encrypted using LUKS. The drives are then used together with
BTRFS.

This system has worked with some minor boot problems (Which I thought were
fixed by adding the bios_grub flag to my partition and the pmbr_boot flag
to my disk) for at least a year.
Today I can not boot the system.  I suspect I've made it worse for trying
to repair it so I will jump to where I am now.

When I boot I get on either device:
error: no such device: (UUID of my decrypted luks volume / btrfs pool)
error: unknown filesystem

I've chrooted onto the system using a usb key.
I can open my encrypted drives and mount the btrfs filesystem.
I suspected a bad kernel or grub update so I:
update initramfs -u -k all
update-grub
grub-install /dev/sda
grub-install /dev/sdb

This gives the above errors when I boot.

When I inspected /boot/grub/grub.cfg I noted it is missing "insmod
cryptodisk" and other encryption related modules. This is despite
/etc/default/grub containing "GRUB_ENABLE_CRYPTODISK=y".  I tried restoring
/boot/grub/grub.cfg from a snapshot that does include those modules and
then grub-install to both drives again but to no avail.

I'm really at a loss and could really use some help in restoring my system.

Thanks,
Joel

Re: hammerfall help request

[Joel Roth]

On Fri, Aug 17, 2018 at 12:12:22AM +, Glenn English wrote:
> On Thu, Aug 16, 2018 at 11:09 PM, Joel Roth  wrote:
> 
> > Unmute and adjust signal levels
> >
> > alsamixer -c 2
> 
> Comes up in curses with the RME selected and a single slider: DDS
> Samp. I raised that a couple steps with the 'up-arrow' button and
> exited.
> 
> > Test playback
> >
> > aplay -D hw:2,0 testfile.wav
> 
> There is no testfile.wav, so I gave it one of mine:
> 
> root@sbox:~# aplay -D hw:2,0
> /blackHole/cdp/BeethovenSix/source_dir/disk001/track02.cdda.wav
> Playing WAVE '/blackHole/cdp/BeethovenSix/source_dir/disk001/track02.cdda.wav'
> : Signed 16 bit Little Endian, Rate 44100 Hz, Stereo
> aplay: set_params:1334: Access type not available

Hmmm.

> Went into HDSPmixer and set it to preset 2 (like I did before to get
> the 'phones to work) and tried again. Still nothing. Files like this
> have worked before, in Audacity.

Looks like you may have to dig into the documentation. This
is a professional card, with some special capabilities 
and complexities.

I also suggest you search the Linux Audio Users mailing list
archives. It's an excellent resource for any Linux audio
issues.

> Not sure what aplay means by 'Access type not available' so I looked
> on the web. I still don't.
> 
> But 'aplay -L' says, about the RME:
> 
> default:CARD=DSP
> Hammerfall DSP, RME Hammerfall HDSP 9632
> Default Audio Device
> sysdefault:CARD=DSP
> Hammerfall DSP, RME Hammerfall HDSP 9632
> Default Audio Device
> dmix:CARD=DSP,DEV=0
> Hammerfall DSP, RME Hammerfall HDSP 9632
> Direct sample mixing device
> dsnoop:CARD=DSP,DEV=0
> Hammerfall DSP, RME Hammerfall HDSP 9632
> Direct sample snooping device
> hw:CARD=DSP,DEV=0
> Hammerfall DSP, RME Hammerfall HDSP 9632
> Direct hardware device without any conversions
> plughw:CARD=DSP,DEV=0
> Hammerfall DSP, RME Hammerfall HDSP 9632
> Hardware device with all software conversions
> 
> if that's any help.
 
The items are audio devices that can be used as 
a -D argument to aplay.

> -- 
> Glenn English
> 

-- 
Joel Roth
  

Re: hammerfall help request

[Joel Roth]

On Thu, Aug 16, 2018 at 06:50:10PM +, Glenn English wrote:
> On Thu, Aug 16, 2018 at 2:42 AM, Joel Roth  wrote:
> 
> > cat /proc/asound/cards
> 
> root@sbox:~# cat /proc/asound/cards
>  0 [Intel  ]: HDA-Intel - HDA Intel
>   HDA Intel at 0xfb6d8000 irq 51
>  1 [HDMI   ]: HDA-Intel - HDA ATI HDMI
>   HDA ATI HDMI at 0xfb8bc000 irq 52
>  2 [DSP]: H-DSP - Hammerfall DSP
>   RME Hammerfall HDSP 9632 at 0xfb7f, irq 16
> 
> That's pretty much what I've seen with all the other card-listers --
> two Intels and one RME. Next?

Unmute and adjust signal levels 

alsamixer -c 2

Test playback

aplay -D hw:2,0 testfile.wav

> -- 
> Glenn English
> 

-- 
Joel Roth
  

Re: hammerfall help request

[Joel Roth]

On Wed, Aug 15, 2018 at 02:30:01PM +, Glenn English wrote:
> Buster, RME Hammerfall, Supermicro box
> 
> I can't get my RME Hammerfall sound card to make noise in the
> headphones. It did a few weeks ago with pretty much the same hardware
> and pretty much the same OS.
> 
> I replaced the DDR3 1066 RAM with DDR3 1333, and I updated Buster.
> 
> I've tried to do what I did before to get the RME going, but either I
> can't remember what I did or something has changed so it doesn't work
> anymore (I doubt it).
> 
> lspci sees the card, as does alsamixer and the HDSP utilities in
> XFCE's multimedia menu (HDSPmixer and HDSPconf):
> 
> root@sbox:~# lspci | egrep -i hammerfall
> 02:01.0 Multimedia audio controller: Xilinx Corporation RME Hammerfall
> DSP (rev 97)
> 
> If I open an audio file in Audacity, I see the 'meters' bounce, but
> there's nothing in the 'phones.
> 
> I've looked at everything I can find on the web and done what they
> suggest. I've installed modules and listed them -- they all seem to be
> there:
> 
> root@sbox:~# cat /proc/modules | egrep hdsp
> snd_hdsp 69632 0 - Live 0xc0997000
> snd_rawmidi 40960 2 snd_seq_midi,snd_hdsp, Live 0xc08b8000
> snd_hwdep 20480 2 snd_hdsp,snd_hda_codec, Live 0xc073c000
> snd_pcm 118784 5
> snd_hdsp,snd_hda_codec_hdmi,snd_hda_intel,snd_hda_codec,snd_hda_core,
> Live 0xc0922000
> snd 94208 17 
> snd_seq,snd_hda_codec_realtek,snd_hdsp,snd_rawmidi,snd_seq_device,snd_hda_codec_generic,snd_hda_codec_hdmi,snd_hda_intel,snd_hda_codec,snd_hwdep,snd_pcm,snd_timer,
> Live 0xc0895000
> 
> And I've checked to make sure the 'phones are plugged in. Looks OK,
> but I think I remember hearing small clicks when I plugged them in
> before. All quiet now. Even if I'm remembering correctly, the lack of
> clicks could be due to having something set wrong.
> 
> I really think there's something I haven't done, but I can't figure
> out what. Anybody have any ideas?
> 
> -- 
> Glenn English

cat /proc/asound/cards 

-- 
Joel Roth
  

Re: USB2 or 3 WiFi dual band adapters

[Joel Wirāmu Pauling]

Huh? Intel cards are numerous and cheap - they come in PCIe / NGFF form
factors (like the easily available ath) - get a PCIe/USB to MiniPCIe
converter card for a few pennies off Aliexpress and you are in business.

On 14 August 2018 at 08:33, Michael Stone  wrote:

> On Wed, Aug 08, 2018 at 08:12:37AM -0700, tony mollica wrote:
>
>> I need to find a good, reliable WiFi adapter.  I have an Alfa AWUS036ACH
>> using
>> a RTL8812au chip
>> and there is support but it's unreliable.  Connects sometimes, mostly
>> not.  My
>> older adapters work
>> but they're slow but maybe that's the compromise I need to resolve.
>>
>> What's being used reliably?
>>
>
> Intel, but as far as I know you can only buy one with a new laptop.
> Your other options are the rtl* stuff which tends to be flaky, or the ath*
> stuff which is impossible to find. It's not a good time for open source
> wifi.
>
> Mike Stone
>
>

Re: USB2 or 3 WiFi dual band adapters

[Joel Wirāmu Pauling]

Basically find one that uses the ath9k Chipset. They are easily the best
supported Wifi Interface.

If you need Wireless AC then ath10k based products are useable too.

The Intel ranges are OK as clients, but are not really very Opensource.
Ath9k has the best Fully Opensource impementation out of any of the
Wireless cards.

https://wikidevi.com/wiki/Ath9k

On 9 August 2018 at 03:12, tony mollica  wrote:

> Hello.
>
> I need to find a good, reliable WiFi adapter.  I have an Alfa AWUS036ACH
> using a RTL8812au chip
> and there is support but it's unreliable.  Connects sometimes, mostly
> not.  My older adapters work
> but they're slow but maybe that's the compromise I need to resolve.
>
> What's being used reliably?
>
> Thanks,
>
> Tony
>

x2go reverse proxy

[BERTRAND Joel]

Bonjour à tous,

	Je cherche à prendre la main sur des machines distantes à l'autre bout 
du monde. Pour ce faire, je pense à quelque chose qui pourrait être un 
reverse proxy pour x2go (ou tout autre chose du même genre). Pour 
l'instant, j'utilise x11vnc, mais c'est très plantogène. Sur un réseau 
local, ça fonctionne sans souci, en passant sur internet, c'est une 
autre affaire. Dès qu'un paquet se perd, la session plante.


	J'ai en entrée un VPN entre un serveur NetBSD (serveur local) et le 
serveur distant (Linux/Debian). Derrière le serveur distant, j'ai une 
dizaine de postes tournant avec Windows 7 (32 ou 64 bits) ou Linux 
Debian (pour les nouvelles machines, celles-ci faisant tourner 
virtualbox pour utiliser des progiciels qui ne tournent 
qu'officiellement avec XP et W7 en bidouillant).


