Re: Does the debian kernel sends the gratuitous arp ?

[Balaji G]

Hi Groeten,
The IP address is already assigned to this interface
device eno5np0.
Please find my reply mail below.


Thanks,
Balaji




On Mon, 9 Oct 2023 at 20:39, Balaji G  wrote:

> Hi Geert,
>
>  >>> During `ip link set down dev eno5np0` and `ip link set up dev eno5np0`
>
> > are NO IP-addresses involved.
>
> This interface eno5np0 is configured with the ip addresses(10.45.10.4 ).
>
> # ifconfig
>
> eno5np0: flags=4163  mtu 1500
> inet 10.45.10.4  netmask 255.255.255.0  broadcast 10.45.10.255
> ether 88:e9:a4:49:18:48  txqueuelen 1000  (Ethernet)
> RX packets 2967346895  bytes 4417546189260 (4.0 TiB)
> RX errors 0  dropped 1  overruns 0  frame 0
> TX packets 1634521866  bytes 2220475762246 (2.0 TiB)
> TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
>
>
>
> Thanks,
>
> Balaji
>
>
>
> On Sun, 8 Oct 2023 at 20:58, Jeffrey Walton  wrote:
>
>> On Sun, Oct 8, 2023 at 4:58 AM Geert Stappers 
>> wrote:
>> >
>> > On Sun, Oct 08, 2023 at 11:21:10AM +0530, Balaji G wrote:
>> > > On Sat, 7 Oct 2023 at 02:50, Jeffrey Walton wrote:
>> > > > On Fri, Oct 6, 2023 at 2:04 PM Balaji G wrote:
>> > > > >
>> > > > > Hi,
>> > > > >
>> > > > > I am using "Debian GNU/Linux 11 (bullseye)" with kernel version
>> 5.16.12.
>> > > > > When i do a link up/down i don't see any Gratuitous ARP being
>> sent.
>> > > > >
>> > > > > # echo 1 > /proc/sys/net/ipv4/conf/eno5np0/arp_notify
>> > > > > # ip link set down dev eno5np0
>> > > > > # ip link set up dev eno5np0
>> > > > >
>> > > > > Captured all the packets via tcpdump & the tcpdump is not showing
>> any
>> > > > > Gratuitous ARP packets.
>> > > > >
>> > > > > But, with the same commands i could see the Gratuitous ARP being
>> sent in
>> > > > > Red hat.9.0 (Plow).
>> > > > >
>> > > > > So, please let me know if this is a specific scenario in Debian
>> 11 ??
>> > > >
>> > > > I think that's now Poettering:
>> > > >
>> > > >
>> https://github.com/systemd/systemd/blob/main/src/libsystemd-network/sd-ipv4acd.c#L302
>> > > >
>> > >
>> > > Do you mean this is a known issue & will be fixed in the future
>> releases ?
>> >
>> > Here not Jeff.  What I do understand from Jeff message is
>> > something like:  "I can bitch on systemd, so I do"
>> > Time will tell how much Jeff contributes in solving
>> > the interesting problem of the invisible gratuitous ARP.
>>
>> Nope, not that.
>>
>> I learned a long time ago to answer the question that was asked in the
>> body of the post.
>>
>> You're welcome to complain about Systemd all you like. You won't get
>> any complaints from me. It's not a discussion I will engage in.
>>
>> Jeff
>>
>

Re: Does the debian kernel sends the gratuitous arp ?

