Re: Temporary failure in name resolution

[Alexander V. Makartsev]

On 31.03.2021 07:21, Dan Norton wrote:

After the 10.9 upgrade, name resolution is not working for me. Does anyone else 
see this?

My desktop is a wifi server for laptop access using windows. That works OK but 
the server, attached by ethernet to the DSL modem does not get names resolved 
since the upgrade. The resolvconf program is not installed according to whereis.

# cat /etc/resolv.conf
domain attlocal.net
search attlocal.net
nameserver 192.168.1.254

If that's not the right nameserver then what is?


Is "192.168.1.254" an IP address of your DSL modem?
If you don't need to resolve hostnames from you local network, like 
"somepc1.attlocal.net" and only want to access the Internet, you can 
configure one or more of the public DNS servers.

From Google [1]:
8.8.8.8
8.8.4.4
From CloudFlare [2]:
1.1.1.1

[1] https://developers.google.com/speed/public-dns/docs/using
[2] https://one.one.one.one/

--
With kindest regards, Alexander.

⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org
⠈⠳⣄

Re: xsane can't see Brother ADS-2700W scanner

[Alexander V. Makartsev]

On 31.03.2021 07:24, Charlie Gibbs wrote:


If anyone has gotten one of these newfangled machines to work
as a slave, rather than a master, please share your secrets.

aTdHvAaNnKcSe...

Have you tried to install a sane backend driver coming as a .deb package 
from Brother website?

https://support.brother.com/g/b/downloadtop.aspx?c=eu_ot=en=ads2700w_us_eu_as


--
With kindest regards, Alexander.

⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org
⠈⠳⣄

Temporary failure in name resolution

[Dan Norton]

After the 10.9 upgrade, name resolution is not working for me. Does anyone else 
see this?

My desktop is a wifi server for laptop access using windows. That works OK but 
the server, attached by ethernet to the DSL modem does not get names resolved 
since the upgrade. The resolvconf program is not installed according to whereis.

# cat /etc/resolv.conf
domain attlocal.net
search attlocal.net
nameserver 192.168.1.254

If that's not the right nameserver then what is?

$ ping google.com
ping: google.com: Temporary failure in name resolution

$ ping -c2 8.8.8.8
PING 8.8.8.8 (8.8.8.8) 56(84) bytes of data.
64 bytes from 8.8.8.8: icmp_seq=1 ttl=117 time=21.1 ms
64 bytes from 8.8.8.8: icmp_seq=2 ttl=117 time=21.2 ms

--- 8.8.8.8 ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 2ms
rtt min/avg/max/mdev = 21.148/21.188/21.229/0.151 ms

# systemctl status systemd-resolved
shows it being active and "Processing requests...".

Nothing in history.log and term.log is obviously wrong to me FWIW.

# journalctl -b0
gives no clue to me.

No doubt it is something simple(tm). Where should I look or what should I 
re-read?

 - Dan Norton

xsane can't see Brother ADS-2700W scanner

[Charlie Gibbs]

I just got a Brother ADS-2700W sheet-fed scanner and am trying
to access it from xsane.  I've done a lot of flatbed scanning,
first with an HP 3970, and lately with an Epson WF-2650 all-in-one,
but I have a lot of old manuals I want to scan and upload to
Bitsavers, and a sheet feeder will speed the process along.

The Brother got a lot of good reviews so I decided to give it a try.
It offers many options, such as e-mail, [S]FTP, etc. over Ethernet,
wi-fi, and USB.  But so far, I haven't been able to get xsane to
recognize it.  My wife tried to get at it from her Macbook (which
accesses the Epson with no trouble), but had no luck either.
It's not a connectivity issue - the scanner happily connects
to my wi-fi and gets an IP address, and I can access it from
a web browser and get at all of its configuration screens.
But neither xsane nor my wife's Macbook can see it.

The one way I did manage to get the scanner to work was to a
USB flash drive.  It quickly sucked in a handful of sheets,
scanned both sides, and wrote them to a file on the stick.
If all else fails, I can work with it that way.  But I'd
really like to let xsane manage the process.

I'm beginning to wonder, though, whether fashions are changing.
Scanners nowadays seem to want to push data to a server, rather
than being commanded to scan by a computer.  Is this really
happening?  If so, whither (or should that be "wither") xsane?

If anyone has gotten one of these newfangled machines to work
as a slave, rather than a master, please share your secrets.

aTdHvAaNnKcSe...

--
/~\  Charlie Gibbs  |  They don't understand Microsoft
\ /|  has stolen their car and parked
 X   I'm really at ac.dekanfrus |  a taxi in their driveway.
/ \  if you read it the right way.  |-- Mayayana

Re: DNS problems on Raspberry Pi 400 (Debian 10.9)

[Nicholas Geovanis]

On Tue, Mar 30, 2021, 5:33 PM Moritz Kempe  wrote:

> Hello,
>
> since i upgraded my Raspberry Pi 400 (with regular Debian 10, not
> Raspberry PI OS) to the latest buster version, i am experiencing
> problems with dns, which i cannot replicate with any of my other devices
> (Debian Stretch amd64 workstation, Debian Buster amd64 server, Raspberry
> Pi 3b armhf raspberrypi OS). But i can replicate this behavior with
> another another micro sd flashcard with Twister OS on the (same)
> Raspberry Pi 400. Twister OS is an Debian / Raspberry Pi OS based OS
> which helps you to game on the Raspberry Pi 400.
>
> I noticed the problem, while i was browsing the internet and got
> confused because after a while some domains could not longer be
> found/connected to by the browser. (On both, Firefox and Chromium)
>
> -- Firefox
> Hmm. We’re having trouble finding that site.
>
> We can’t connect to the server at github.com.
>
> If that address is correct, here are three other things you can try:
>
>  Try again later.
>  Check your network connection.
>  If you are connected but behind a firewall, check that Firefox has
> permission to access the Web.
> --
>
> -- Chromium
> This site can’t be reachedCheck if there is a typo in github.com.
> DNS_PROBE_FINISHED_NXDOMAIN
> --
>
> And as time passes, more and more website will disappear.
> Which means, that after a hour only barely visited domains are left.
>
> And even stranger is the output of different dns tools:
>

Never saw that before said the man on fire :-) So just guesses:

Some weird DNS TTL issue? Or TTL incompatibility?
Strange DNS local caching issue? Maybe check resolv.conf and name service
config for new options or something.
Do you see any unexpected DHCP messages in the logs, maybe further upstream
from the pi's?