	Je peux installer un serveur x2go sur chaque client. Mais je trouve 
plus propre d'installer un genre de proxy sur le serveur distant. Est-ce 
possible ?


Bien cordialement,

JKB

[résolu] Installation sur Asus Prime B250M-K was Re: Installation sur CPU kaby lake

[BERTRAND Joel]

BERTRAND Joel a écrit :

BERTRAND Joel a écrit :

 Bonjour à tous,

 J'essaie vainement d'installer Debian sur deux machines pour un 
client. Les cartes-mères sont des Asus Prime 250M-K avec un CPU 
Pentium G4560.


 J'ai essayé de booter sur le réseau (mon installation tftp 
fonctionne puisque toutes les machines de mon réseau bootent ainsi). 
Le client récupère le pxelinux.0 puis d'autres fichiers que je vois 
passer dans les logs et... je n'obtiens qu'un écran noir sur le PC que 
j'essaie d'installer.


 J'ai donc téléchargé une image netinst de stable. Je ne sais pas 
si ça boote, en fonction des options du bios, soit le système ouvre 
l'UEFI, soit l'écran reste noir.


 J'ai donc pensé à un problème de firmware et j'ai téléchargé 
l'image avec les firmware (de stable). Même problème, que la 
configuration soit legacy ou UEFI. J'ai désactivé tout ce qui pouvait 
poser problème (secure boot et autres joyeusetés).


 Je ne sais plus par où chercher.

 Une idée ?

 Bien cordialement,

 JKB



Quelques précisions :
- Le secure boot est désactivé.
- la clef USB est bootable sur mon PC portable (pas UEFI). J'ai essayé 
toutes les options côté machines à installer sans succès.

- je n'arrive pas à avoir le menu de grub.
- en désespoir de cause, j'ai mis à jour le bios.


	Je me réponds à moi-même. J'ai enfin réussi à faire booter la chose. 
J'ai commencé à 8h00 ce matin, c'est donc mon record. ;-)


Pour ce faire :
- mettre le bios à jour pour la dernière version (1205) ;
- désactiver le secure boot en virant la clef PK (et que celle-là, 
sinon, le machin se réactive dans le dos de l'utilisateur) ;
- désactiver l'UEFI (pas capable de booter sur le réseau sinon, plus 
exactement avec l'UEFI, la machine boote à moitié, s'arrêtant après le 
chargement du PXE) en ne gardant que les options "legacy" ;
- utiliser le netboot de buster, celui de stretch a un noyau un peu 
ancien (il faut visiblement un 4.10 pour supporter correctement la 
carte-mère).


Moyennant cela, on finit par arriver à démarrer la chose par le réseau.

	Je n'ai pas réussi à démarrer la machine à partir d'un port USB (USB2 
ou 3), le comportement étant quelque peu erratique. Naturellement, la 
clef utilisée fonctionne parfaitement sur d'autres machines.


Cordialement,

JKB

Re: Installation sur CPU kaby lake

[BERTRAND Joel]

BERTRAND Joel a écrit :

 Bonjour à tous,

 J'essaie vainement d'installer Debian sur deux machines pour un 
client. Les cartes-mères sont des Asus Prime 250M-K avec un CPU Pentium 
G4560.


 J'ai essayé de booter sur le réseau (mon installation tftp 
fonctionne puisque toutes les machines de mon réseau bootent ainsi). Le 
client récupère le pxelinux.0 puis d'autres fichiers que je vois passer 
dans les logs et... je n'obtiens qu'un écran noir sur le PC que j'essaie 
d'installer.


 J'ai donc téléchargé une image netinst de stable. Je ne sais pas si 
ça boote, en fonction des options du bios, soit le système ouvre l'UEFI, 
soit l'écran reste noir.


 J'ai donc pensé à un problème de firmware et j'ai téléchargé 
l'image avec les firmware (de stable). Même problème, que la 
configuration soit legacy ou UEFI. J'ai désactivé tout ce qui pouvait 
poser problème (secure boot et autres joyeusetés).


 Je ne sais plus par où chercher.

 Une idée ?

 Bien cordialement,

 JKB



Quelques précisions :
- Le secure boot est désactivé.
- la clef USB est bootable sur mon PC portable (pas UEFI). J'ai essayé 
toutes les options côté machines à installer sans succès.

- je n'arrive pas à avoir le menu de grub.
- en désespoir de cause, j'ai mis à jour le bios.

JKB

--
Dr. BERTRAND Joël
SYSTELLA S.A.S., 7, la Sudrie, 19130 VIGNOLS FRANCE
Siège: +33 (0)973870210, Paris: +33 (0)973870201, GSM: +33 (0)616018060
http://www.systella.fr

Installation sur CPU kaby lake

[BERTRAND Joel]

Bonjour à tous,

	J'essaie vainement d'installer Debian sur deux machines pour un client. 
Les cartes-mères sont des Asus Prime 250M-K avec un CPU Pentium G4560.


	J'ai essayé de booter sur le réseau (mon installation tftp fonctionne 
puisque toutes les machines de mon réseau bootent ainsi). Le client 
récupère le pxelinux.0 puis d'autres fichiers que je vois passer dans 
les logs et... je n'obtiens qu'un écran noir sur le PC que j'essaie 
d'installer.


	J'ai donc téléchargé une image netinst de stable. Je ne sais pas si ça 
boote, en fonction des options du bios, soit le système ouvre l'UEFI, 
soit l'écran reste noir.


	J'ai donc pensé à un problème de firmware et j'ai téléchargé l'image 
avec les firmware (de stable). Même problème, que la configuration soit 
legacy ou UEFI. J'ai désactivé tout ce qui pouvait poser problème 
(secure boot et autres joyeusetés).


Je ne sais plus par où chercher.

Une idée ?

Bien cordialement,

JKB

Re: Get the external IP address from a Linux box

[Joel Rees]

(Erk. Sorry, Joe.)

On Thu, May 24, 2018 at 6:29 PM, Joe  wrote:
> On Thu, 24 May 2018 08:13:54 +0100
> André Rodier  wrote:
>
>> On Thu, 2018-05-24 at 09:07 +0200, Alberto Luaces wrote:
>> > Joe writes:
>> >
>> > > On the assumption that you are using a router of some kind, your
>> > > public
>> > > IP address will be that of the router WAN port (cable, ADSL, etc.)
>> > > and
>> > > there will be a method of determining that by connecting to the
>> > > router
>> > > as an administrator. That method will depend entirely on the
>> > > router.
>> >
>> > If the router supports upnp and it is activated, you can check the
>> > external IP in an device-independent way with
>> >
>> > upnpc -l | grep ExternalIPAddress
>> >
>>
>> Thank you, finally an answer that make sense and is not pedantic.

Two people have already tried to point out that UPNP is vulnerable by design.

If you have any interest in your local security, your router to the
outside should simply not respond to UPNP at all.

Block/ignore UPNP at every interface, internal and external, on your
external router, at bare minimum.

This is not pedantry, this is trying to save you from being attacked
from your inside.

> How is it possible to avoid being pedantic? You told us nothing about
> your Internet connection, or Debian version, so we had to guess at what
> information you actually wanted and which device to ask.
>
>> I tried this, but it is not 100% reliable. For instance, with the
>> firewall / router I use, upnp id not activated. I suppose I will have
>> to write a custom python script.
>
> So presumably it isn't your computer's external address that you want,
> but that of your router. I was a bit surprised to see upnp mentioned, I
> thought it was only game-players who were willing to run that, and
> Debian would not be their OS of choice.
>
> From (not recent) experience of talking to routers, you may have telnet
> or ssh available, otherwise it's an http admin login, followed by one or
> two router-specific commands. You might be lucky, and the default
> router status page without login may contain the WAN address.
>
> A couple of lines of bash should do it: use curl, and you'll probably
> have to provide the admin password, so the script should probably be
> stored in /root. My routers in years gone by used to need an occasional
> reboot, so I had a script running every ten minutes to check multiple
> websites for connectivity, and if none were found, to issue a reboot
> command.

What Joe says here.

I had ten or twenty lines of moderately careful code in a two
hundred-line perl script I used to update my dyndns.com domain name
back before dyn.com decided they had to kick all the freeloaders like
me off.

Resolution to *some* domain name really ought to be part of an ISP's
basic package, but the Internet got taken over by the poachers.


--
Joel Rees

http://reiisi.blogspot.jp/p/novels-i-am-writing.html

Re: utf

[Joel Roth]

On Wed, Apr 04, 2018 at 02:20:17PM -0400, rhkra...@gmail.com wrote:
> On Wednesday, April 04, 2018 12:58:57 PM deloptes wrote:
> > And regarding the mbox thing, well mbox was depreciated for many reasons. I
> > guess if it was that good it wouldn't be depreciated.

> Oh, I wasn't aware that mbox was deprecated--can you shed more light on that. 
>  
> AFAIK, it is not defined in an RFC and is used by quite a few email programs.
 
Not exactly deprecated, but it's considered a less reliable storage format, 
because of potential problems. 

https://en.wikipedia.org/wiki/Mbox

I converted to Maildir for better compatibility with the mu
indexing programs (package maildir-utils). 

cheers,
 

-- 
Joel Roth
  

Re: Simple spreadsheet program.

[Joel Roth]

On Tue, Mar 13, 2018 at 02:13:33PM +1100, terryc wrote:
> What is a simple spreadsheet program that can be installed under
> Stretch. I need to do some work quickly

Here's Instacalc, a free-to-use DOS program that I 
once used heavily.

http://www.bttr-software.de/freesoft/dbase.htm#instacalc
 
Note that you'll need to run this under some type of 
DOS emulator. 