[Geert Stappers]

On Mon, Oct 09, 2023 at 08:39:02PM +0530, Balaji G wrote:
> Hi Geert,

Hello mailinglist (archive )readers of debian-user@lists.debian.org


> >>> During `ip link set down dev eno5np0` and `ip link set up dev eno5np0`
> >>> are NO IP-addresses involved.
>
> This interface eno5np0 is configured with the ip addresses(10.45.10.4 ).
>
> # ifconfig
}( mismatch between `ifconfig` command expected `ifconfig` output )
> eno5np0: flags=4163  mtu 1500
> inet 10.45.10.4  netmask 255.255.255.0  broadcast 10.45.10.255
> ether 88:e9:a4:49:18:48  txqueuelen 1000  (Ethernet)
> RX packets 2967346895  bytes 4417546189260 (4.0 TiB)
> RX errors 0  dropped 1  overruns 0  frame 0
> TX packets 1634521866  bytes 2220475762246 (2.0 TiB)
> TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0


I did missed  "LOWER_UP", then I noticed that `ifconfig` is used. More
on that later.  Back to 'Subject: Re: Does the debian kernel sends the
gratuitous arp ?'

To me is the answer "Yes, Debian kernel sends gratuitous ARP when it
supposed to do".  But we, this mailinglist, have this discussion because
somewhere is the gratuitous ARP missing.  What triggered the "I wonder
why there is no gratuitous ARP?" is yet unknown.

Let me try to capture a gratuitous ARP.  I will be using only one
interface  (it is OK if the computer has many other network interfaces)
Using two logins as root, one for "tcpdump", the other "set address".

In "tcpdump session"

  tcpdump -n -i eth0  arp


In "set address session"

  sudo ip address add 198.51.100.23/24 dev eth0


Back to "tcpdump session".

And to my surprise:   **no** gratuitous ARP !

It should be there.  I even tried after:

$ cat /proc/sys/net/ipv4/conf/eth0/arp_notify
0
$ echo 1 | sudo tee -a /proc/sys/net/ipv4/conf/eth0/arp_notify
1
$ cat /proc/sys/net/ipv4/conf/eth0/arp_notify
1
$


I'm gonna sleep about.  Expect me back over 48 hours.



> Thanks,

Yeah, thanks for the good question.


> Balaji



Groeten
Geert Stappers


About ifconfig :
  
https://serverfault.com/questions/633087/where-is-the-statement-of-deprecation-of-ifconfig-on-linux

--
Silence is hard to parse

Re: debian.org - broken Download link.

[David Christensen]

On 10/9/23 05:07, Dmitry wrote:

Hi, Brad.

The issue with a broken download link was in browser cache. It preserved 
link

to previous 12.1 version in html. After force update by F5 the issue was
resolved.

P.S. It is so uncommon when robust technologies with low resource 
consumption

are used, with SinglePageApplications no need to press F5, full data set
downloaded per each request.



There are two hard problems in computer science:

1.  What to name things.

2.  When to clear caches.

3.  Off-by-one errors.


See also:

* "The Mess We're In" by Joe Armstrong 
https://www.youtube.com/watch?v=lKXe3HUG2l4



David

Re: Single-page application [was: debian.org - broken Download link.]

[tomas]

On Mon, Oct 09, 2023 at 09:56:43PM +0700, Dima Estudiante wrote:
> > Tastes can be so different :)
> 
> Looks like our tastes quite the same.
> - Plain HTML with Caching is a robust but now days rare used.
> - SPA widely used, but with high resource consumption.

Glad I'm not alone :)

Cheers
-- 
t


signature.asc
Description: PGP signature

Re: Single-page application [was: debian.org - broken Download link.]

[tomas]

On Mon, Oct 09, 2023 at 07:39:44AM -0700, Mike Castle wrote:
> On Mon, Oct 9, 2023 at 6:11 AM  wrote:
> 
> > Gah, no. As a user I hate those with all my guts. Page "state" is
> > distributed in some intransparent way across client and server and
> > there is no way to refer to "something" via an URL.
> 
> Many modern SPAs track state via URL, so they can be referenced.  And
> not just as a ?query or #fragment either.

Sometimes they get it right, yes. Most of the time not.

The problem with web "applications" is that every framework seems
intent in reinventing every wheel, in slightly different ways. Web
"programmers" using those frameworks use them in, again, slightly
different ways. The upshot:

 - slightly different UI "languages" for every application
   (e.g. copy-paste works sometimes, sometimes it doesn't,
   yadda, yadda)
 - at the time the fattest bugs are shaken out, the framework
   dies and is replaced by an even chee^H^H^H^H cooler one.