-- host github.com
> moke@rpi4-20201112:~$ host github.com
> Host github.com not found: 2(SERVFAIL)
> --
>
> -- dig github.com
> ; <<>> DiG 9.16.13-Debian <<>> github.com
> ;; global options: +cmd
> ;; Got answer:
> ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 30900
> ;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 1
>
> ;; OPT PSEUDOSECTION:
> ; EDNS: version: 0, flags:; udp: 1232
> ;; QUESTION SECTION:
> ;github.com.IN  A
>
> ;; ANSWER SECTION:
> github.com. 27  IN  A   140.82.121.3
>
> ;; Query time: 343 msec
> ;; SERVER: 10.0.0.1#53(10.0.0.1)
> ;; WHEN: Tue Mar 30 23:31:01 CEST 2021
> ;; MSG SIZE  rcvd: 55
> --
>
> Dig seems to be able to get the ip from github.com but (the program)
> host, firefox and chromium can't.
>
> Some system information
>
> --
> moke@rpi4-20201112
> --
> OS: Debian GNU/Linux bullseye/sid aarch64
> Host: Raspberry Pi 400 Rev 1.0
> Kernel: 5.10.0-5-arm64
> Resolution: 1440x900, 1280x1024
> WM: awesome
> CPU: (4) @ 1.800GHz
> Memory: 606MiB / 3791MiB
> --
>
> My Raspberry Pi 400 is connected over Ethernet to my home router.
>
> Is there someone else, who experiences / has experienced this problem?
>
> PS: After i reboot the system all domains get unlocked (until they get
> locked again).
>
> Kind Regards
> Moritz Kempe
>
>

DNS problems on Raspberry Pi 400 (Debian 10.9)

[Moritz Kempe]

Hello,

since i upgraded my Raspberry Pi 400 (with regular Debian 10, not 
Raspberry PI OS) to the latest buster version, i am experiencing 
problems with dns, which i cannot replicate with any of my other devices 
(Debian Stretch amd64 workstation, Debian Buster amd64 server, Raspberry 
Pi 3b armhf raspberrypi OS). But i can replicate this behavior with 
another another micro sd flashcard with Twister OS on the (same) 
Raspberry Pi 400. Twister OS is an Debian / Raspberry Pi OS based OS 
which helps you to game on the Raspberry Pi 400.


I noticed the problem, while i was browsing the internet and got 
confused because after a while some domains could not longer be 
found/connected to by the browser. (On both, Firefox and Chromium)


-- Firefox
Hmm. We’re having trouble finding that site.

We can’t connect to the server at github.com.

If that address is correct, here are three other things you can try:

Try again later.
Check your network connection.
If you are connected but behind a firewall, check that Firefox has 
permission to access the Web.

--

-- Chromium
This site can’t be reachedCheck if there is a typo in github.com.
DNS_PROBE_FINISHED_NXDOMAIN
--

And as time passes, more and more website will disappear.
Which means, that after a hour only barely visited domains are left.

And even stranger is the output of different dns tools:

-- host github.com
moke@rpi4-20201112:~$ host github.com
Host github.com not found: 2(SERVFAIL)
--

-- dig github.com
; <<>> DiG 9.16.13-Debian <<>> github.com
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 30900
;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 1

;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 1232
;; QUESTION SECTION:
;github.com.IN  A

;; ANSWER SECTION:
github.com. 27  IN  A   140.82.121.3

;; Query time: 343 msec
;; SERVER: 10.0.0.1#53(10.0.0.1)
;; WHEN: Tue Mar 30 23:31:01 CEST 2021
;; MSG SIZE  rcvd: 55
--

Dig seems to be able to get the ip from github.com but (the program) 
host, firefox and chromium can't.


Some system information

--
moke@rpi4-20201112
--
OS: Debian GNU/Linux bullseye/sid aarch64
Host: Raspberry Pi 400 Rev 1.0
Kernel: 5.10.0-5-arm64
Resolution: 1440x900, 1280x1024
WM: awesome
CPU: (4) @ 1.800GHz
Memory: 606MiB / 3791MiB
--

My Raspberry Pi 400 is connected over Ethernet to my home router.

Is there someone else, who experiences / has experienced this problem?

PS: After i reboot the system all domains get unlocked (until they get 
locked again).


Kind Regards
Moritz Kempe

Re: Possible?! A Debian public repository for all complex code lines with examples and scripts?

[Charles Curley]

On Tue, 30 Mar 2021 17:20:11 +0300
George Shuklin  wrote:

> My point was that 'sh' approach (including 
> bash and dash) is really, really bad. Bash as a programming language
> is super tricky, with lots of landmines and almost no validation
> whatsoever for errors (there is a bit, but way less than expected
> from any modern system language).

Something that may help this problem is shellcheck. If you use emacs,
you can have emacs check on the fly with package elpa-flycheck. I
expect other editors have similar support.

-- 
Does anybody read signatures any more?

https://charlescurley.com
https://charlescurley.com/blog/

Re: Creating my first LAN

[David Wright]

On Tue 30 Mar 2021 at 16:51:05 (+0100), mick crane wrote:
> On 2021-03-24 15:34, Richard Owlett wrote:
> > This is essentially a reading list request.
> > I have never administered a LAN and believe in "learning by doing".
> > I have two laptops with clean installs of Buster. During installation
> > server software was installed on *ONE* of them. Communication will be
> > via WiFi. The MATE panel's network connection icon for each laptop
> > shows the existence of WiFi sources in the neighborhood.
> > 
> > Any specific suggestions for reading?
> > What internet search terms will lead to productive results?
> 
> I'd get the wired connection to work first.
> You might need a x-over RJ45 cable to connect the 2 together.
> A x-over cable has the red and green wires swapped over
> red > green
> green > red
> red stripey > green stripey
> green stripey > red stripey

s/red/orange/

But if one of the machines is the OP's Dell E6410, then its
ethernet port will happily cooperate with the other end anyway.

> If you put them in the same network address range
> 10.0.0.1, 10.0.0.2
> subnetmask 255.255.255.0
> you might be able to ping each other then you could get ssh working.

They should be able to interoperate without any manual configuration
because an IPv6 local loop address should configure itself
automatically at both ends.

So you just need to broadcast ping, specifying the outgoing interface,
ie that of the machine you're typing on:

$ ip a
 [ … ]
2: enp3s0: [ … ]   ← en… is the ethernet interface.
 [ … ] ↘ copy it after the % (no colon).
$↘
   ↘
 ↘
   ↘
$ ping6 ff02::1%enp3s0
64 bytes from fe80::212:3456:fe12:3456%enp3s0: icmp_seq=1 ttl=64 time=0.099 ms
 [ … ]
^C

Cheers,
David.

Re: Creating my first LAN

[Brian]

On Tue 30 Mar 2021 at 19:29:49 +0100, Brian wrote:

> On Tue 30 Mar 2021 at 11:19:48 -0500, Richard Owlett wrote:
> 
> [...]
> 
> > What well meaning members of this list have not picked up on is that I am
> > very literal minded and careful in phrasing my questions. I've said
> > somewhere in this thread that my "universe of discourse" is explicitly
> > limited to Laptop-1 and Laptop2. I've not been a sophomore B.S.E.E. student
> > for over a half-century ;} But I can see a Sophomore lab being set up with
> > my specified constraints. Since this thread started I have been reading
> > about "ad-hoc networks". That's my goal. Real world has temporarily
> > interrupted my pursuit.
> 
> You have to bear in mind that other users read these posts and maybe
> have similar objectives in mind. Posts veering away from the caefully
> phrased question could spark their interest and lead them towards a
> solution to an issue

Ten minutes after writing that I recollected I possess a non-working
router. It is the first I bought (> 100 GBP) and has not yet beem
re-cycled.

My network also hasn't any spare LAN ports to connect to and some of
the devices have beem wireless-enabled to communicate with them.

Now - could I use this non-internet-capable router as a switch?

See what I was getting at? An idea that is unrelated to your first
post. Carefully phrased or not, a thread cannot be actively channelled
to meet the demands of its originator.

