Re: Telecommunications network drafting software

[Måns Nilsson]

Subject: Re: Telecommunications network drafting software Date: Wed, Sep 01, 
2021 at 03:26:08PM -0400 Quoting Eric Kuhnke (eric.kuh...@gmail.com):
> For logical diagrams of networks, on MacOS, I recommend Omnigraffle.

OmniGraffle is what Visio would be if Visio was cool, looked good and
didn't hate its users. Only drawback -- to some -- is that it's OS X only.

-- 
Måns Nilsson primary/secondary/besserwisser/machina
MN-1334-RIPE   SA0XLR+46 705 989668
... I think I'd better go back to my DESK and toy with a few common
MISAPPREHENSIONS ...


signature.asc
Description: PGP signature

Re: Reminder: Never connect a generator to home wiring without transfer switch

[Forrest Christian (List Account)]

Let me clarify since this thread has resurrected itself.

In the northern climates where I live, almost 100% of the heat during
winter is either natural gas or propane.   It's either fan forced or hot
water.

In each case,  the amount of electricity consumed by a typical furnace is
well under 15 amps.  Like a few hundred watts for all but the biggest
furnaces.

When the power fails during the coldest part of the year the most critical
thing to keep running is the furnace.  You  can have frozen pipes and other
cold related damage in a relatively short time.   The amount of time is
irrelevant but think multiple hours not days.

We all have flashlights,  the fridge isn't a big issue (natural freezer
outdoors, and usually free ice) and so on.But the furnace having power
is critical.

The quickest fastest way to get that furnace back on is to fire up a
generator or some other suitable power source and then hook the generator
up to it.

The problem is that people rarely think ahead,  so they're trying to come
up with a quick solution,  and most of them don't really understand
wiring.   Because,  by code,  the furnace may not be connected with a plug
and socket the homeowner has no option but to open up an electrical box and
try to figure out how to hook his generator up.   With the not uncommon
enough result of the generator trying to power the neighborhood.

The solution is dirt cheap.  Instead of requiring a hardwired connection,
move to a standard 120V 15A plug and socket connection.   Based on other
requirements in the code such as for a disconnecting means near the furnace
which this would replace as well,  the cost is likely to be zero.

Its not uncommon to see this done even though it is against code.  Whether
it will pass inspection depends on the attitude of the inspector.


On Wed, Sep 1, 2021, 10:13 PM Peter Beckman  wrote:

> On Tue, 31 Aug 2021, Forrest Christian (List Account) wrote:
>
> > I just wish the electrical code would permit or require certain low cost
> > things which make temporary generator connections more likely to be safe.
> >
> > For example, code requires most furnaces to be hardwired.  But a furnace
> is
> > one of the first things you want on a generator in an extended winter
> power
> > outage.   If instead of hardwired, the code required plug and socket
> > connections at each 120v furnace  then Joe homeowner would be more likely
> > to run an extension cord from his generator to his furnace instead of
> > trying to rig up his generator with a suicide cord.
>
>   Is $40-60 low cost enough for you for safe, temporary generator
> connections?
>
>  - Generator Interlock Kit: $20-25 (Safety)
>  - Breaker: $5 (30amp 120v) to $20 (60amp 240v) (Dedicated Power
> connection)
>  - Generator Power Inlet Input: $15 (indoor 120v) to $50 (outdoor 240v)
>
>  A Generator Interlock Kit is a few pieces of metal that, once
>  installed on your existing electrical panel, allows one to run
>  a properly-sized circuit and breaker to an outlet that you can plug
> your
>  120v or 240v generator inverter RV output into.
>
>  Add a Generator Power Inlet Input (indoor or outdoor) rated at 30Amp
>  240v NEMA L6-30P, for example, then plug your generator into that.
>
>  The Generator Interlock Kit physically prevents the mains from being
> on
>  when the generator Breaker is on. This is the safety component.
>
>  This seems affordable ($60 plus some wire and a few minutes inside
> your
>  electrical panel) and safe.
>
>  Add a few bucks to have your locality inspect and certify the work.
>
>   If this is too much, why? What would be easier while also being equally
> as
>   safe? This is work that, with a few minutes on YouTube, could do safely,
>   as long as the power is disconnected at the meter outside the home during
>   installation.
>
>   PS - I suppose you could also move all of your emergency 120v stuff to
> one
>   side of your panel and also provide only 120V to one side of your panel.
>   This would also reduce costs a bit.
>
>   Why believe me? In 2019 I read the NEC code and learned how to install a
>   60amp circuit for an electric charger. I did the work myself. I had it
>   inspected and certified by the county. I did so for about $100 total for
>   all parts and wire.
>
> Beckman
> ---
> Peter Beckman  Internet Guy
> beck...@angryox.com
> http://www.angryox.com/
> ---
>

Re: Reminder: Never connect a generator to home wiring without transfer switch

[Jay Hennigan]

On 9/1/21 21:13, Peter Beckman wrote:

On Tue, 31 Aug 2021, Forrest Christian (List Account) wrote:


  Is $40-60 low cost enough for you for safe, temporary generator 
connections?