I've been using my $COMPANY's webmail (a single-pager) for a while.
You can change to calender. If you do a reload, you're thrown back
into your inbox.

People put up with this instead of screaming at their provider.

(I discovered there's an IMAP. Thankfully).

Cheers
-- 
t


signature.asc
Description: PGP signature

Re: Intermittent WiFi on Network Manager

[tomas]

On Mon, Oct 09, 2023 at 03:06:14PM +, Ottavio Caruso wrote:

[...]

> So, we are all running the same frequencies on mobile phones. We all use
> mobile phones, so we should be all cancelling each other out? Something
> doesn't add up.

Perhaps you should read up on how mobile phones work.

> T H E R E   A R E   N O   M I C R O W A W EO V E N SH E R E !!!

If there's no background noise... why are you shouting?

;-)

Cheers
-- 
t


signature.asc
Description: PGP signature

Re: Does the debian kernel sends the gratuitous arp ?

[Balaji G]

Hi Geert,

 >>> During `ip link set down dev eno5np0` and `ip link set up dev eno5np0`

> are NO IP-addresses involved.

This interface eno5np0 is configured with the ip addresses(10.45.10.4 ).

# ifconfig

eno5np0: flags=4163  mtu 1500
inet 10.45.10.4  netmask 255.255.255.0  broadcast 10.45.10.255
ether 88:e9:a4:49:18:48  txqueuelen 1000  (Ethernet)
RX packets 2967346895  bytes 4417546189260 (4.0 TiB)
RX errors 0  dropped 1  overruns 0  frame 0
TX packets 1634521866  bytes 2220475762246 (2.0 TiB)
TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0



Thanks,

Balaji



On Sun, 8 Oct 2023 at 20:58, Jeffrey Walton  wrote:

> On Sun, Oct 8, 2023 at 4:58 AM Geert Stappers  wrote:
> >
> > On Sun, Oct 08, 2023 at 11:21:10AM +0530, Balaji G wrote:
> > > On Sat, 7 Oct 2023 at 02:50, Jeffrey Walton wrote:
> > > > On Fri, Oct 6, 2023 at 2:04 PM Balaji G wrote:
> > > > >
> > > > > Hi,
> > > > >
> > > > > I am using "Debian GNU/Linux 11 (bullseye)" with kernel version
> 5.16.12.
> > > > > When i do a link up/down i don't see any Gratuitous ARP being sent.
> > > > >
> > > > > # echo 1 > /proc/sys/net/ipv4/conf/eno5np0/arp_notify
> > > > > # ip link set down dev eno5np0
> > > > > # ip link set up dev eno5np0
> > > > >
> > > > > Captured all the packets via tcpdump & the tcpdump is not showing
> any
> > > > > Gratuitous ARP packets.
> > > > >
> > > > > But, with the same commands i could see the Gratuitous ARP being
> sent in
> > > > > Red hat.9.0 (Plow).
> > > > >
> > > > > So, please let me know if this is a specific scenario in Debian 11
> ??
> > > >
> > > > I think that's now Poettering:
> > > >
> > > >
> https://github.com/systemd/systemd/blob/main/src/libsystemd-network/sd-ipv4acd.c#L302
> > > >
> > >
> > > Do you mean this is a known issue & will be fixed in the future
> releases ?
> >
> > Here not Jeff.  What I do understand from Jeff message is
> > something like:  "I can bitch on systemd, so I do"
> > Time will tell how much Jeff contributes in solving
> > the interesting problem of the invisible gratuitous ARP.
>
> Nope, not that.
>
> I learned a long time ago to answer the question that was asked in the
> body of the post.
>
> You're welcome to complain about Systemd all you like. You won't get
> any complaints from me. It's not a discussion I will engage in.
>
> Jeff
>

Re: Single-page application [was: debian.org - broken Download link.]