-- 
A well-meaning -debian member.

Re: [Résolu] Re: suivre bullseye en stable

[ajh-valmer]

On Tuesday 30 March 2021 20:22:38 Benoit B wrote:
> > > Y a-t-il quelque chose de spécial à faire pour suivre bullseye quand
> > > ça va basculer en stable ?

Bonsoir,

Je préfère attendre quelques semaines avant de basculer
vers la nouvelle version stable, souvent sage attitude :-)

Re: Creating my first LAN

[Brian]

On Tue 30 Mar 2021 at 11:19:48 -0500, Richard Owlett wrote:

[...]

> What well meaning members of this list have not picked up on is that I am
> very literal minded and careful in phrasing my questions. I've said
> somewhere in this thread that my "universe of discourse" is explicitly
> limited to Laptop-1 and Laptop2. I've not been a sophomore B.S.E.E. student
> for over a half-century ;} But I can see a Sophomore lab being set up with
> my specified constraints. Since this thread started I have been reading
> about "ad-hoc networks". That's my goal. Real world has temporarily
> interrupted my pursuit.

You have to bear in mind that other users read these posts and maybe
have similar objectives in mind. Posts veering away from the caefully
phrased question could spark their interest and lead them towards a
solution to an issue.

Personally, I see the idea of an ad-hoc network as interesting, but
haven't the incentive to set one up. Not yet anyway!
 
> More will eventually follow.

Hopefully it will basically be a HOWTO and not a flurry of questions
on how to do it :).

-- 
A well-meaning -user contributor.
> 
> 
> 
> 

[Résolu] Re: suivre bullseye en stable

[Benoit B]

Merci pour vos réponses !

--
Benoit

Le mar. 30 mars 2021 à 19:50, Bruno Volpi  a écrit :
>
> bonsoir,
>
> tu peux te mettre sur la liste de distribution
> debian-annou...@lists.debian.org afin d'être prévenu de la date de
> basculement à stable.
>
>
> Le 30/03/2021 à 19:39, Benoit B a écrit :
> > Bonsoir à toutes et tous,
> >
> > Y a-t-il quelque chose de spécial à faire pour suivre bullseye quand
> > ça va basculer en stable ?
> > Voici mon source.list
> >
> > deb https://deb.debian.org/debian/ bullseye main #contrib non-free
> > deb-src https://deb.debian.org/debian/ bullseye main
> > deb https://deb.debian.org/debian/ bullseye-updates main
> >
> > Je voudrais passer en stable à ce moment là.
> >
> > Merci d'avance.
> >
> > --
> > Benoît
> >
>

Re: suivre bullseye en stable

[Bruno Volpi]

bonsoir,

tu peux te mettre sur la liste de distribution 
debian-annou...@lists.debian.org afin d'être prévenu de la date de 
basculement à stable.



Le 30/03/2021 à 19:39, Benoit B a écrit :

Bonsoir à toutes et tous,

Y a-t-il quelque chose de spécial à faire pour suivre bullseye quand
ça va basculer en stable ?
Voici mon source.list

deb https://deb.debian.org/debian/ bullseye main #contrib non-free
deb-src https://deb.debian.org/debian/ bullseye main
deb https://deb.debian.org/debian/ bullseye-updates main

Je voudrais passer en stable à ce moment là.

Merci d'avance.

--
Benoît

Re: suivre bullseye en stable

[Gaëtan Perrier]

Le mardi 30 mars 2021 à 19:39 +0200, Benoit B a écrit :
> Bonsoir à toutes et tous,
> 
> Y a-t-il quelque chose de spécial à faire pour suivre bullseye quand
> ça va basculer en stable ?
> Voici mon source.list
> 
> deb https://deb.debian.org/debian/ bullseye main #contrib non-free
> deb-src https://deb.debian.org/debian/ bullseye main
> deb https://deb.debian.org/debian/ bullseye-updates main
> 
> Je voudrais passer en stable à ce moment là.
> 
> Merci d'avance.

Bonsoir,

Si tu veux passer en stable quand Bullseye deviendra stable, c'est bon.

Gaëtan


signature.asc
Description: This is a digitally signed message part

suivre bullseye en stable

[Benoit B]

Bonsoir à toutes et tous,

Y a-t-il quelque chose de spécial à faire pour suivre bullseye quand
ça va basculer en stable ?
Voici mon source.list

deb https://deb.debian.org/debian/ bullseye main #contrib non-free
deb-src https://deb.debian.org/debian/ bullseye main
deb https://deb.debian.org/debian/ bullseye-updates main

Je voudrais passer en stable à ce moment là.

Merci d'avance.

--
Benoît

Re: Creating my first LAN

[Andrew M.A. Cater]

On Tue, Mar 30, 2021 at 11:28:07AM -0500, Richard Owlett wrote:
> On 03/30/2021 10:28 AM, Andrew M.A. Cater wrote:
> > Richard,
> > 
> > So: You have two laptops. Both installed from the first DVD. Never connected
> > to the Internet.
> > 
> > Did you actually manage to install the firmware to make the WiFi cards work?
> 
> Yes ;>
> One was old enough that the installer had the correct driver.

Which driver?

How installed? From firmware-linux-free?

> The driver for the other was available from the repository.
> 
_Which_ repository? 

For the one that required firmware:

How did you install it? 

What firmware did you actually install

> > 
> > Outwith that firmware, it is very unlikely that anything will work.
> > [And this goes back to one of the previous email chains].
> 
> Justify that claim.

Most modern cards will require some firmware: the oldest of mine here requres 
firmware-intel-iwlwifi 
for example. If you are lucky, that firmware will be in firmware-linux-free.
If you are less lucky, the driver will be non free and in 
firmware-linux-nonfree / firmware-misc-nonfree 
[Which require Debian's non-free repositories to be enabled and an Internet 
conection / use of the 
unofficial installer which includes firmware.]

If you're very unlucky, you may have to download the firmware from somewhere on 
the Internet, build kernel
modules etc. etc.

> The older unit had no problem connecting to the internet via the local
> library's public WiFi portal. [Haven't explicitly tried the other yet.]
> 

See the questions above: so one of the laptops _has_ connected to the Internet 
- it's not just laptop1 and
laptop2 and no Internet connection ever?

> > 
> > If you fire up one of the laptops, and run " dmesg | less " - do you find
> > references to WiFi hardware (and corresponding firmware being loaded) ?
> >

Have you got not only hardware physically in the laptop but the 
firmware/drivers needed to use it and is the
detection of the WiFi hardware working?
 
> > It doesn't really matter what ad-hoc networking you want to put in place:
> > absent the firmware, almost all Wifi hardware is a brick.

If you don't have the right firmware, WiFi won't work - see, for example, the 
Pine64 ARM machine sat by my
feet which has a strange Realtek wireless driver. The hardware is there, but 
the software will never work to
initialise it - for all intents and purposes, it doesn't exist and the firmware 
from Debian doesn't initialise
it at all. To make it work, I'd have to go and find source code in Github 
somewhere and hand build modules.

> The older (and
> > significantly slower) WiFi hardware where the firmware is contained in free
> > drivers is very rare now.