> hint, if your answer is LibreOffice or similar read the question again.
> I'm frustrated that the last few time I wanted to do a simple
> spreadsheet layout, it was easier and faster to craft a LaTex document
> then try and unfathom LibreOffice methods.

I'm a dinosaur, too, but the bar to using LibreOffice calc
seems pretty low, basically one 5 minute youtube tutorial.

cheers
 

-- 
Joel Roth
  

Re: Multichannel audio listening

[Joel Roth]

Rodolfo Medina wrote:
>  I want to buy one of those multichannel
> soundcards...  Do you think this one could be all right...?
> 
> 
> https://www.strumentimusicali.net/product_info.php/products_id/51790/behringer-umc404hd.html?gclid=Cj0KCQiAuP7UBRDiARIsAFpxiRKgsptPW1qxkEw793ahs684ltlhyh5dcIgzLJXtDh39CZA8IEX3qgIaAnquEALw_wcB
> 
> Rodolfo

I looked around a bit, and the UMC404HD seems to be 'class
compliant' which means that the linux USB audio drivers can
access the most important functions. Behringer is definitely
on the low-cost end of things, but a lot of their hardware
seems to be of decent quality. 

A useful feature of this card is hardware monitoring, so
that you can listen to the audio during recording without a
time lag. 

Now that you've described your application (recording piano,
possibly with vocals) I think -- unless you're especially
comfortable and patient with low-level commands such as
ecasound provides -- you may like to run some kind of
multitrack recorder or DAW application. This will let you
adjust volume levels and add various plugins to the
different tracks. It's also common to duplicate a mono
signal to stereo and adjust the position right or left in a
stereo mix (e.g. panning).  

Someone already mentioned Audacity, which is quite easy to
use.

More sophisticated software lets you apply effects in
realtime, so it's easier to diddle with parameters. 

For professional quality with all possible features,
there is Ardour. Another very well developed application is
Qtractor. 

I'll also shamelessly mention a lightweight DAW called Nama
that provides the most important functions for recording and
mixing.  It's an application layer driven by text commands,
hotkeys and/or a simple GUI that runs Ecasound to provide
the signal processing. The debianized version is slightly
out of date, but you can at least get an idea of what it
looks like.  One of several unusual features is a preset
system that lets you set plugin defaults, re-use chains of
plugins, and create templates for groups of tracks or entire
projects. I'm the biased author ;-) 

For hardware or software questions you cannot answer by
reading the docs or doing a web search, I'll refer you to
the many experts on the Linux Audio Users mailing list
(LAU).

cheers,

-- 
Joel Roth
  

Re: Multichannel audio listening (was: Live recording)

[Joel Roth]

On Mon, Mar 05, 2018 at 10:39:19AM +0100, Rodolfo Medina wrote:
> Hi all.
> 
> After learning, some months ago, thanks to listers' help, how to live record
> into a multi channel audio file, I was wondering about the reverse problem: 
> now
> I have my multi channel audio file, e.g. composed by three different channels.
> Is it possibile (I guess it is), and how?, to send each of the three outputs
> into a different loud speaker and so listen to the song...?
> 

Hi Rodolfo,

Ecasound is pretty convenient for routing audio.

If you connect three powered speakers to the first three
channels of a (sufficiently capable) soundcard and you're
using ALSA (the default low-level linux audio API), it's
pretty simple:

For example,

ecasound -i:3ch.wav -f:16,3,44100 -o:alsa,default
 
If you need to fool around with the routing, you can do
almost anything. For example, assuming  you have an
8-channel soundcard and want to route ch1 to ch5, ch2 to ch6
and ch3 to ch7:

ecasound -i:3ch.wav \
-f:16,8,44100  \
-chmove:1,5 \
-chmove:2,6 \
-chmove:3,7 \
-o:alsa,default

Maybe you need to add separate effects (e.g. volume control) to each channel:

ecasound \
-a:in -i:3ch.wav -o:loop,3ch \
-a:ch1,ch2,ch3 -i:loop,3ch \
-a:ch1 -chmove:1,5 -ea:80 \
-a:ch2 -chmove:2,6 -ea:70 \
-a:ch3 -chmove:3,7 -ea:60 \
-a:ch1,ch2,ch3 -f:16,8,44100 -o:alsa,default

There's a lot more you can do with ecasound, and there are
various front ends that provide a higher level of
abstraction.

Hope this helps,

> I hope I was clear enough...

> Thanks for any help.
> 
> Rodolfo
> 

-- 
Joel Roth
  

Re: DTrace GPLed?

[Joel Wirāmu Pauling]

eBPF makes dtrace less interesting.

On 21 February 2018 at 08:27, Weaver  wrote:

> Of interest to some?
>
> https://www.theregister.co.uk/2018/02/19/oracle_open_
> sources_dtrace_changes_licence_to_gpl/
>
> Cheers!
>
> --
> `The difference between friendship and love is how much you can hurt
> each other’.
> ― Ashleigh Brilliant
>
> Registered Linux User: 554515
>
>

Re: How to create a PDF-Printer from the command line

[Joel Wirāmu Pauling]

Works fine for txt, although as it rasterizes things it's not going to be
optimized for size.

On 18 January 2018 at 10:33, Ben Caradoc-Davies <b...@transient.nz> wrote:

> On 18/01/18 10:15, Joel Wirāmu Pauling wrote:
>
>> **cough** $convert
>> imagemagick
>> $convert somefile.whatever somefile.pdf
>>
>
> +1 for ImageMagick convert to generate PDFs from scanned pages (images). I
> found that this works best with -page and -density specified. However, I
> have not tried using it for text, which I think was the problem facing the
> original poster.
>
> Kind regards,
>
> --
> Ben Caradoc-Davies <b...@transient.nz>
> Director
> Transient Software Limited <https://transient.nz/>
> New Zealand
>
>

Re: How to create a PDF-Printer from the command line

[Joel Wirāmu Pauling]

**cough** $convert

imagemagick

$convert somefile.whatever somefile.pdf

---

On 18 January 2018 at 09:04,  wrote:

> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
>
> On Wed, Jan 17, 2018 at 06:21:31PM +, Brian wrote:
> > On Wed 10 Jan 2018 at 21:01:13 +0100, to...@tuxteam.de wrote:
>
> [...]
>
> > > Thanks for actually trying out. I stand corrected...
> >
> > A gracious response. However, my data were in the context of using a2ps
> > to go from text to PS. Your "hefty PDFs" would be entirely correct if
> > paps had been used for the conversion. The result is a 156551 sized file
> > for me. gs2pdf comes up with a whopping 11540827 and takes 18 s to do so.
>
> Heh. I've been called all sort of names, but gracious... :-)
>
> > That was in 2015, Debian's paps does not relect the existence of a
> > 7.0 version. I wonder why?
>
> Yes, that might be the root of my dim memories.
>
> Thanks for your thorough investigation
>
> cheers
> - -- tomás
> -BEGIN PGP SIGNATURE-
> Version: GnuPG v1.4.12 (GNU/Linux)
>
> iEYEARECAAYFAlpfrDMACgkQBcgs9XrR2kYFGACeL5J7MFpQSDa3F96kHNjq6D7H
> DFMAn0w5jLhnXV492+EJQeORz3LhQqoX
> =t1V6
> -END PGP SIGNATURE-
>
>

Re: Experiences with BTRFS -- is it mature enough for enterprise use?

[Joel Wirāmu Pauling]

The reason Redhat dropped btrfs support is because it currently has no
native cryptographic function. And from the various threads I've read on
the topic there is no easy answer to the problem.

On 1 January 2018 at 06:44, Sven Hartge  wrote:

> David Christensen  wrote:
> > On 12/30/17 14:38, Matthew Crews wrote:
>
> >> The main issue I see with using BTRFS with MDADM is that you lose the
> >> benefit of bit-rot repair. MDADM can't correct bit rot, but
> >> BTRFS-Raid (and ZFS raid arrays) can, but only with native raid
> >> configurations.
>
> > AFAIK:
>
> > 1.  mdadm RAID1 can fix bit rot, so long as one drive has a good block
> > to fix the others.
>
> Yes, but it can't fix silent bit-rot, where incorrect bytes are read
> from the drive without the drive noticing. In that case the Kernel has
> no way of knowing which bytes are the correct ones, you need some sort
> of checksum for that.
>
> Grüße,
> Sven.
>
> --
> Sigmentation fault. Core dumped.
>
>

Re: Pidgin et webcam

[BERTRAND Joel]

Michel Memeteau - EKIMIA a écrit :

Salut , avec protocol ? Xmpp ?


	Oui, j'ai oublié de préciser (et prosody comme serveur, ejabber me 
faisait des misères).


JKB

--
BERTRAND Joël
SYSTELLA S.A.S.
http://www.systella.fr

Pidgin et webcam

[BERTRAND Joel]

Bonjour à tous,

	J'essaie sans succès d'installer une webcam sous testing et pidgin. Sur 
deux machines différentes, j'ai exactement le même souci :


(10:54:51) mediamanager: gst pipeline error: Internal data stream error.
(10:54:51) mediamanager: Debug details: gstbasesrc.c(2939): 
gst_base_src_loop (): 
/GstPipeline:pipeline0/GstBin:pidgindefaultvideosrc/GstBin:pidgindefaultvideosrc/GstAutoVideoSrc:autovideosrc0/GstV4l2Src:autovideosrc0-actual-src-v4l:

streaming stopped, reason not-negotiated (-4)
(10:54:51) backend-fs2: gst error Internal data stream error.
debugging: gstbasesrc.c(2939): gst_base_src_loop (): 
/GstPipeline:pipeline0/GstBin:pidgindefaultvideosrc/GstBin:pidgindefaultvideosrc/GstAutoVideoSrc:autovideosrc0/GstV4l2Src:autovideosrc0-actual-src-v4l:

streaming stopped, reason not-negotiated (-4)
(10:54:51) media: Problème avec votre webcam
(10:54:51) media: Erreur de conférence

	Naturellement, j'ai vérifié que les deux flux sont accessibles (avec 
cheese ou gst) et que les droits sont les bons. Malgré cela, je n'arrive 
à rien et j'avoue que la lecture de la documentation disponible ne m'a 
pas vraiment fait avancer. Je ne comprends pas pourquoi je me prends un 
"not-negotiated". Pour être tout à fait franc, je me demande si gst est 
bien appelé dans l'histoire mais je n'ai pas trouvé e moyen de le savoir.