[snip]


     Add a Generator Power Inlet Input (indoor or outdoor) rated at 30Amp
     240v NEMA L6-30P, for example, then plug your generator into that.


You'll want a neutral. L14-30P.

--
Jay Hennigan - j...@west.net
Network Engineering - CCIE #7880
503 897-8550 - WB6RDV

Re: Reminder: Never connect a generator to home wiring without transfer switch

[Mark Tinka]

On 9/2/21 02:16, Eric Germann via NANOG wrote:

15kW is 15kVA (not 1.5 kVA) at a power factor of 1.0, if the heat is 
all resistive.


Right! Even at a 0.8pf, 15kW is not 1.5kVA.

I just didn't have the energy to get into it with him.

Mark.

Re: Reminder: Never connect a generator to home wiring without transfer switch

[Peter Beckman]

On Tue, 31 Aug 2021, Forrest Christian (List Account) wrote:


I just wish the electrical code would permit or require certain low cost
things which make temporary generator connections more likely to be safe.

For example, code requires most furnaces to be hardwired.  But a furnace is
one of the first things you want on a generator in an extended winter power
outage.   If instead of hardwired, the code required plug and socket
connections at each 120v furnace  then Joe homeowner would be more likely
to run an extension cord from his generator to his furnace instead of
trying to rig up his generator with a suicide cord.


 Is $40-60 low cost enough for you for safe, temporary generator connections?

- Generator Interlock Kit: $20-25 (Safety)
- Breaker: $5 (30amp 120v) to $20 (60amp 240v) (Dedicated Power connection)
- Generator Power Inlet Input: $15 (indoor 120v) to $50 (outdoor 240v)

A Generator Interlock Kit is a few pieces of metal that, once
installed on your existing electrical panel, allows one to run
a properly-sized circuit and breaker to an outlet that you can plug your
120v or 240v generator inverter RV output into.

Add a Generator Power Inlet Input (indoor or outdoor) rated at 30Amp
240v NEMA L6-30P, for example, then plug your generator into that.

The Generator Interlock Kit physically prevents the mains from being on
when the generator Breaker is on. This is the safety component.

This seems affordable ($60 plus some wire and a few minutes inside your
electrical panel) and safe.

Add a few bucks to have your locality inspect and certify the work.

 If this is too much, why? What would be easier while also being equally as
 safe? This is work that, with a few minutes on YouTube, could do safely,
 as long as the power is disconnected at the meter outside the home during
 installation.

 PS - I suppose you could also move all of your emergency 120v stuff to one
 side of your panel and also provide only 120V to one side of your panel.
 This would also reduce costs a bit.

 Why believe me? In 2019 I read the NEC code and learned how to install a
 60amp circuit for an electric charger. I did the work myself. I had it
 inspected and certified by the county. I did so for about $100 total for
 all parts and wire.

Beckman
---
Peter Beckman  Internet Guy
beck...@angryox.com http://www.angryox.com/
---

Re: Reminder: Never connect a generator to home wiring without transfer switch

[Eric Germann via NANOG]

> On Aug 31, 2021, at 2:33 PM, Owen DeLong via NANOG  wrote:
> 
> 
> 
> ...

> 15kW is 1.5kVA in a simple radiant electric heat application. (it’s a simple 
> resistive load with no power factor weirdness). Whether you could do this 
> with 4-8kVA depends on what else you’re trying to run.
> 
> 
> 
> Owen


15kW is 15kVA (not 1.5 kVA) at a power factor of 1.0, if the heat is all 
resistive.

Eric

Re: The great Netflix vpn debacle! (geofeeds)

[Justin Krejci]

Well apparently there are VPN applications that rely on fellow VPN users in a 
P2P fashion to share network connectivity. I guess it is like a commercialized 
version of Tor to some extent. Excluding any potential legal risks for illegal 
behavior tunneled through an unsuspecting fellow user, this has great potential 
to cause a contaminating spread of VPN flagged IP addresses, even with just 
normal usage.


One such VPN application is Hola VPN which also has a premium version using 
their VPN server gateways instead of or perhaps in addition to the community 
method.


Dynamic IP address assignments by an ISP could easily allow for one such user 
to get many IP addresses flagged as a VPN gateway. I have communicated with 
some IP reputation companies and they track VPN users and can even supply the 
specific VPN brand associated with certain IP addresses, with timestamps, they 
have observed and added to their reputation databases as VPN users. How they 
obtain their data I do not know for sure but I can think of a few ways.


So we seem to have a battle between

  *   users
  *   streaming content providers
  *   streaming content owners / copyright holders
  *   ISPs
  *   VPN providers
  *   restrictive/invasive governments or network operators
  *   ??

There is definitely collateral damage from their use that should be considered, 
especially if very prominent streaming content providers take a more 
restrictive posture towards users of these kinds of VPN services.