A couple of very old chipsets - some of them are sold by the FSF as Respects 
Your Hardware chipset, for 
example.. One of them is an old TP-Link - the chipsets may only support 2.4GHz 
/ slower speeds.
> 
> Those two paragraphs do not {so to speak} compute.
> For more detail, see my response to David.
> 
> > 
> > All the very best, as ever,
> > 
> > Andy C.
> > 
> > 
> > 
> 
> 

Re: Creating my first LAN

[mick crane]

On 2021-03-30 17:28, IL Ka wrote:



You might need a x-over RJ45 cable to connect the 2 together.



There is a technology called "Auto MDI-X". With it NIC can detect cable
type hence it can support both cross-cord and patch-cord.
This technology is optional for 10Mb and 100Mb (although widely 
implemented

in 100).
For 1000Mb cross cable is never required.

https://en.wikipedia.org/wiki/Medium-dependent_interface#Auto_MDI-X


I suppose it depends on the age, I recall having a PC needed a x-over 
cable for whatever it was I was doing.

mick

--
Key ID4BFEBB31

Re: Creating my first LAN

[IL Ka]

>
>
> You might need a x-over RJ45 cable to connect the 2 together.
>

There is a technology called "Auto MDI-X". With it NIC can detect cable
type hence it can support both cross-cord and patch-cord.
This technology is optional for 10Mb and 100Mb (although widely implemented
in 100).
For 1000Mb cross cable is never required.

https://en.wikipedia.org/wiki/Medium-dependent_interface#Auto_MDI-X

Re: Creating my first LAN

[Richard Owlett]

On 03/30/2021 10:28 AM, Andrew M.A. Cater wrote:

Richard,

So: You have two laptops. Both installed from the first DVD. Never connected
to the Internet.

Did you actually manage to install the firmware to make the WiFi cards work?


Yes ;>
One was old enough that the installer had the correct driver.
The driver for the other was available from the repository.



Outwith that firmware, it is very unlikely that anything will work.
[And this goes back to one of the previous email chains].


Justify that claim.
The older unit had no problem connecting to the internet via the local 
library's public WiFi portal. [Haven't explicitly tried the other yet.]




If you fire up one of the laptops, and run " dmesg | less " - do you find
references to WiFi hardware (and corresponding firmware being loaded) ?

It doesn't really matter what ad-hoc networking you want to put in place:
absent the firmware, almost all Wifi hardware is a brick. The older (and
significantly slower) WiFi hardware where the firmware is contained in free
drivers is very rare now.


Those two paragraphs do not {so to speak} compute.
For more detail, see my response to David.



All the very best, as ever,

Andy C.

Re: Button events from headphone+micro combo

[tomas]

On Tue, Mar 30, 2021 at 11:18:11AM -0400, Stefan Monnier wrote:
> > I'm sure this can be made in software, by looking at the mic
> > input. Most probably even in user space.
> > Now to find someone with enough time in her/his hands and who
> > can be convinced into trying :-)
> 
> What I find puzzling here is that this is a 100% standard feature on
> pretty much all Android tablet and smartphones.  It's clearly cheap
> to implement.  Why isn't it standard on laptops and (mini) desktops?

Perhaps because the need is more acute when you have the device
somewhere in a shirt pocket, which doesn't happen that often with
laptops and/or desktops.

But don't take my word for it. I'm a well-known Luddite and try to
steer clear of as many gadgets as I can ;-)

Cheers
 - t


signature.asc
Description: Digital signature

Re: Creating my first LAN

[Richard Owlett]

On 03/30/2021 09:11 AM, David Wright wrote:

On Thu 25 Mar 2021 at 17:02:36 (+), Andrew M.A. Cater wrote:


You still have he magic Mifi hotspot from T-Mobile that you were connecting
to via USB. That may well have DHCP and DNS functionality built in - assuming
you have wifi drivers installed on your two machines, it may well "just work"
in assigning addresses, allwoing for communication between machines.


That would involve taking the cover off, removing the battery, then
the SIM, refitting the battery and back cover. That's assuming it
still provides a LAN service when disconnected from its WAN (which
it ought). And then reversing the above whenever you *want* the Internet.


That's what it's _designed_ for :-)


I would be surprised if it didn't have DHCP, as it's a portable device
and would expect to be seeing a lot of new devices. But I have no idea
why they would go to the trouble and expense of providing a DNS server.
How would you use it?

Cheers,
David.


In a way, every thing you say is true. It is unclear how much of the 
intelligence displayed is physically in the T-Mobile supplied physical 
device and how much is in the system that connects the far end of the 
cell channel to the rest of the world. The format of error messages I 
get with a malformed URL leads me to suspect much is at the other end of 
the cell network connection.


What well meaning members of this list have not picked up on is that I 
am very literal minded and careful in phrasing my questions. I've said 
somewhere in this thread that my "universe of discourse" is explicitly 
limited to Laptop-1 and Laptop2. I've not been a sophomore B.S.E.E. 
student for over a half-century ;} But I can see a Sophomore lab being 
set up with my specified constraints. Since this thread started I have 
been reading about "ad-hoc networks". That's my goal. Real world has 
temporarily interrupted my pursuit.


More will eventually follow.

Re: Creating my first LAN

[mick crane]

On 2021-03-24 15:34, Richard Owlett wrote:

This is essentially a reading list request.
I have never administered a LAN and believe in "learning by doing".
I have two laptops with clean installs of Buster. During installation
server software was installed on *ONE* of them. Communication will be
via WiFi. The MATE panel's network connection icon for each laptop
shows the existence of WiFi sources in the neighborhood.

Any specific suggestions for reading?
What internet search terms will lead to productive results?
TIA


I'd get the wired connection to work first.
You might need a x-over RJ45 cable to connect the 2 together.
A x-over cable has the red and green wires swapped over
red > green
green > red
red stripey > green stripey
green stripey > red stripey
If you put them in the same network address range
10.0.0.1, 10.0.0.2
subnetmask 255.255.255.0
you might be able to ping each other then you could get ssh working.

mick
--
Key ID4BFEBB31

Re: Re: Creating my first LAN

[Andrew M.A. Cater]

Richard,

So: You have two laptops. Both installed from the first DVD. Never connected
to the Internet.

Did you actually manage to install the firmware to make the WiFi cards work?

Outwith that firmware, it is very unlikely that anything will work.
[And this goes back to one of the previous email chains].

If you fire up one of the laptops, and run " dmesg | less " - do you find 
references to WiFi hardware (and corresponding firmware being loaded) ?

It doesn't really matter what ad-hoc networking you want to put in place:
absent the firmware, almost all Wifi hardware is a brick. The older (and 
significantly slower) WiFi hardware where the firmware is contained in free
drivers is very rare now.

All the very best, as ever,

Andy C. 

Re: Button events from headphone+micro combo

[Stefan Monnier]

> I'm sure this can be made in software, by looking at the mic
> input. Most probably even in user space.
> Now to find someone with enough time in her/his hands and who
> can be convinced into trying :-)

What I find puzzling here is that this is a 100% standard feature on
pretty much all Android tablet and smartphones.  It's clearly cheap
to implement.  Why isn't it standard on laptops and (mini) desktops?