Une idée ?

Bien cordialement,

JKB

Re: [HS] Choix registrar, conseils

[BERTRAND Joel]

andre_deb...@numericable.fr a écrit :

On Monday 04 December 2017 10:19:58 fab wrote:
...

Enfin, concernant OVH et c'est mon petit coup de gueule,
cette façon (très récente) d'obliger les clients à lier un compte ou une
carte bancaire à leurs services a fini de me convaincre que ce n'était
plus un hébergeur intérresant.


Je confirme,

OVH a besoin d'investir et cherche de l'argent chez le client.
- Politique d'inscription à la financière,
- Augmentation des factures sur motifs pipeau,
- Impossibilité d'avoir le type de contrat, CDD ou CDI, et réponses
   biaisantes,
- Panne grave (effacement données) sans la moindre excuse,
- Graves pannes électrique récentes sur 2 sites (Strasbourg et Roubaix)...


	Guère mieux que Gandi. Je me souviens avoir débarqué au siège social de 
Gandi pour taper du point sur la table pour un client. Il fallait 
simplement que quelqu'un se penche sur le problème.


Bref, en dessous d'un certain prix, on n'en a que pour son argent.

	J'avoue ne pas les avoir tous testés, mais j'ai eu des déboires avec 
Gandi, OVH, 1and1, Ikoola (instabilités électriques, pertes de données, 
migrations impossibles, prérequis étranges sur des mutualisés...). Rien 
que cette année, OVH m'a planté trois fois des serveurs (IPBX + services 
divers). Pour l'instant, pas de problème avec Amen, Equinix ou Nérim 
(là, j'ai une baie vide louée où je colle mon matériel).


	Je n'ai jamais trouvé un vrai service en dessous de 50 ou 60 €/mois. En 
dessous, ce sont des prix d'appel. Tout peut parfaitement bien se 
passer, mais en cas de problème, on n'a que ses yeux pour pleurer. De 
toute façon, il est possible de faire un calcul simple : coût mensuel du 
serveur et de la maintenance sur cinq ans plus électricité et accès 
internet, plus climatisation et techniciens de maintenance. Ça fait un 
coût incompressible pour un hébergeur sérieux.


Bien cordialement,

JKB

Re: tesing et teamviewer

[BERTRAND Joel]

Bernard Schoenacker a écrit :



- Mail original -

De: "Gaëtan Perrier" 
À: "Liste Debian" 
Envoyé: Mercredi 15 Novembre 2017 22:02:56
Objet: tesing et teamviewer

Bonjour,

Arrivez-vous à faire fonctionner teamviewer 12 sur testing
actuellement ?
Quand je le lance j'ai ça:

$ teamviewer

Init...
CheckCPU: SSE2 support: yes
XRandRWait: No value set. Using default.
XRandRWait: Started by user.
Checking setup...
Launching TeamViewer ...
Launching TeamViewer GUI ...


puis plus rien. Aucun affichage, aucune erreur.

Est-ce de même pour vous ?

Gaëtan


bonjour,

j'ai le même problème sur 3 ordis, et j'ai envoyé les logs
à teamviewer (j'attends une réponse) 

comme solution temporaire, je propose de passer par le
navigateur chromium : Chrome Remote Desktop

https://chrome.google.com/webstore/detail/chrome-remote-desktop/gbchcmhmhahfdphkhkmpfmihenigjmpp

https://alternativeto.net/software/teamviewer/

désolé de ne pouvoir en dire plus, mais je n'ai pas eu les moyens de
faire des essais

slt
bernard




Bonsoir,

Ni 11 ni 10 d'ailleurs...

Cordialement,

JKB

Re: No Sound - Puzzle

[Joel Roth]

A lot of folks (myself included) manage just fine without
pulseaudio.  ALSA is already complicated and capable enough for them.
Some sophisticated users route pulseaudio through JACK, so
that the pulseaudio daemon doesn't grab exclusive use of the ALSA device. 

Good luck!

On Wed, Nov 15, 2017 at 06:21:07PM -0500, Thomas George wrote:
> From a terminal run alsamixer, f6, select Xonar DSX
> 
> On another terminal cd to a directory of ogg files, start a playback of a
> file.
> 
> Oh the desktop start pulse audio volume control. Monitor shows strong active
> playback levels
> 
> Switch to terminal running alsamixer, mute master front which mutes all
> output channels.
> 
> Return to desktop pulse audio volume control, monitor continues to show
> strong active playback levels even though alsamixer thinks all output
> channels muted!
> 
> The music goes down and around and comes out nowhere!
> 
> 
> On 11/15/2017 03:35 PM, Thomas George wrote:
> >I just installed an Asus Xonar DSX pcie card. Alsamixer recognizes it and
> >PulseAudio Volume Control show a strong signal level during playback but
> >no sound from the speakers. The system works with the onboard Nvidia
> >sound, no problems.
> >
> >I assume the problem is a missing linux driver for the Xonar but I have
> >seen claims that Xonar works out of the box on linux systems.
> >
> >Does anyone have this working?
> >
> >I am using Debian Stretch.
> >
> >
> >
> 

-- 
Joel Roth
  

Re: Rsync

[Joel Roth]

Brian wrote:
> I would take "base system" to mean what the installer means by it. In
> which case, it is should not be associated with "standard system
> utililites". They are different sets of packages.

Generally it is up to the author of a liveCD distribution to
include packages relevant for utility use and I think it's
fair that the net installer focus on minimizing its own
requirements. Equally, rsync is small, useful, and
comprehensible, IMO a good target for teaching the command
line literacy. 

cheers

-- 
Joel Roth
  

Re: Codec Xvid

[BERTRAND Joel]

Dominique Dumont a écrit :

On Thursday, 5 October 2017 13:53:27 CEST BERTRAND Joël wrote:

ffmpeg -i te donne quelles infos?


Pour vérifier si ffmpeg arrive à décoder le flux vidéo, est-ce que tu peux
essayer ffplay et nous montrer la sortie sur le terminal ?



Je ne peux pas faire cela avant lundi.

Cordialement,

JKB

Re: [POSTFIX] On est trolldi, j'en ai une bien bonne :

[BERTRAND Joel]

Ph. Gras a écrit :

Hello les gens !

Saviez-vous ce que signifie SMTP ?

Simple Mail Transfert Protocol

MDR -D

Ph. Gras


	Ehoh ! Il y a longtemps qu'on est passé à ESMTP.Et sais-tu ce que 
signifie le E ? ;-)


JKB

Re: Performance RAID instable

[BERTRAND Joel]

Pascal Hambourg a écrit :

Le 05/10/2017 à 13:34, BERTRAND Joël a écrit :

 Je pense (mais je n'ai pas regardé, il y a longtemps que je ne
bidouille plus le noyau Linux) qu'avec moins de 4 Go de mémoire, même
avec PAE, le système ne va pas essayer de mapper la mémoire sur plus
de 32 bits d'adresses, donc n'essayera pas d'étendre l'adressage au-delà.


D'après ce que j'ai compris en lisant

l'activation de PAE modifie la structure des tables de pages (avec
notamment un niveau supplémentaire), quelle que soit la quantité de
mémoire physique à adresser. Je ne pense pas que le fait que la totalité
de la mémoire physique soit adressable avec 32 bits y change quelque chose.


	Personnellement, je ne fais pas d'hypothèse sur la complexité de la 
grouille rajoutée par PAE (j'ai des vieux souvenirs des problèmes de 
gestion de la mémoire sur sparc32 entre le 2.0 et le 2.2...). J'ai vu 
des trucs tellement ma fichus dans les noyaux Linux... D'autant que les 
noyaux i686 avec plus de 4Go de mémoire doivent être aujourd'hui 
marginaux. Je ne suis pas sûr qu'il y ait encore un réel effort là-dessus.


Cordialement,

JKB

Re: Performance RAID instable

[BERTRAND Joel]

Frédéric MASSOT a écrit :

Le 05/10/2017 à 11:12, BERTRAND Joël a écrit :

Seb a écrit :


Bonjour Pascal,



Je n'ai pas le courage de relire le fil pour voir si tu l'as déjà fait
: utiliser le noyau 3.16 de Debian 8 sur Debian 9 ?


Ah oui, c'est une bonne idée.

Si j'appelle sur une machine en Debian 9
apt-cache search linux-image
je ne vois pas de noyau 3.16 .

Dans la FAQ Debian, chapitre "Debian and the kernel", je ne vois pas de
solution packagée indiquée; d'un autre côté, elle ne mentionne pas non
plus les backports pour avoir, au contraire, un noyau plus récent.

Y a-t-il une méthode standard Debian qui me permettrait d'installer sans
trop souffrir le noyau 3.16 sur la Debian 9 ? (Un pointeur vers la bonne
page web m'irait amplement.)


Avec un init systemV, c'est assez trivial. Avec la grouille
systemd/udev, c'est assez casse-gueule. La seule fois où j'ai dû faire
ça, cela s'est soldé par un échec.


À cause du bug (865866) de LibreOffice Writer j'utilise un noyau 3.16
i386 sur une Buster avec systemd et udev sans problème. Il y a aucun
problème à changer de noyau.