[Dima Estudiante]

> Tastes can be so different :)

Looks like our tastes quite the same.
- Plain HTML with Caching is a robust but now days rare used.
- SPA widely used, but with high resource consumption.

Re: Single-page application [was: debian.org - broken Download link.]

[Mike Castle]

On Mon, Oct 9, 2023 at 6:11 AM  wrote:

> Gah, no. As a user I hate those with all my guts. Page "state" is
> distributed in some intransparent way across client and server and
> there is no way to refer to "something" via an URL.

Many modern SPAs track state via URL, so they can be referenced.  And
not just as a ?query or #fragment either.

mrc

Re: Intermittent WiFi on Network Manager

[gene heskett]

On 10/9/23 10:08, Lee wrote:

On 10/8/23, gene heskett  wrote:

On 10/8/23 07:43, Lee wrote:

On 10/7/23, Ottavio Caruso  wrote:

Am 07/10/2023 um 11:11 schrieb gene heskett:

Another possibility is a leaky microwave oven in the vicinity


This is an urban legend and an excuse I was using when I was in tech
support.


It's real.  Try it yourself - run iperf for 2 minutes, display the
bandwidth report every second and then start the microwave for 1
minute.

I get the thruput cut in half or or more when the microwave is on.
Which is an improvement on the previous microwave which used to kill a
wireless connection. (which was super annoying when the wife was doing
work-from-home & I wasn't allowed to use the microwave _at_all_ during
the day.  I suspect that's the reason she got a toaster oven)

Is it fairly well-known that microwave ovens interfere the most on channel
11?
I just tried linssid again and there's a bunch of APs on channel 1 &
6, one on channel 2 and two on channel 8.  Nothing on channel 11.


That, again probably, would be because the microwave does NOT transmit
an SID,


It doesn't transmit anything resembling a wifi frame (packet?), it's
just noise as far as the wifi interface knows.. and not something that
shows up on a wifi analyzer like linssid.

You need a spectrum analyzer to see wifi noise/interference.  I just
took a quick look again for an affordable spectrum analyzer & didn't
see anything.  Then again, my definition of "affordable" is under $50
so I suppose that's not to surprising.

True, a decent spec analyzer starts with the Siglent offering at a 
little over $3G's, so your $50 is missing some zero's.  Really broadband 
that can see beyond 3 gigahertz costs real money and will spend time in 
the cal lab annually because their front end is that fragile, some can 
be destroyed by a 10 milli-watt input signal.

Regards
Lee
.


Cheers, Gene Heskett.
--
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author, 1940)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 

Re: Intermittent WiFi on Network Manager

[Lee]

On 10/9/23, Ottavio Caruso  wrote:
> Am 08/10/2023 um 11:42 schrieb Lee:
>> On 10/7/23, Ottavio Caruso  wrote:
>>> Am 07/10/2023 um 11:11 schrieb gene heskett:
 Another possibility is a leaky microwave oven in the vicinity
>>>
>>> This is an urban legend and an excuse I was using when I was in tech
>>> support.
>>
>> It's real.  Try it yourself - run iperf for 2 minutes, display the
>> bandwidth report every second and then start the microwave for 1
>> minute.
>>
>> I get the thruput cut in half or or more when the microwave is on.
>> Which is an improvement on the previous microwave which used to kill a
>> wireless connection. (which was super annoying when the wife was doing
>> work-from-home & I wasn't allowed to use the microwave _at_all_ during
>> the day.  I suspect that's the reason she got a toaster oven)
>>
>> Is it fairly well-known that microwave ovens interfere the most on channel
>> 11?
>> I just tried linssid again and there's a bunch of APs on channel 1 &
>> 6, one on channel 2 and two on channel 8.  Nothing on channel 11.
>>
>> Lee
>>
>>
>
> So the microwave should be running 100% 24/7? What is it? Am I
> surrounded by 24/7 greasy spoons? I'm more inclined to believe in a
> buggy driver implementation. All the nearby Windows laptops run fine.