Stefan

A script for downloading browser files, Re: [?]Are Realtek Audio Drivers for Linux available for use

[David Wright]

On Sat 13 Mar 2021 at 09:26:32 (+0530), Susmita/Rajib wrote:
> On 13/03/2021, David Wright wrote:
> > ... ... [snipped]   ... ... 
> > [snipped]   ... ...
> > I (≠ OP) don't use pulseaudio as a matter of course. It's meant to
> > give benefits, but can add undesired complexity. So I generally
> > stick with ALSA.
> ...   ... [snipped]   ... ... [snipped]   
> ... ...
> 
> It's nice to hear from you, Mr. Wright. I remember your support.
> 
> > Most of the time, this doesn't matter, as I have developed several
> > techniques over the years for capturing Transport Stream files from
> > the cache, and assembling them into seamless live video, and even
> > slurping files out of the browser's /proc//fd thingies when
> > youtube_dl fails to download a video.

I was asked to share this technique of stealing files from the FF
browser's cache, so I've attached the script that I use.

I normally browse sites as a different user (named flash for
historical reasons), so a typical invocation command line looks like:

$ sudo -u flash /home/flash/bin/steal-from-cache.sh 
/home/flash/.cache/mozilla/firefox/jzq80501.default/cache2/entries/ 
/tmp/.0flashcache

The directories firefox/jzq80501.default/ will obviously need
adjustment. In my own case, even */*/ would suffice, as the flash
"user" only runs one browser in one profile.

There are several idiosyncrasies:

 The script sets file permissions to world-writeable so that I can
 steal the swag from user flash.

 The filenames are formatted to be readable but short.

 If you delete the file that stops it (Target/0), it only actually
 stops when the next file triggers.

 The program chatters as it captures files, so that you know whether
 at least your end of the process is working.

 It can sometimes capture and timestamp files in the wrong order
 (particularly when starting), or copy duplicates, so the internal
 times should always be used, where available, for assembling a
 set of files.

Finally, you might not capture much if you don't have this in your
magic file:

# MPEG Transport Stream (MPEG-2 Part 1) and perhaps others

0   string  G
>0xBC   string  G
>>0x178 string  G
>>>0x234string  G
0x2F0   string  G
>0x3AC  string  G
>>0x468 string  G   TS transport stream

Cheers,
David.


steal-from-cache.sh
Description: Bourne shell script

Re: problem with debian Buster and debugging thread in Qt5

[Vincent charente]

Hi Lisandro,

Great, it works with handle SIGSTOP pass nostop noprint. I think I had
already tried but I forgot "pass".

Thank you very much

Le mar. 30 mars 2021 à 15:04, Lisandro Damián Nicanor Pérez Meyer <
perezme...@gmail.com> a écrit :

> Hi again!
>
> On Tue, 30 Mar 2021 at 09:55, Lisandro Damián Nicanor Pérez Meyer
>  wrote:
> >
> [snip]
> >
> > In fact I can reproduce your issue and at the same time I can make it
> > work under gdb itself, so the issue seems to be in Qt Creator. I've
> > reproduced the issue with Qt Creator from unstable, so let me suggest
> > you to file a bug upstream: http://https://bugreports.qt.io Do not
> > forget to upload this test code, it's very straightforward and so a
> > nice example for an upstream issue.
>
> Actually it is not an issue but a feature. Qt Creator is stopping the
> execution when it receives a signal from gdb, in this case SIGSTOP.
> Add a breakpoint inside the thread code and press F5 to start
> debugging. If you accept the message and press F10 (step) some times
> you will be able to reach the code.
>
> You can tell Qt Creator to ignore the signal. Check
>
> https://forum.qt.io/topic/100958/catching-unix-interrupt-signal-on-console-application-when-debugging-with-qtcreator/2
>
> I've added
>
>   handle SIGSTOP pass nostop noprint
>
> And it just worked :-)
>
> --
> Lisandro Damián Nicanor Pérez Meyer
> http://perezmeyer.com.ar/
> http://perezmeyer.blogspot.com/
>

Re: Possible?! A Debian public repository for all complex code lines with examples and scripts?

[George Shuklin]

On 26/03/2021 16:59, Dan Ritter wrote:

George Shuklin wrote:

But it's all software. Debian can't change sh to be 'not sh'. And any
changes in a build stack are touching vast amount of software with extremely
complex use-cases, so it's almost impossible to 'replace'. You can 'add' a
new one, but it just make xkcd #927.

To be completely clear, Debian did change sh from Bourne-ish sh
to dash in Squeeze -- 2006 or so.

Thank you for clarification. My point was that 'sh' approach (including 
bash and dash) is really, really bad. Bash as a programming language is 
super tricky, with lots of landmines and almost no validation whatsoever 
for errors (there is a bit, but way less than expected from any modern 
system language).


I use a lot of Debian packaging in proprietary settings (sigh, I get 
paid for this), and it's love-hate relationship. I love that I can do 
whatever I need to; I hate those implicit magic rules and obscure 
environment variables.

Re: Cannot see update to recent linux kernel 5.10.19-1~bpo10+1 (from 5.10.13-1~bpo10+1)

[David Wright]

On Sun 28 Mar 2021 at 19:46:07 (+0200), l0f...@tuta.io wrote:
> 
> By the way, what is image-5.10.0-0.bpo.4* serie for?Is it only related to 
> "Change ABI number to 0.bpo.4"? Do I need that?

I guess you have to read the changes file to find that out.
https://lists.debian.org/debian-backports-changes/2021/03/msg00053.html
might be of use here.

> 23 mars 2021, 17:45 de deb...@lionunicorn.co.uk:
> 
> > But the OP is, perhaps, relying on the upgrade mechanism provided by
> > the linux-image-amd64 metapackage. However, that's supporting only
> > the signed versions, and buster-backports doesn't (yet) have any
> > 5.10.19 signed versions:
> >
> So you think that's the correct answer (no update because the package is not 
> ready yet)?
> 
> If so, I'll simply wait for it. I just wanted to make sure my conf is OK (and 
> obviously it wasn't configured in such a way to get the latest kernel 
> versions from backports)...

That would depend on there being one.

On Mon 29 Mar 2021 at 01:09:25 (+0200), l0f...@tuta.io wrote:
> 22 mars 2021, 01:12 de wea...@riseup.net:
> 
> > You may be looking at the unsigned version and may have to wait 24 hours
> > for the signed one to come through the repositories, if that's what you
> > are running.
> >
> Thanks for your message but 24 hours don't seem to be enough here...
> 
> Maybe there are some issues experienced by the maintainer?
> Is there an easy way to see that?
> 
> With what everyone says so far, I have the feeling that 5.10.19-1~bpo10+1 is 
> indicated as available via backports but not for every flavours. Everybody is 
> not in the same boat ;)
> 
> I have no problem with that but maybe this subtlety could be stated more 
> explicitely somewhere? Just a suggestion so one doesn't have to dig too much 
> to understand that...

Perhaps it's worth subscribing to debian-backports-changes and
debian-backports. You can post to the latter.

Cheers,
David.

Re: Creating my first LAN

[David Wright]

On Thu 25 Mar 2021 at 17:02:36 (+), Andrew M.A. Cater wrote:
> 
> You still have he magic Mifi hotspot from T-Mobile that you were connecting
> to via USB. That may well have DHCP and DNS functionality built in - assuming
> you have wifi drivers installed on your two machines, it may well "just work"
> in assigning addresses, allwoing for communication between machines.