Bonjour,

	Ça se fait, mais de là à dire sans problème... J'ai dû faire un 
rétrofit de matériel qui avait été installé avec un 4.x (j'ai oublié le 
numéro mineur). Le noyau 3.16 s'installe, boote, mais se baugeait lors 
de l'appel à udev et systemd. Il y a une dépendance forte entre udev, la 
version de la bouse systemd et celle du noyau. Ça peut se passer bien, 
avec juste une ligne de warning ou rien du tout, ça peut aussi mettre le 
système en vrac. On ne peut plus aujourd'hui garder des versions du 
noyau différentes pour se sortir d'un mauvais pas de manière fiable.


	Lorsque tu prends une Debian avec un 4.11 ou 4.12 et que tu veux 
remettre un 3.16, il convient d'installer toutes les dépendances du 
3.16. Et ça peut faire beaucoup de monde puisque de plus en plus 
d'outils tiers dépendent de la saloperie systemd.


	Dans le sens contraire, tu peux _aussi_ avoir des problèmes. Pas plus 
tard qu'hier, j'ai fait un dist-upgrade sur un serveur de test. 
Changement de noyau et, forcément, changement de udev et systemd. 
Impossible à redémarrer sans passer au bouton ! shutdown -r ne faisait 
rien (sauf bouffer 100% d'un CPU durant plus d'une heure). Le noyau 
initial était un 4.11, le nouveau un 4.12 avec la grouille qui venait 
avec lui. Là, ça allait, j'étais à côté de la machine. Mais je sens que 
je vais me déplacer pour mes machines hébergées qui n'ont pas de bandeau 
de prise commandé à distance.


Cordialement,

JKB

Re: Recommended editor for novice programmers?

[Joel Roth]

I'm dropping in late to say that running 'vimtutor' in a 
terminal is an easy way to interactively get to know how vim
works.

-- 
Joel Roth
  

Re: no /etc/inittab

[Joel Rees]

2017/08/15 1:05 "Brian" <a...@cityscape.co.uk>:
>
> On Mon 14 Aug 2017 at 13:22:45 +0200, Nicolas George wrote:
>
> > Le septidi 27 thermidor, an CCXXV, Pierre Frenkiel a écrit :
> > > I just wanted to know why the Debian wiki is not updated, 2 months
after the Stretch release.
> >
> > Jessie also used systemd, so that is more two years than two months.
> >
> > The answer to that question is simple: it is a wiki, it has not been
> > updated because you did not update it. Me neither. Please proceed.
>
> That looked to be a five minute job. Replace the "Overview" section
> with the single line "The system initialization process is handled
> by systemd" and delete all the links except the one to init(1).
>
> That's until you get to thinking what the purpose of the page is and
> look at where it is linked from and what has to be done to make all
> the parts form a coherent whole.

I wonder if you will now begin to recognize why the forced universal
upgrade to systemd was a thoroughly ill-conceived bit of social
engineering.

--
Joel Rees

http://reiisi.blogspot.com

Re: Yes you have standing to sue GRSecurity.

[Joel Rees]

On Mon, Jul 31, 2017 at 5:01 PM,  <to...@tuxteam.de> wrote:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
>
> On Mon, Jul 31, 2017 at 09:23:06AM +0200, deloptes wrote:
>
> [...]
>
>> Recently I found out that when ideology stands higher than pragmatics the
>> whole thing affected dies ... numerous examples like communism or democracy
>> in our modern understanding.
>
> This is so... unpragmatic that it could itself be called an ideology.

Precisely. We can only do what we can, and part of what we can do is
educate ourselves. If we don't educate ourselves, we end up being
able to do less.

Getting depressed at the real world is also not going to help us do what
we can. And that's the way the world has been for as long as I've been
alive, and, near as I can tell, for as far back as we have history.

> best
> - -- t

And I'll echo that. Best.

-- 
Joel Rees

One of these days I'll get someone to pay me
to design a language that combines the best of Forth and C.
Then I'll be able to leap wide instruction sets with a single #ifdef,
run faster than a speeding infinite loop with a #define,
and stop all integer size bugs with my bare cast.
http://defining-computers.blogspot.com/2017/06/reinventing-computers.html

More of my delusions:
http://reiisi.blogspot.com/2017/05/do-not-pay-modern-danegeld-ransomware.html
http://reiisi.blogspot.jp/p/novels-i-am-writing.html

Re: Yes you have standing to sue GRSecurity.

[Joel Rees]

On Mon, Jul 31, 2017 at 3:33 AM, deloptes <delop...@gmail.com> wrote:
> Nicolas George wrote:
>
>> How do you know you can trust that "legal clarification" better than
>> what any of us could have written? I do not have any legal training, and
>> I know approximatively what is written in the first message, but you
>> would be wrong to take my opinion at face value.
>>
>> Have you checked « ni...@redchan.it »'s credentials as an attorney?
>>
>> Regards,
>
> Why do you think I trust you or him? I mean - every one can put forward a
> proposition. The statement was clear and from argumentative point of you
> OK. It was definitely informative. I do not take it as granted but as a
> fair stand point - same as your statement.
>
> There is a lot the community could do, but it spends time being and arguing
> about political correctness and trying to bring all parties together
> etc ... if those guys are too far right, you are too far left - IMO.
>
> regards

One thing that might be worth saying here:

Partial truths are sometimes more damaging than outright lies.

Specifically, GRSecurity is, in fact, misusing the GPL in a way which we
do not want to see becoming common. This is the part the troll is quite
willing to tell.

What the troll is not telling is that they are doing so in response to certain
parties  who are in flagrant violation of the GPL, specifically regarding
their (GrSecurity's) contributions. (And have a record of other, more
general violations.)

Since the original offenders seem to be more willing to throw lawyers
and legal filings at the problem than simply come into compliance
regarding their use of GRSecurities patches, I would question the motives
of the trolls.

Frankly, the large corporations who are doing this with GRSecurity
have no logical reason to be so recalcitrant. The old expression,
"Cutting off their noses to spite their own faces," seems to apply.

If the patches are useful, they should be willing to support the source
of the patches. And if the patches are useful, they should be willing
to help their customers keep their firewalls and other infrastructure
equipment up-to-date.

We may not support GRSecurity's questionable methods, but we may
well decide we should boycott the companies who induced GRSecurity's
stance.

If you want to know who those companies are, you can find out pretty
quickly by searching the web.

-- 
Joel Rees

One of these days I'll get someone to pay me
to design a language that combines the best of Forth and C.
Then I'll be able to leap wide instruction sets with a single #ifdef,
run faster than a speeding infinite loop with a #define,
and stop all integer size bugs with my bare cast.
http://defining-computers.blogspot.com/2017/06/reinventing-computers.html

More of my delusions:
http://reiisi.blogspot.com/2017/05/do-not-pay-modern-danegeld-ransomware.html
http://reiisi.blogspot.jp/p/novels-i-am-writing.html

Re: From dual- to single-boot

[Joel Rees]

On Tue, Jul 25, 2017 at 2:40 PM, solitone <solit...@mail.com> wrote:
> On Tuesday, 25 July 2017 14:25:59 CEST Joel Rees wrote:
>> Can you boot without the Mac OS partition?
>
> I'm using grub to boot debian.
>
> To boot MacOS, I need to press the option key (⌥)  to start up to Apple's
> Startup Manager, rather than grub. Startup Manager allows me to choose the
> MacOS partition, and boot that one.

The reason I ask is that, at least in the past, at least in some
configurations, you needed a bootable Mac partition to boot
anything else. And it might be easy to forget if you had such a
setup and had not been using Mac OS for a while.

Just one more thing to check.

Personally, I've been bitten by a botched partition move in the past,
so I'd tend to avoid moving partitions anyway, if not using LVM.

And if I needed the extra 20G, I'd be foreseeing needing more pretty
soon, so I'd be planning on buying a second drive pretty soon.

Or I could delete the Mac OS partition (after backing anything important
up, of course) and make sure it still boots after formatting the partition for
Linux. If it doesn't it will be much easier to re-install the Mac OS and
necessary boot stuff before you try anything fancy.

Then, instead of moving partitions around, I'd look for what needs the
extra space, and mount the former Mac OS partition there. For instance
if it's space for backups, mount the partition as /backup2/ or
/home/sharedbackup/ or something.

Or, if I could wait, I'd hold things off until I upgrade to a new version,
and restructure my drive(s) while doing the upgrade.

LVM helps avoid botching things when you move things and resize
a lot, and it has better recovery options than when you move things
with a partition editor, so if you think you might be moving things
around a lot, you should look into LVM, as someone else has already
suggested, either with a second drive or when you back up your data
and re-install from scratch on your next upgrade.

-- 
Joel Rees

One of these days I'll get someone to pay me
to design a language that combines the best of Forth and C.
Then I'll be able to leap wide instruction sets with a single #ifdef,
run faster than a speeding infinite loop with a #define,
and stop all integer size bugs with my bare cast.
http://defining-computers.blogspot.com/2017/06/reinventing-computers.html

More of my delusions:
http://reiisi.blogspot.com/2017/05/do-not-pay-modern-danegeld-ransomware.html
http://reiisi.blogspot.jp/p/novels-i-am-writing.html

Re: From dual- to single-boot

[Joel Rees]

On Mon, Jul 24, 2017 at 11:13 PM, solitone <solit...@mail.com> wrote:
> I never use MacOs, so I want to just keep debian, so at least I'll put its 22
> GB space to better use. I used to keep it just for some sporadic firmware
> update, but frankly I don't think I'll need this again in the future.

Can you boot without the Mac OS partition?