In other words, you didn't try running iperf and then starting the
microwave, right?
Or you did and don't want to admit that your microwave interferes with wifi.

Either way, take a look at
  
https://www.acrylicwifi.com/en/blog/performing-wifi-spectrum-analysis-information-provided/#How_to_Perform_a_Professional_Site_Survey

scroll down just a bit and see
The most common devices that create interference and noise in a
wireless infrastructure are:

Some of those do run 24/7.  And finally
  
https://www.zdnet.com/article/usb-3-and-usb-c-devices-can-cause-problems-with-wi-fi-and-bluetooth-connections-but-theres-a-solution/
which I've never seen in action, just read about.

Lee

Re: Intermittent WiFi on Network Manager

[gene heskett]

On 10/9/23 07:47, Eduardo M KALINOWSKI wrote:

On 07/10/2023 08:11, gene heskett wrote:
Another possibility is a leaky microwave oven in the vicinity. As 


Also check the color of the microwave: if it's bright magenta, change it 
to a black one.


No, wait, that's SATA cables. Never mind.


Actually Eduardo it can be any cable. First observed in the 1970's at 
the height of the CB radio craze, and the then Japanese dominant radios 
used that color of wire for the microphones push to talk switch. I was 
benching at Norfolk 2-Way Radio, a 2nd afternoon job while careing for 
and feeding a uhf transmitter for the Nebraska ETV Commission and about 
the time we were celebrating our 200th anniversary as a nation, I had 
around 75 radios on the back shelf of the service room, waiting on 
microphone cable replacements I couldn't get cuz by then nearly everyone 
in the cable making business had offshored the making to the J.A.Pan co.


That was the only cable we could get thru our supply chaanels, and its 
life in a large car was much less than a year. Belden finally came to 
our rescue by starting up a line to make coil cord versions, but they 
thought we were roping cattle with them so it took a 40 lb pull on a 2 
foot cable to make it 3 feet. Complaints to the J.A.Pan src's must have 
been read by employee's who spoke or read no English as it took several 
years to get that fixed.


Then the same plastic die started showing up in computers at about the 
same time sata became the interface std.  As the outer jacket color cuz 
it was purty. The actual wire inside may have a different color jacket 
inside the magenta sleeve, but that only prolongs the failure to 3 to 5 
years time frame.


The initial failure shows up in the logs as drive resets.  So if the log 
blows up when one of those colored cables is touched by a pencil, its 
gone, but for a long term fix, use any other color.  It will last 
longer.  A lot longer.


Folks laugh at me, and I guess I'm an urban legend. I'll soon be gone as 
I turned 89 last week. I've done pretty good on an 8th grade education, 
but I tested at IQ of 147 in the 7nth grade. quit school and went to 
work fixing them new things called tv's in 1948.  Got a 1st phone in 
1962 and a job at the local tv station. Saw a notice in the local fish 
wrap that the local Community College was testing for Certified 
Electronics Technician certificates in 1972, walked into the classroom 
of the prof teaching the class the next morning and put my 20 dollar 
bill on his desk. Had 4 hours to do it. I was down with that years flu 
so I spent some time in the john, but finished the test in 45 minutes. 
Raised his eyebrows a long ways when he laid the answer stencil on my 
test papers and saw a sea of black. He had been teaching that class for 
several years. I, a total stranger with a $20 bill for his time, was the 
first to pass that test. Not a single one of his students had passed it. 
That certificate, has gotten me every job I've asked for since, its 
testimony that I do know what the hell I'm doing. Yes, I have made 
mistakes, and I'm honest about it when I do, but this isn't one of them. 
 There is one CET for every hundred EE's who can't pass it.  They 
haven't gotten their hands dirty enough. I've been there, and done that, 
got my hands dirty and learned by doing.