That would involve taking the cover off, removing the battery, then
the SIM, refitting the battery and back cover. That's assuming it
still provides a LAN service when disconnected from its WAN (which
it ought). And then reversing the above whenever you *want* the Internet.

> That's what it's _designed_ for :-)

I would be surprised if it didn't have DHCP, as it's a portable device
and would expect to be seeing a lot of new devices. But I have no idea
why they would go to the trouble and expense of providing a DNS server.
How would you use it?

Cheers,
David.

Re: Pending kernel upgrade

[didier gaumet]

Le 29/03/2021 à 13:31, David BERCOT a écrit :

Bon, j'ai rebooté et... aucun changement...
Bizarrement, au démarrage, on dirait que la config est "gélée" et ne 
tient pas compte des mises à jour précédentes.


Suis-je le seul dans ce cas ?


Une autre hypothèse: si tu as joué avec plusieurs installations de grub 
(multiboot Linux en laissant chaque distro installer son propre grub, ou 
usage un peu expérimental en spécifiant l'endroit où Grub doit 
s'installer, etc...), peut-être le fichier /boot/grub/grub.cfg n'est-il 
pas celui utilisé par grub et dans ce cas un simple grub-install avec en 
paramètre ton disque (pas une partition) pourrait suffire à faire rentre 
les choses dans l'ordre...

Re: problem with debian Buster and debugging thread in Qt5

[Lisandro Damián Nicanor Pérez Meyer]

Hi again!

On Tue, 30 Mar 2021 at 09:55, Lisandro Damián Nicanor Pérez Meyer
 wrote:
>
[snip]
>
> In fact I can reproduce your issue and at the same time I can make it
> work under gdb itself, so the issue seems to be in Qt Creator. I've
> reproduced the issue with Qt Creator from unstable, so let me suggest
> you to file a bug upstream: http://https://bugreports.qt.io Do not
> forget to upload this test code, it's very straightforward and so a
> nice example for an upstream issue.

Actually it is not an issue but a feature. Qt Creator is stopping the
execution when it receives a signal from gdb, in this case SIGSTOP.
Add a breakpoint inside the thread code and press F5 to start
debugging. If you accept the message and press F10 (step) some times
you will be able to reach the code.

You can tell Qt Creator to ignore the signal. Check
https://forum.qt.io/topic/100958/catching-unix-interrupt-signal-on-console-application-when-debugging-with-qtcreator/2

I've added

  handle SIGSTOP pass nostop noprint

And it just worked :-)

-- 
Lisandro Damián Nicanor Pérez Meyer
http://perezmeyer.com.ar/
http://perezmeyer.blogspot.com/

Re: problem with debian Buster and debugging thread in Qt5

[Lisandro Damián Nicanor Pérez Meyer]

Hi Vincent!

On Tue, 30 Mar 2021 at 05:29, Vincent charente  wrote:
>
> Dear All,
>
>
>
> I am sending you this email because I encounter a problem with Debian Buster 
> and default Qt5 package.
>
> I installed packages below :
>
> ·debian 10.8 x86_64 (amd64)
>
> ·apt-get install build-essentials
>
> ·apt-get install gdb
>
> ·apt-get install qt5-default

This one is really not needed at all, feel free to remove it.

> ·apt-get install qtcreator
>
> I have tried to debug an application which contains a thread with GBD but GDB 
> stops when thread starts. I tried to build GDB with last sources but the 
> problem still remains.
>
> This problem doesn’t exist with Debian 9 with the same procedure of 
> installation.
>
> You can find a simple example which increments a counter in a thread every 
> 500ms at the URL below https://pastebin.com/WegsL9vd.
>
>
> Do you know if there is a way to solve this problem ?

In fact I can reproduce your issue and at the same time I can make it
work under gdb itself, so the issue seems to be in Qt Creator. I've
reproduced the issue with Qt Creator from unstable, so let me suggest
you to file a bug upstream: http://https://bugreports.qt.io Do not
forget to upload this test code, it's very straightforward and so a
nice example for an upstream issue.


-- 
Lisandro Damián Nicanor Pérez Meyer
http://perezmeyer.com.ar/
http://perezmeyer.blogspot.com/

Re: O.T. : instalar alternativa a Google Calc en celular con Android 6

[Debian]

El 30/3/21 a las 00:06, Simeón Ignacio Martirén escribió:
Necesito abandonar las herramientas de Google para trabajarlas desde mi 
celular. LibreOffice no parece estar disponible para dispositivos 
móviles. Gracias desde ya.


--
*/_Ignacio_/*
*/_
_/*
*/_
_/*



¿Quién dijo que no? En Android por lo menos, sí.

Lector:

https://play.google.com/store/apps/details?id=at.tomtasche.reader

Editor / Lector:

https://play.google.com/store/apps/details?id=com.collabora.libreoffice


JAP

Re: MX patito feo 19.3 : trying to apt upgrade

[Thomas Schmitt]

Hi,

ellanios82 wrote:
> : perhaps if i wait a day or 2 , i might get lucky  :)

On the risk to be proven wrong by more experienced bystanders, i assume
that you need to find a mirror with the new package, update your package
list from that mirror, and use it for upgrading.

(My theory is that your local package list does not match the list of the
mirror at ftp.cc.uoc.gr, but rather stems from a server with newer packages.)


Have a nice day :)

Thomas

Re: MX patito feo 19.3 : trying to apt upgrade

[ellanios82]

On 3/30/21 1:15 PM, Thomas Schmitt wrote:

Hi,


E: Failed to fetch
http://ftp.cc.uoc.gr/mirrors/linux/mx/mx/repo/pool/main/m/mx-docs/mx-doc
s_20200817mx19+1_all.deb

This server only has the predecessor package
   mx-docs_20200817mx19_all.deb

It seems that mx-docs_20200817mx19+1_all.deb already has a successor named
   mx-docs_20200817mx19+2.deb
   http://mxrepo.com/mx/repo/pool/main/m/mx-docs/mx-docs_20200817mx19+2_all.deb

Google found me the currently desired package name as

http://mxl.mirror.root.lu/mx/repo/pool/main/m/mx-docs/mx-docs_20200817mx19+1_all.deb

(If every attempt fails to get this working as mirror, i'd download the
package manually and install it by "dpkg -i".
This might be a bad idea, though. I'm very bad as Debian sysadmin.)



Err:5 http://ftp.gr.debian.org/mirrors/linux/mx/mx/repo buster Release
   404  Not Found [IP: 147.102.222.211 80]"
Did i type something wrong, please??

That server does not expose a top level directory named "mirrors".
The directory path for accessing MX repos seems not to be much standardized.


Have a nice day :)

Thomas


- many thanks, Thomas


 : perhaps if i wait a day or 2 , i might get lucky  :)


best,

  Richard

.

Re: MX patito feo 19.3 : trying to apt upgrade

[Thomas Schmitt]

Hi,

> E: Failed to fetch
> http://ftp.cc.uoc.gr/mirrors/linux/mx/mx/repo/pool/main/m/mx-docs/mx-doc
> s_20200817mx19+1_all.deb

This server only has the predecessor package
  mx-docs_20200817mx19_all.deb

It seems that mx-docs_20200817mx19+1_all.deb already has a successor named
  mx-docs_20200817mx19+2.deb
  http://mxrepo.com/mx/repo/pool/main/m/mx-docs/mx-docs_20200817mx19+2_all.deb

Google found me the currently desired package name as
   
http://mxl.mirror.root.lu/mx/repo/pool/main/m/mx-docs/mx-docs_20200817mx19+1_all.deb

(If every attempt fails to get this working as mirror, i'd download the
package manually and install it by "dpkg -i".
This might be a bad idea, though. I'm very bad as Debian sysadmin.)