> The issue is that MacOs is at the start of the disc:
>
> ~$
> ~$ sudo /sbin/parted /dev/sda print
> Model: ATA APPLE SSD SM0128 (scsi)
> Disk /dev/sda: 121GB
> Sector size (logical/physical): 512B/4096B
> Partition Table: gpt
> Disk Flags:
>
> Number  Start   End SizeFile system Name  Flags
>  1  20.5kB  210MB   210MB   fat32   EFI System Partition  boot,
> esp
>  2  210MB   22.6GB  22.4GB  hfs+
>  3  22.6GB  23.2GB  650MB   hfs+
>  4  23.2GB  31.2GB  8000MB  linux-swap(v1)  swap
>  5  31.4GB  121GB   89.8GB  ext4linux
> ~$
>
> I would use parted from the installation media to delete partitions 1-4,
> recreate the swap at the start (unless I decide to usa a file for the swap),
> and move/extend the ext4 partition.
>
> This seems a bit risky, though. I already asked this, but is there a way to
> completely backup my current system, so that I could quickly restore it on a
> blank new partition, in case everything goes wrong?
>
> I have daily backups of /home, /usr/local, and /etc. But in case I need to
> reinstall from scratch I think I need more.
>
> What's the best approach?
> --
> ⢀⣴⠾⠻⢶⣦⠀
> ⣾⠁⢠⠒⠀⣿⡁Sent from my brain using neurons fueled by glucose.
> ⢿⡄⠘⠷⠚⠋⠀
> ⠈⠳⣄
>



-- 
Joel Rees

One of these days I'll get someone to pay me
to design a language that combines the best of Forth and C.
Then I'll be able to leap wide instruction sets with a single #ifdef,
run faster than a speeding infinite loop with a #define,
and stop all integer size bugs with my bare cast.
http://defining-computers.blogspot.com/2017/06/reinventing-computers.html

More of my delusions:
http://reiisi.blogspot.com/2017/05/do-not-pay-modern-danegeld-ransomware.html
http://reiisi.blogspot.jp/p/novels-i-am-writing.html

Re: Can "PulseAudio Volume Control" devs be redeemed?

[Joel Rees]

On Sat, Jul 22, 2017 at 1:51 PM, Erik Christiansen
<dva...@internode.on.net> wrote:
> After two days of trying to google ways to get audio on the hdmi output
> on a shiny new Udoo X86 running debian 9.0.0, sheer gritted-teeth
> determination, smacking the walls of the GUI rat's maze lucked onto the
> deeply concealed interface.
>
> On the LXDE desktop, the "Sound & Video" -> "PulseAudio Volume Control"
> menu item has only 3 widely spaced tabs, underutilising the chosen
> window width, and the "Output Ports" tab offered no management or
> configuration possibilities.
>
> But there are two tiny dark triangles in the corners. Clicking on the RH
> one leads to an unnecessarily hidden tab, "Input Devices", and clicking
> again reveals "Configuration". There, in a "Profile" selection box, it
> is possible to select "Digital Stereo (HDMI) Output". Once selected, it
> even seems to be the power-on default. (Much to my surprise, given the
> user-hostile perversity of the devious GUI design, deliberately made
> unnecessarily narrow, so that two vital tabs could be hidden from the
> user, without the most tenuous rational reason for doing so.)
>
> Granted, the purpose of a GUI is to put access to necessary functions
> at the end of deep maze rat runs - but invisible secret tabs with double
> blind access?! I do believe that some of these devs are being paid by
> Microsoft to paralyse linux. (If not, we know that they were born
> arse-backwards, and have never turned around.)
>
> There is no rational explanation for failing to make all 5 tabs visible.
>
> Erik
> (Who in 30 years of s/w development never let a team member produce crap
> like that.)
>

I think there are two things going on here.

One is that many devs get huge displays to make it easier to code, and
then forget what the ordinary-sized displays are like. So they get careless
about the constraints imposed by ordinary-sized screens.

Another is that many devs are trying to support tablets without
designing a separate UI for them. So they cram too much in.

-- 
Joel Rees

One of these days I'll get someone to pay me
to design a language that combines the best of Forth and C.
Then I'll be able to leap wide instruction sets with a single #ifdef,
run faster than a speeding infinite loop with a #define,
and stop all integer size bugs with my bare cast.
http://defining-computers.blogspot.com/2017/06/reinventing-computers.html

More of my delusions:
http://reiisi.blogspot.com/2017/05/do-not-pay-modern-danegeld-ransomware.html
http://reiisi.blogspot.jp/p/novels-i-am-writing.html

Re: funding & viability questions of GPL enforcement.

[Joel Rees]

On Sat, Jul 22, 2017 at 3:29 AM, Alessandro Vesely <ves...@tana.it> wrote:
> On Thu 20/Jul/2017 22:18:25 +0200 Fungi4All wrote:
>>[...]
>
>> For linux we all need to agree before we decide.
>
> Yeah, that's a pita.  It's hard to change anything if everyone can veto.

That's sure indication that everything is getting too big -- the companies,
of course, but also the projects, the software, ...

... and the egos.

-- 
Joel Rees

One of these days I'll get someone to pay me
to design a language that combines the best of Forth and C.
Then I'll be able to leap wide instruction sets with a single #ifdef,
run faster than a speeding infinite loop with a #define,
and stop all integer size bugs with my bare cast.
http://defining-computers.blogspot.com/2017/06/reinventing-computers.html

More of my delusions:
http://reiisi.blogspot.com/2017/05/do-not-pay-modern-danegeld-ransomware.html
http://reiisi.blogspot.jp/p/novels-i-am-writing.html

Re: Free software

[Joel Rees]

On Fri, Jul 21, 2017 at 3:46 AM, Doug <dmcgarr...@optonline.net> wrote:
>
> On 07/20/2017 06:32 AM, rhkra...@gmail.com wrote:
>>
>> On Wednesday, July 19, 2017 09:27:16 PM Gene Heskett wrote:
>>>
>>> Doug is correct. Every shop had a subscription to SAM's and toward the
>>> end as many as 9 or 10, tall 4 drawer fileing cabinets to keep the stuff
>>> in if the subscription was for all of the stuff.
>>
>> Ahh, Sam's was a good clue (for me)--I think the series was called Sam's
>> Photofacts.
>>
>>
>>
> You can look it up on the Internet. Just put Sam's Photofact into
> your browser!
>
> (I remember it as Photofacts, just like rhkramer, but the 'net
> has it as singular.)

Well, a couple of filing cabinets full would be plural, wouldn't it?

-- 
Joel Rees

One of these days I'll get someone to pay me
to design a language that combines the best of Forth and C.
Then I'll be able to leap wide instruction sets with a single #ifdef,
run faster than a speeding infinite loop with a #define,
and stop all integer size bugs with my bare cast.
http://defining-computers.blogspot.com/2017/06/reinventing-computers.html

More of my delusions:
http://reiisi.blogspot.com/2017/05/do-not-pay-modern-danegeld-ransomware.html
http://reiisi.blogspot.jp/p/novels-i-am-writing.html

Re: Free software

[Joel Rees]

On Thu, Jul 20, 2017 at 8:04 AM, Doug <dmcgarr...@optonline.net> wrote:
>
> On 07/19/2017 05:44 PM, Joel Rees wrote:
>>
>>
>> This is another aspect of "closed source" gratis technology that is
>> often swept under the rug.
>>
>> It used to be, for instance, that a TV in the US had a full diagram
>> of working parts in the back case, so that the TV could still be
>> fixed even if the manufacturer suddenly wiped their books and
>> disappeared.
>>
>>
> Not at all true! As a sideline I was a TV serviceman in the 1960s.
> There usually was  a drawing of the tube numbers and positions

Thus, the working parts. Except it was not just the tubes, it usually
included whatever an independent technician could get as a "part".

> somewhere in the set--more usually on an inside surface of the
> wooden box.

Yeah, in the back [of the] case. :-/

> There certainly was no schematic diagram.

What do you call a schematic diagram, then?

> However, it was almost always possible to obtain real service
> information including schematic diagrams of the circuits from
> a paid service,

SAMS was one of the services which provide more detailed
schematic diagrams. Their existence owe no small debt to the
fact that "intellectual property" rights of inventors of
usually-not-all-that-new art were not allowed by the law then
to take precedence over the rights of the inventors of prior art,
nor of the community that gives birth to art over the existing
state-of-the-art.

And there's another word whose legal definition seems to have
changed: state-of-the-art seems to now mean to lawyers
something it cannot logically mean to those who practice the art.
Too many salescrew-turned-lawyers, maybe.

> the name of which escapes me now. (The pages
> always included useless ones for record players and such that
> nobody ever heard of!)

Funny. I often used those useless diagrams for record players
that I guess you never heard of. Maybe it was because I used them
in the seventies, not the sixties? ;-)

For those who are missing the allegories, source code is somewhat
the parallel of those diagrams, and we don't have them now except
in very rare cases.

And, where those diagrams kept still-usable electronics out of the
landfill for a few more years, lack of the source code results in a lot
of waste in the current economy, contributing to pollution and other
things that don't, ultimately, help the economy.

-- 
Joel Rees

One of these days I'll get someone to pay me
to design a language that combines the best of Forth and C.
Then I'll be able to leap wide instruction sets with a single #ifdef,
run faster than a speeding infinite loop with a #define,
and stop all integer size bugs with my bare cast.
http://defining-computers.blogspot.com/2017/06/reinventing-computers.html

More of my delusions:
http://reiisi.blogspot.com/2017/05/do-not-pay-modern-danegeld-ransomware.html
http://reiisi.blogspot.jp/p/novels-i-am-writing.html

Re: Free software

[Joel Rees]

FTR

On Wed, Jul 19, 2017 at 5:46 PM, Fungi4All <fungil...@protonmail.com> wrote:
>
>>> From: dmcgarr...@optonline.net
>>> To: Gene Heskett <ghesk...@shentel.net>, debian-user@lists.debian.org
>>> [ important stuff, check the archives]
>>> Cheers, Gene Heskett
>> I thought I put this to bed, but apparently not.   
>>
>
> You thought well because if we are clarifying the description
> of reality we can not utilize the grocery store logic of choice. Either
> one final thought convinces everyone that it is correct or there
> still is room for discussion.