Cheers, Gene Heskett.
--
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author, 1940)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 

Re: Intermittent WiFi on Network Manager

[Lee]

On 10/8/23, gene heskett  wrote:
> On 10/8/23 07:43, Lee wrote:
>> On 10/7/23, Ottavio Caruso  wrote:
>>> Am 07/10/2023 um 11:11 schrieb gene heskett:
 Another possibility is a leaky microwave oven in the vicinity
>>>
>>> This is an urban legend and an excuse I was using when I was in tech
>>> support.
>>
>> It's real.  Try it yourself - run iperf for 2 minutes, display the
>> bandwidth report every second and then start the microwave for 1
>> minute.
>>
>> I get the thruput cut in half or or more when the microwave is on.
>> Which is an improvement on the previous microwave which used to kill a
>> wireless connection. (which was super annoying when the wife was doing
>> work-from-home & I wasn't allowed to use the microwave _at_all_ during
>> the day.  I suspect that's the reason she got a toaster oven)
>>
>> Is it fairly well-known that microwave ovens interfere the most on channel
>> 11?
>> I just tried linssid again and there's a bunch of APs on channel 1 &
>> 6, one on channel 2 and two on channel 8.  Nothing on channel 11.
>
> That, again probably, would be because the microwave does NOT transmit
> an SID,

It doesn't transmit anything resembling a wifi frame (packet?), it's
just noise as far as the wifi interface knows.. and not something that
shows up on a wifi analyzer like linssid.

You need a spectrum analyzer to see wifi noise/interference.  I just
took a quick look again for an affordable spectrum analyzer & didn't
see anything.  Then again, my definition of "affordable" is under $50
so I suppose that's not to surprising.

Regards
Lee

Single-page application [was: debian.org - broken Download link.]

[tomas]

On Mon, Oct 09, 2023 at 07:07:09PM +0700, Dmitry wrote:

[...]

> P.S. It is so uncommon when robust technologies with low resource consumption
> are used, with SinglePageApplications no need to press F5, full data set
> downloaded per each request.

Gah, no. As a user I hate those with all my guts. Page "state" is
distributed in some intransparent way across client and server and
there is no way to refer to "something" via an URL.

Tastes can be so different :)

Cheers
-- 
t


signature.asc
Description: PGP signature

Re: debian.org - broken Download link.

[Dmitry]

Hi, Brad.

The issue with a broken download link was in browser cache. It preserved link
to previous 12.1 version in html. After force update by F5 the issue was
resolved.

P.S. It is so uncommon when robust technologies with low resource consumption
are used, with SinglePageApplications no need to press F5, full data set
downloaded per each request.

Re: Understanding package dependencies

[Max Nikulin]

On 08/10/2023 01:45, Greg Wooledge wrote:

On Sat, Oct 07, 2023 at 08:27:11PM +0200, Sven Joachim wrote:

In the current case, "aptitude search '~Plsb-base'" does the trick.

Why on EARTH was this not ported to apt-patterns(7)?


It is one of two features I miss in "apt list".

Another one is --showformat similar to dpkg-query. "apt" is not for 
scripting, but sometimes I do not mind to add package descriptions.

Re: Intermittent WiFi on Network Manager

[Eduardo M KALINOWSKI]

On 07/10/2023 08:11, gene heskett wrote:
Another possibility is a leaky microwave oven in the vicinity. As 


Also check the color of the microwave: if it's bright magenta, change it 
to a black one.


No, wait, that's SATA cables. Never mind.

--
Wedding rings are the world's smallest handcuffs.

Eduardo M KALINOWSKI
edua...@kalinowski.com.br

Re: [OT] Mutt y Gmail

[alfon]

> El 28/9/23 a las 20:20, Camaleón escribió:
>
> > Desde hace ¿años¹? Gmail ya no permite usar la contraseña que usas
> > cuando accedes desde el webmail o aplicaciones que admitan sistemas de
> > autentificación avanzados (OAuth2).
>
> parece que desde mayo del 2022 no funciona.
>

Yo actualmente conecto a gmail dede Mutt usando OAuth, y me funciona
sin problemas.

http://www.mutt.org/doc/manual/#oauth