> Err:5 http://ftp.gr.debian.org/mirrors/linux/mx/mx/repo buster Release
>   404  Not Found [IP: 147.102.222.211 80]"
> Did i type something wrong, please??

That server does not expose a top level directory named "mirrors".
The directory path for accessing MX repos seems not to be much standardized.


Have a nice day :)

Thomas

Re: Button events from headphone+micro combo

[David]

On Tue, 30 Mar 2021 at 19:35,  wrote:

> Anyway. It is a classical "hack" which means taking something you
> have and extending it in some unexpected and usually very creative
> way. Enjoyable, most of the time extremely useful, in a nutshell:
> the engineers core value :-)

> But I stand by "ugly". Which isn't intended to be dismissive.

> P.S: I've no aversion to analog electronics, but I think it's
> rim-full of ugly hacks "we just pretend this transistor has
> no VBE0 and is a linear device" or something :-)

Nooo, please stop digging this hole deeper :)

Those are not hacks (by your definition above).
Rather, they're ugly lies, which if believed will result in tears :)

Simplified models can assist learning, like those
1-dimensional models in the physics class ;p

But there is much more learning required beyond this level, for
anyone who wants to achieve real-world results.

Re: MX patito feo 19.3 : trying to apt upgrade

[ellanios82]

On 3/30/21 11:32 AM, john doe wrote:

On 3/30/2021 10:27 AM, ellanios82 wrote:

Hello List ,

  - since 3 or 4 days, i have not been able to

  "apt upgrade" , and --fix-missing does not solve problem



  this is what am seeing :

"Need to get 50.7 MB/51.0 MB of archives.
After this operation, 3,768 kB disk space will be freed.
Do you want to continue? [Y/n]
Ign:1 http://ftp.cc.uoc.gr/mirrors/linux/mx/mx/repo buster/main amd64
mx-docs all 20200817mx19+1
Err:1 http://ftp.cc.uoc.gr/mirrors/linux/mx/mx/repo buster/main amd64
mx-docs all 20200817mx19+1
   404  Not Found [IP: 2001:648:2c00:6c05::2 80]
E: Failed to fetch
http://ftp.cc.uoc.gr/mirrors/linux/mx/mx/repo/pool/main/m/mx-docs/mx-docs_20200817mx19+1_all.deb 


404  Not Found [IP: 2001:648:2c00:6c05::2 80]
E: Unable to fetch some archives, maybe run apt-get update or try with
--fix-missing?"

...

  - What, please, may i try next, to resolve problem?


  thank you

.




For some reasons, the mirror is not working.
I would try an other mirror.

-- John Doe


- Thanks John : have just now tried 
 But , get this output :

"Ign:4 http://ftp.gr.debian.org/mirrors/linux/mx/mx/repo buster InRelease
Err:5 http://ftp.gr.debian.org/mirrors/linux/mx/mx/repo buster Release
  404  Not Found [IP: 147.102.222.211 80]"

 Did i type something wrong, please??

 thanks
  Richard
.

Re: MX patito feo 19.3 : trying to apt upgrade

[ellanios82]

On 3/30/21 11:32 AM, john doe wrote:

On 3/30/2021 10:27 AM, ellanios82 wrote:

Hello List ,

  - since 3 or 4 days, i have not been able to

  "apt upgrade" , and --fix-missing does not solve problem



  this is what am seeing :

"Need to get 50.7 MB/51.0 MB of archives.
After this operation, 3,768 kB disk space will be freed.
Do you want to continue? [Y/n]
Ign:1 http://ftp.cc.uoc.gr/mirrors/linux/mx/mx/repo buster/main amd64
mx-docs all 20200817mx19+1
Err:1 http://ftp.cc.uoc.gr/mirrors/linux/mx/mx/repo buster/main amd64
mx-docs all 20200817mx19+1
   404  Not Found [IP: 2001:648:2c00:6c05::2 80]
E: Failed to fetch
http://ftp.cc.uoc.gr/mirrors/linux/mx/mx/repo/pool/main/m/mx-docs/mx-docs_20200817mx19+1_all.deb 


404  Not Found [IP: 2001:648:2c00:6c05::2 80]
E: Unable to fetch some archives, maybe run apt-get update or try with
--fix-missing?"

...

  - What, please, may i try next, to resolve problem?


  thank you

.




For some reasons, the mirror is not working.
I would try an other mirror.

--
John Doe


- thanks vm. John :

- myself, stumbling & bumbling in my dark, i did try to change my 
sources.list :-


"# cat /etc/apt/sources.list.d/mx.list
# MX Community Main and Test Repos
# line below changed March 24, 2021
deb http://ftp.cc.uoc.gr/mirrors/linux/mx/mx/repo/ buster main non-free
# deb http://ftp.cc.uoc.gr/mirrors/linux/mx/mx/testrepo/ buster test
# deb http://ftp.auth.gr/mirrors/linux/mx/mx/repo/ buster main non-free
# ahs hardware stack repo
# deb http://ftp.cc.uoc.gr/mirrors/linux/mx/mx/repo/ buster ahs
# line below added March 24, 2021
# deb http://ftp.ntua.gr/mirrors/linux/mx/mx/repo/ buster main non-free
# deb http://ftp.uni-sofia.bg/mirrors/linux/mx/mx/repo/ buster main non-free
# deb http://ftp.gr.debian.org/mirrors/linux/mx/mx/repo/ buster main 
non-free"



 But, unfortunately, just none of my attempts did any good  :(


...

 thanks, Richard

.

Re: Button events from headphone+micro combo

[tomas]

On Mon, Mar 29, 2021 at 09:58:19PM -0400, Stefan Monnier wrote:
> > It's entirely possible that the Librem doesn't have a sound chip
> > that translates impedance changes to button clicks. You could
> > ask the hardware folks, right?
> 
> So, that's what I did:
> 
> https://forums.puri.sm/t/headphone-buttons-on-the-librem-mini/
> 
> so you were right: my chip doesn't support that feature :-(
> Bummer!

I'm sure this can be made in software, by looking at the mic
input. Most probably even in user space.

Now to find someone with enough time in her/his hands and who
can be convinced into trying :-)

Cheers
 - t


signature.asc
Description: Digital signature

Re: Button events from headphone+micro combo

[tomas]

On Tue, Mar 30, 2021 at 10:40:45AM +1100, David wrote:
> On Sat, 27 Mar 2021 at 06:29,  wrote:
> > On Fri, Mar 26, 2021 at 03:05:58PM -0400, Dan Ritter wrote:
> > > Stefan Monnier wrote:
> 
> > > > >> My Librem mini comes with a an audio jack in the front into which I 
> > > > >> can
> > > > >> connect the same headphones-with-micro as used typically on phones.
> 
> > > It's entirely possible that the Librem doesn't have a sound chip
> > > that translates impedance changes to button clicks. You could
> > > ask the hardware folks, right?
> 
> > Ugh. A construction like this [1], where the mic is shorted with
> > varying resistors to transmit signals?
> 
> > Ugly hacks folks come up with, that :-/
> 
> I do not consider this to be ugly, at all. I am a hardware engineer.