... or we decide we are tired of discussion, which sometimes is not a
good thing.

> I have no beef with nvidia in specific,
> they are just as bad as any for-profit organization, so it is not
> specific.
>
>> I"ve stated my case. Let"s drop it here!

You have a lousy attitude. You get to air your armpits, but no one
who disagrees with you does.

> But this is the problem, you can not be one sided in having
> the authority to end a conversation.  If Nvidia one day decides
> it is not profiting from their little monopoly and decides to
> fold its gc production and r, they have the "legal right" to
> destroy all knowledge produced and owned by them.  The
> work and findings of hundreds of people developing nvidia
> products is in the hands of one entity (stock-holders).
> The state gives them the right to conduct this atrocity of
> destroying knowledge and deny society access to it, even
> when it has no value for them anymore.

This is another aspect of "closed source" gratis technology that is
often swept under the rug.

It used to be, for instance, that a TV in the US had a full diagram
of working parts in the back case, so that the TV could still be
fixed even if the manufacturer suddenly wiped their books and
disappeared.

> I believe society has the right to criticize this mutual practice
> by state and owner of knowledge.  Any industry is a school for
> society and it exists due to lack of alternatives for society.
> This is what we are talking about, I think.  Internalizing the
> logic of the capitalist and its puppet the state, the logic of
> market, into our conversation as "logical" is a product of
> propaganda and demagogues of the media and other
> "institutions".  A society must be able to survive having
> alternatives past capitalism.
>
> [...]

Again, just for the record, politics itself may be off topic, but we
have to give each other a little leeway where politics intersects
with the license which allows us to cooperate peacefully in this
community.

-- 
Joel Rees

One of these days I'll get someone to pay me
to design a language that combines the best of Forth and C.
Then I'll be able to leap wide instruction sets with a single #ifdef,
run faster than a speeding infinite loop with a #define,
and stop all integer size bugs with my bare cast.
http://defining-computers.blogspot.com/2017/06/reinventing-computers.html

More of my delusions:
http://reiisi.blogspot.com/2017/05/do-not-pay-modern-danegeld-ransomware.html
http://reiisi.blogspot.jp/p/novels-i-am-writing.html

Re: so much for your ascii only emails and 80 char lines :)

[Joel Rees]

On Wed, Jul 19, 2017 at 4:58 AM, Dominik George <n...@naturalnet.de> wrote:
> it? Thanks!
> topic and stop
> gone horribly off-
> that this thread has
> all just agree
> So, can we
> Sent from my very colourful mailer which encodes as it pleases.
>

ROFCOL :)

Just for reference, this thread just misses paralleling a thread on
misc@openbsd:

https://marc.info/?l=openbsd-misc=149984510728808=2

-- 
Joel Rees

One of these days I'll get someone to pay me
to design a language that combines the best of Forth and C.
Then I'll be able to leap wide instruction sets with a single #ifdef,
run faster than a speeding infinite loop with a #define,
and stop all integer size bugs with my bare cast.
http://defining-computers.blogspot.com/2017/06/reinventing-computers.html

More of my delusions:
http://reiisi.blogspot.com/2017/05/do-not-pay-modern-danegeld-ransomware.html
http://reiisi.blogspot.jp/p/novels-i-am-writing.html

Re: perl system integrity?

[Joel Roth]

Miles Fidelman wrote:
> On 7/18/17 12:52 PM, Jude DaShiell wrote:
> 
> >Does any means exist within perl to detect that a perl base installed on a
> >system is completely intact?  Perl base is perl binary and all support
> >packages that binary uses before cpan or cpanp is run.  The perl base is
> >put on newly installed systems.
> >I am interested in this since sometimes system updates break perl and
> >steps have to be taken to clear the breakage then repair steps need to be
> >taken.  If a means exists to let a user know this list of packages came
> >with your installed system and are now missing perl system repairs could
> >be expedited.
> >Aside from perl, such capabilities added to other base systems like python
> >for one example could help out repairing those other systems too.
> 
> Perl has the best collection of regression tests that I've ever seen.  When
> you install from a source tarball, and run "make test" - everything gets
> checked.  Same again when you install/update modules via cpan.  It takes a
> while, but you know everything is working.
> 
> Personally, when I'm using perl, I always install from scratch, and maintain
> using cpan - packagers & packaging just muck it up.

The Debian Perl team does an amazing job of packaging -- and
patching -- the base perl as well as thousands of
distributions from CPAN. Their bug reports and patches
are a significant contribution to upstream maintenance.

Debian packaging handles external dependencies of perl
modules, which the perl toolchain is not capable of
handling on its own. For casual, trouble-free installations,
the system perl is excellent. 

That said, I generally compile perl from sources and install
it in my $HOME directory, along with any CPAN modules that I
need.

There are various ways to install non-system perl. 
The newest and neatest IMO is plenv, which allows multiple 
perl versions to coexist. You can select a global perl
version to use, and override it in individual directories.
That floats my boat, however perlbrew and local::lib
can also be useful.

I find that cpanm (App::cpanminus)  is the most convenient
way to install CPAN modules.

Coming to the OP's question, I agree with the poster who
said that perl's testing framework is awesome. If there is a
problem upgrading system perl, it's a bug. If reported, the
Debian perl team will investigate it thoroughly.

Problems are most likely with an unintended mix of system
perl, perl installed by root in /usr/local, or unprivileged
perl installed in $HOME.

Which perl is picked up by a shebang line #!/usr/bin/env
perl will depend on your $PATH variable. 

cheers,

Joel
 
> Miles Fidelman
> 
> 
> -- 
> In theory, there is no difference between theory and practice.
> In practice, there is.   Yogi Berra
> 

-- 
Joel Roth
  

Re: funding & viability questions of GPL enforcement.

[Joel Rees]

On Sun, Jul 16, 2017 at 8:47 PM, Alessandro Vesely <ves...@tana.it> wrote:
> There are salient discrepancies in copylefting collective work —as there are
> mismatches in working as a free software developer in a western economic 
> model.
>

There are salient discrepancies in every licensing model, so-called
free/libre, free/open, free-as-in-beer, sell-your-first-child-shrinkwrap,
etc.

Shoot, the entire concept of property is still not worked out well.

It only works if we agree to cooperate.

And people and companies who receive excess and refuse to return
it to the market simply are not cooperating, irregardless of either license
or external economic system. (Excess includes not only money and less
tangible proxies for value, but also control, which is one place where
communism and socialism historically fail.)

The whole idea that they have to protect themselves from users of their
so-called intellectual property is where we, as a society are failing to do
the reality check.

>  Let me just say that this discussion, working out the legal details of the
> problem, is very interesting.  I guess that's how every inch of freedom has to
> be conquered, and I'm delighted that this list allows me to witness it.

I think you meant "fought for" and not "conquered"?

> Please go ahead.
>
> On Sat 15/Jul/2017 19:24:56 +0200 Bradley M. Kuhn wrote:
>> [...]
>> Finally, this is probably a good moment -- since this thread has erupted on
>> a Debian Mailing List -- to let everyone know that Conservancy also
>> organizes a GPL copyright aggregation project for Debian contributors as
>> well, see: https://sfconservancy.org/copyleft-compliance/#debian and
>> https://sfconservancy.org/news/2015/aug/17/debian/.

Crowd-funding without the middleman!

> [...]

-- 
Joel Rees

One of these days I'll get someone to pay me
to design a language that combines the best of Forth and C.
Then I'll be able to leap wide instruction sets with a single #ifdef,
run faster than a speeding infinite loop with a #define,
and stop all integer size bugs with my bare cast.
http://defining-computers.blogspot.com/2017/06/reinventing-computers.html

More of my delusions:
http://reiisi.blogspot.com/2017/05/do-not-pay-modern-danegeld-ransomware.html
http://reiisi.blogspot.jp/p/novels-i-am-writing.html

Re: Why does no one care that Brad Spengler of GRSecurity is blatantly violating the intention of the rightsholders to the Linux Kernel?

[Joel Rees]

On Tue, Jun 20, 2017 at 7:12 AM, deloptes <delop...@gmail.com> wrote:
> Richard Stallman wrote:

I don't think you should assume RMS is monitoring the debian list.

>> I am not trying to study the GRsecurity case because (0) it's
>> complicated, and it would take a lot of time to think about, (1) the
>> FSF has no say in the matter (it is about Linux) and (2) I don't think
>> the copany would heed whatever I might say.
>
> Could you explain why it should be complicated?

I don't want to pretend to be answering for Mr. Stallman, but I have done
a small bit of reading now that I see Bruce Perens is taking the time to
get involved.

> GPL states the rights
> obtained should be passed to the recipient, so the recipient should be
> allowed to redistribute the code (IMO) even if he/she is paying for
> improvements.
>
> It would be really nice if GRSec could help improve the kernel security in
> some way acceptable by and for the benefit of all. I don't think someone
> wants to punish them for what they are doing. It would be better to have
> mutual benefit if possible as the GPL does not prohibit modifying and
> redistributing the code and demanding a fee, it however does guarantee the
> right to redistribute is passed to the recipient, which is not the case
> here.

GRSecurity has posted their complaints here:

https://grsecurity.net/announce.php

They have a point, although bending the rules is not a good way to make
your point, in general. (And I'll note that they seem to be thinking they are
following RedHat's example here. At this point I'm more than half inclined
to think they might be following RedHat's example, for what that's worth.)

TheReg's recent article

https://www.theregister.co.uk/2017/04/26/grsecurity_linux_kernel_freeloaders/

indicates they think they might know which large, well-heeled, well-
financed major embedded industry player provided the straw that broke
GRSecurity's camel's back. But it is not just one player. The entire
embedded industry does not seem to understand how their products
came to be or how their future products will come to be.