I am a physicist, so in some way one OSI layer below you ;-P  [1]

> Given that the headset cable must be flexible, robust, and tiny diameter,
> it would be suboptimal to add dedicated wires for signalling. The reliability
> of both connector and cable would be compromised.

I think the real thing isn't the cable, but the connector and its
compatibility. You only have the two rings for stereo, one for the
mic and a common return. This was already stretching good-ol' jack
too far. Introducing yet another ring would've been killing poor
old audio jack :-)

And killing Jack takes "courage", as Apple's Tim Cook once said.
Which is an euphemism for "we can afford to gouge our users".

Anyway. It is a classical "hack" which means taking something you
have and extending it in some unexpected and usually very creative
way. Enjoyable, most of the time extremely useful, in a nutshell:
the engineers core value :-)

But I stand by "ugly". Which isn't intended to be dismissive.

P.S: I've no aversion to analog electronics, but I think it's
rim-full of ugly hacks "we just pretend this transistor has
no VBE0 and is a linear device" or something :-)

> I don't understand why anyone would think this is ugly, or what would be
> less ugly. If it is an aversion to analog electronics, don't forget
> that "digital"
> electronics is an abstraction. And underneath that, everything is analog.

And then, underneath that, again, digital. In a very twisted
way. Or then, perhaps not... depending on whether you're looking
at it. So physicists tell me ;-D

Cheers

[1] Tongue-in-cheek. I don't take myself too seriously. Seriously!

 - t


signature.asc
Description: Digital signature

Re: MX patito feo 19.3 : trying to apt upgrade

[john doe]

On 3/30/2021 10:27 AM, ellanios82 wrote:

Hello List ,

  - since 3 or 4 days, i have not been able to

  "apt upgrade" , and --fix-missing does not solve problem



  this is what am seeing :

"Need to get 50.7 MB/51.0 MB of archives.
After this operation, 3,768 kB disk space will be freed.
Do you want to continue? [Y/n]
Ign:1 http://ftp.cc.uoc.gr/mirrors/linux/mx/mx/repo buster/main amd64
mx-docs all 20200817mx19+1
Err:1 http://ftp.cc.uoc.gr/mirrors/linux/mx/mx/repo buster/main amd64
mx-docs all 20200817mx19+1
   404  Not Found [IP: 2001:648:2c00:6c05::2 80]
E: Failed to fetch
http://ftp.cc.uoc.gr/mirrors/linux/mx/mx/repo/pool/main/m/mx-docs/mx-docs_20200817mx19+1_all.deb
404  Not Found [IP: 2001:648:2c00:6c05::2 80]
E: Unable to fetch some archives, maybe run apt-get update or try with
--fix-missing?"

...

  - What, please, may i try next, to resolve problem?


  thank you

.




For some reasons, the mirror is not working.
I would try an other mirror.

--
John Doe

problem with debian Buster and debugging thread in Qt5

[Vincent charente]

Dear All,



I am sending you this email because I encounter a problem with Debian
Buster and default Qt5 package.

I installed packages below :

·debian 10.8 x86_64 (amd64)

·apt-get install build-essentials

·apt-get install gdb

·apt-get install qt5-default

·apt-get install qtcreator



I have tried to debug an application which contains a thread with GBD but
GDB stops when thread starts. I tried to build GDB with last sources but
the problem still remains.

*This problem doesn’t exist with Debian 9 *with the same procedure of
installation.

You can find a simple example which increments a counter in a thread every
500ms at the URL below https://pastebin.com/WegsL9vd.



Do you know if there is a way to solve this problem ?



Best regards

MX patito feo 19.3 : trying to apt upgrade

[ellanios82]

Hello List ,

 - since 3 or 4 days, i have not been able to

 "apt upgrade" , and --fix-missing does not solve problem



 this is what am seeing :

"Need to get 50.7 MB/51.0 MB of archives.
After this operation, 3,768 kB disk space will be freed.
Do you want to continue? [Y/n]
Ign:1 http://ftp.cc.uoc.gr/mirrors/linux/mx/mx/repo buster/main amd64 
mx-docs all 20200817mx19+1
Err:1 http://ftp.cc.uoc.gr/mirrors/linux/mx/mx/repo buster/main amd64 
mx-docs all 20200817mx19+1

  404  Not Found [IP: 2001:648:2c00:6c05::2 80]
E: Failed to fetch 
http://ftp.cc.uoc.gr/mirrors/linux/mx/mx/repo/pool/main/m/mx-docs/mx-docs_20200817mx19+1_all.deb 
404  Not Found [IP: 2001:648:2c00:6c05::2 80]
E: Unable to fetch some archives, maybe run apt-get update or try with 
--fix-missing?"


...

 - What, please, may i try next, to resolve problem?


 thank you

.

Re: O.T. : instalar alternativa a Google Calc en celular con Android 6

[Camaleón]

El 2021-03-30 a las 00:06 -0300, Simeón Ignacio Martirén escribió:

> Necesito abandonar las herramientas de Google para trabajarlas desde mi
> celular. LibreOffice no parece estar disponible para dispositivos móviles.
> Gracias desde ya.

Listado de alternativas, por si te sirve:

Google Docs / Sheets / Slides alternative
https://restoreprivacy.com/google-alternatives/

Saludos,

-- 
Camaleón 

Re: Cannot see update to recent linux kernel 5.10.19-1~bpo10+1 (from 5.10.13-1~bpo10+1)

[Andrei POPESCU]

On Ma, 30 mar 21, 00:06:26, l0f...@tuta.io wrote:
> Hi Andrei,
> 
> 29 mars 2021, 22:09 de andreimpope...@gmail.com:
> 
> > Sorry, quoted wrong line, I meant this:
> >
> >> linux-image-5.10.0-0.bpo.4-amd64-unsigned/buster-backports 
> >> 5.10.19-1~bpo10+1 amd64
> >>
> Haha, I was wondering why you were comparing 5.10 with 5.9^^
> 
> Actually, I maintain my position. The package you are quoting (lastly) is 
> maybe more recent but *unsigned*.
> However, I need the signed one because I use Secure Boot... ;)

$ rmadison linux-signed-amd64
linux-signed-amd64 | 4.19.118+2+deb10u1~bpo9+1 | stretch-backports | source
linux-signed-amd64 | 4.19.171+2| stable| source
linux-signed-amd64 | 4.19.181+1| stable| source
linux-signed-amd64 | 5.9.15+1~bpo10+1  | buster-backports  | source
linux-signed-amd64 | 5.10.13+1~bpo10+1 | buster-backports  | source
linux-signed-amd64 | 5.10.19+1~bpo10+1 | backports-policy  | source
linux-signed-amd64 | 5.10.24+1 | testing   | source
linux-signed-amd64 | 5.10.24+1 | unstable  | source


The packages appears to be stuck in the backports-policy. If I recall 
correctly (see recent -backports archives) this is related to the 
freeze.

Kind regards,
Andrei
-- 
http://wiki.debian.org/FAQsFromDebianUser


signature.asc
Description: PGP signature