That said, GRSecurity needs to find a different way to seek redress.

And someone in the community needs to find them a lawyer who will
take their case.

The GPL is a gentlemen's agreement.

When the members of the market refuse to behave like gentlemen, it
destroys the value of the agreement. It makes the agreement useless.

It will also destroy the market, so the moneyed players who are not
paying their fair share back into supporting the "small players" are
basically shortening the life-span of their own companies -- cutting
off their own noses to spite their faces, as we used to say.

I am personally not impressed with GRSecurity's hubris. Their tech
is only impressive in that it sort of helps make up a little bit for the
serious lacks in the security of every major CPU available today,
but especially the ones from Intel. So they do make a contribution,
or have until recently.

If the OP wants to solve this problem, (s)he has done enough
rabble-rowsing this way. He needs to start asking everyone he
knows if they know a good lawyer willing to work on contingency.
The rest of us who are concerned probably should, as well.

Or, perhaps, GRSecurity should go to one of the crowdfunding
sites. Or maybe someone could start a new crowdfunding site to
specialize in providing legal relief for the small guy. (Not sure how
that would work.)

(Yeah, I am willing to name and shame Intel here. If our civilization
survives the next two decades, our children will remember Intel's
processors with the same phrase that Ralph Nader made popular
relative to the auto industry. Who can compete when Intel refuses
to pay the price of making CPUs that are unsafe at progressively
higher speeds?)

-- 
Joel Rees

One of these days I'll get someone to pay me
to design a language that combines the best of Forth and C.
Then I'll be able to leap wide instruction sets with a single #ifdef,
run faster than a speeding infinite loop with a #define,
and stop all integer size bugs with my bare cast.
http://defining-computers.blogspot.com/2017/06/reinventing-computers.html

More of my delusions:
http://reiisi.blogspot.com/2017/05/do-not-pay-modern-danegeld-ransomware.html
http://reiisi.blogspot.jp/p/novels-i-am-writing.html

Re: FYI: systemd session logging - no solution

[Joel Rees]

On Mon, Jul 10, 2017 at 11:52 PM, Václav Ovsík <vaclav.ov...@gmail.com> wrote:
> Lately I found this:
> https://github.com/systemd/systemd/issues/1291
> so no nice solution unfortunately :(.
> --
> Zito
>

I want to show that to Poettering's manager.

-- 
Joel Rees

One of these days I'll get someone to pay me
to design a language that combines the best of Forth and C.
Then I'll be able to leap wide instruction sets with a single #ifdef,
run faster than a speeding infinite loop with a #define,
and stop all integer size bugs with my bare cast.
http://defining-computers.blogspot.com/2017/06/reinventing-computers.html

More of my delusions:
http://reiisi.blogspot.com/2017/05/do-not-pay-modern-danegeld-ransomware.html
http://reiisi.blogspot.jp/p/novels-i-am-writing.html

Re: N’hésites pas de commencer une conversation avec moi Jade

[JOEL BOSCHERON]

J attend ton appel

Envoyé de mon iPhone

> Le 4 juil. 2017 à 12:15, Jade Sovansi  a écrit :
> 
> Je ne mords pas tu sais. Bon, sauf si t le demande… 
> http://bit.ly/2sHRIiy

Re: How to gain control over the system?

[Joel Rees]

On Mon, Jul 10, 2017 at 7:11 AM, Kaj Persson <kape_...@algonet.se> wrote:
> Hi Jimmy,
> Well, I did not follow your suggestion exactly, but as people has said, the
> root account is already and always  there, even it has not been assigned a
> password. So, against my real whish, not to activate the root account, I
> gave the command sudo passwd root, and entered a password. And now I suppose
> I have burned my ships and have no way back...

Of course you have a way back.

   man vipw

   man 5 passwd

   man 5 shadow

and note the -s option.

   man nologin

   man false

Then

   sudo vipw

and change the line for root (should be the very first line) to give it
either /bin/false or /sbin/nologin as the default shell for root.

(That's the last field.)

   cat /etc/passwd

after you're done, to make sure you saved it. Then,

   sudo vipw -s

and replace the encrypted password (second ffield) there with '*'.

 > But! Nothing has changed. I can still not enter program icons to the panel,
> and not define keyboard shortcuts. If I sort the icons on the desktop they
> still, after a cold start, come back in a completely other order, dispite I
> had marked "Keep ajusted" (right click on desktop).
>
> So...?
> /Kaj

Have you checked group ownership and  permissions?

Also, have you checked your mount parameters?

And have you checked whether you have established SELinux or acl
permissions or anything of that ilk?

(BTW, do you keep a backup of your /home partition? I usually find
myself using cp -p or tar to copy the files from the old /home to the
new one, instead of keeping an old /home around.)

-- 
Joel Rees

One of these days I'll get someone to pay me
to design a language that combines the best of Forth and C.
Then I'll be able to leap wide instruction sets with a single #ifdef,
run faster than a speeding infinite loop with a #define,
and stop all integer size bugs with my bare cast.
http://defining-computers.blogspot.com/2017/06/reinventing-computers.html

More of my delusions:
http://reiisi.blogspot.com/2017/05/do-not-pay-modern-danegeld-ransomware.html
http://reiisi.blogspot.jp/p/novels-i-am-writing.html

Re: Installing Debian on an android device

[Joel Rees]

On Mon, Jul 10, 2017 at 12:00 AM, deloptes <delop...@gmail.com> wrote:
> [...]
> As for some conspiracy chips with embedded rom  if you have basic
> engineering knowledge you could easily identify all of it and to my
> knowledge it is not trivial to embed such a chip into a mass product,
> especially a mobile phone.

Have you read Ken Thompson's On Trusting Trust?

If it is true of software, it is true of hardware.

> There were servers in the past, where NSA plugged in special chips before
> those machines were shipped to china, but it did not last long and the
> chineese found out.

Either that was in the days when the NSA didn't really have a lot of skilled
engineers, or they wanted the special chips to be found.

I'd tend to guess at the latter, because I know how easy it would be to
bundle a custom soc with an extra processor and private memory with
any standard bit of circuit, package it in a standard-looking package,
and label it with the name and package numbering of any company
of choice. This has been easy for a long time.

> So keep your eyes open and think twice before you buy something and use it
-
> this is my advise

-- 
Joel Rees

One of these days I'll get someone to pay me
to design a language that combines the best of Forth and C.
Then I'll be able to leap wide instruction sets with a single #ifdef,
run faster than a speeding infinite loop with a #define,
and stop all integer size bugs with my bare cast.
http://defining-computers.blogspot.com/2017/06/reinventing-computers.html

More of my delusions:
http://reiisi.blogspot.com/2017/05/do-not-pay-modern-danegeld-ransomware.html
http://reiisi.blogspot.jp/p/novels-i-am-writing.html

Re: shadow spam (was Re: stop your mail)

[Joel Rees]

On Sun, Jul 9, 2017 at 7:53 PM, Thomas Schmitt <scdbac...@gmx.net> wrote:
> Hi,
>
[...]
>
> Joel Rees wrote:
>> (1) These messages may be a sort of generator for phishing targets.
>
> You mean that those who hit the "Smack Sender" button of their mail
> app show up as flotsam here and can be harvested without reveiling
> the harvester's mail address ?
> (This theory would imply that the reflector senders are real people
>  or their watchdog apps.)
>
> Eek. That would mean we would really have to take measures to not
> let appear most of the messages in subscriber mailboxes and archive.
> If we let this continue then we create a commercial incentive to
> flood us.

Of course, if the hypothetical "they" are looking for a commercially viable
way to harvest addresses from this list and are doing this, they've missed
something much more obvious.

And?

>> they might be setting up a noise
>> background against which to send steganographically encoded messages.
>
> That's a good one.
> We are testing ground for a novel low-bandwidth method to control
> bot nets or remote spies.

Not likely a testing ground.

> Ten hops over iPads, Galaxies, or WinPhones would be nearly as
> effective in hiding the sender as a Tor onion would be.
>
>
> Have a nice day :)
>
> Thomas
>

Did I say something about onions?

-- 
Joel Rees

One of these days I'll get someone to pay me
to design a language that combines the best of Forth and C.
Then I'll be able to leap wide instruction sets with a single #ifdef,
run faster than a speeding infinite loop with a #define,
and stop all integer size bugs with my bare cast.
http://defining-computers.blogspot.com/2017/06/reinventing-computers.html

More of my delusions:
http://reiisi.blogspot.com/2017/05/do-not-pay-modern-danegeld-ransomware.html
http://reiisi.blogspot.jp/p/novels-i-am-writing.html

shadow spam (was Re: stop your mail)

[Joel Rees]

Can I suggest two possibilities not apparently being considered?

(1) These messages may be a sort of generator for phishing targets.
(This is not currently a likely scenario, but you want to consider it.)

(2) These might be either the body of a message sent by a spatter
steganography technique, or they might be setting up a noise
background against which to send steganographically encoded
messages.

I'd suggest a third, which is true tin-foil-hat stuff, but you who are into
conspiracy theories can work that out yourselves.

Whenever I see a sudden rise in odd-looking spam, I tend to
assume something like the second possibility.

https://en.wikipedia.org/wiki/Steganography

-- 
Joel Rees

One of these days I'll get someone to pay me
to design a language that combines the best of Forth and C.
Then I'll be able to leap wide instruction sets with a single #ifdef,
run faster than a speeding infinite loop with a #define,
and stop all integer size bugs with my bare cast.
http://defining-computers.blogspot.com/2017/06/reinventing-computers.html

More of my delusions:
http://reiisi.blogspot.com/2017/05/do-not-pay-modern-danegeld-ransomware.html
http://reiisi.blogspot.jp/p/novels-i-am-writing.html