From: NANOG  on behalf of Haudy 
Kazemi via NANOG 
Sent: Wednesday, September 1, 2021 4:44 PM
To: Owen DeLong; nanog list
Cc: b...@theworld.com
Subject: Re: The great Netflix vpn debacle! (geofeeds)

Some TVs may also try to rescale the inputs, or enhance/process the image in 
ways that can improve perceived video quality. Things like increasing frame 
rates of sources that are lower frame rates (thus the 120 Hz and 240 Hz TVs 
that attempt to make 24, 30, and 60 FPS sources look better), or deinterlacing 
1080i ATSC sources.

Some of this image processing may not work well in specific monitor use cases.

I have had generally good results with using a TV as an HTPC monitor.  Only 
issues I've run into over the years are

1.) a 1080p Sony TV with a VGA input that could not handle 1920x1080 (using 
HDMI worked)
and
2.) a 720p Toshiba that could not show the BIOS screen of the attached computer 
(I think this was either an unsupported resolution issue, or a timing issue 
where the TV couldn't wake up fast enough from the 'signal lost' message to 
display a brand new signal input).

YMMV.


VPNs: there is a race going on between streaming services who want to block 
VPNs, and VPN services who have customers who want to be able to watch streams 
(whether in or out of their regions). Some VPN customers buy VPN services 
because they do not trust their ISP to not do stuff like selling browsing 
histories.

I think ISPs are getting caught in the middle, maybe when they have IP ranges 
near or in the middle of ranges that are suspected by IP reputation companies 
as being used by VPN services. I'd guess the problem is more likely to affect 
smaller ISPs, and not the Comcast/Cox/Charter/Spectrum/CenturyLinks of the 
world. There are also 'distributed VPN' services that let people share their 
connections with others.

We are also seeing fragmentation in the cable/streaming service space, similar 
to what happened in the cable/Dish Network/DirecTV wars. Add it all up, some 
customers may throw up their hands in annoyance at the various platforms and 
then revert to other means of obtaining the content they seek.



On Wed, Sep 1, 2021, 15:13 Owen DeLong via NANOG < 
nanog@nanog.org> wrote:


> On Sep 1, 2021, at 11:25 , b...@theworld.com wrote:
>
>
> Every time I've read a thread about using TVs for monitors several
> people who'd tried would say don't do it. I think the gist was that
> the image processors in the TVs would fuzz text or something like
> that. That it was usable but they were unhappy with their attempts, it
> was tiring on the eyes.

That was definitely true of 480 TVs and older 1080p units, but modern sets
are almost designed to be monitors first and everything else second.

> Maybe that's changed or maybe people happy with this don't do a lot of
> text? Or maybe there are settings involved they weren't aware of, or
> some TVs (other than superficial specs like 4K vs 720p) are better for
> this than others so some will say they're happy and others not so
> much?

There are some tradeoffs… For example, sitting normal computer monitor
distance from a 44” 4K screen, you can damn near see the individual pixels
and that can make text look fuzzy, especially if your GPU or OS are stupid
enough to use a technique called anti-aliasing on text (which is the most
probable source of the fuzziness in your originally quoted complaint).

Older TVs would try to sm

Re: The great Netflix vpn debacle! (geofeeds)

[Michael Thomas]

On 9/1/21 3:17 PM, Warren Kumari wrote:



On Wed, Sep 1, 2021 at 2:28 PM > wrote:



Every time I've read a thread about using TVs for monitors several
people who'd tried would say don't do it.


And everytime I see an email thread about the difference or not 
between monitors and TVs I'm taken over by an all consuming rage...
I have a **monitor** I purchased it from Dell, and it clearly said 
"monitor" on the box, it identifies itself somewhere display settings 
as a "monitor", and even says "monitor" in small letters somewhere on 
the back It's a MONITOR dagnabit... but, for some unfathomable 
reason it has some tiny little speakers in it, and every time I 
connect it via HDMI to my Mac laptop, the machine decides to 
completely ignore the fact that I've told it that I want to use a 
specific sound output, and starts playing all audio though the 
monitors speakers. Oh, and because this is HDMI, and Apple apparently 
follows the HDMI spec, the Mac volume controls won't work ("This 
device has no audio level control" or something...) and I have to go 
scrummaging around in some horrendous on-screen monitor menu to make 
it less obnoxiously loud...


Huh. I have a Mac and my monitor was definitely marketed as a TV and all 
I do is just turn the volume down on the TV remote and don't have issues 
with the Mac not honoring where its audio output is. So there is 
obviously something different between our two setups. It does like you 
say not have the ability to control volume which I don't understand 
because my chromecast can do that and its only cable is HDMI so 
obviously the Mac can too.





All attempts to get this less stupid result in Apple pointing at the 
HDMI spec and saying that if a device advertises audio capabilites 
they list it as an output device, and Dell pointing out that they 
simply advirtise the fact that the device has a speaker, and, well, 
shrug, not thier issue if things try and use it.


I can understand why they have speakers and all of that even if it's 
just a monitor because it's probably cheaper to just have one model to 
manufacture and just rebrand it. There was some device -- gad I want to 
think it was an old DEC terminal server -- that just filled in the 
serial ports with glue or something so that you couldn't use them. That 
was pretty shameless.


Mike

Re: The great Netflix vpn debacle! (geofeeds)

[Owen DeLong via NANOG]

> On Sep 1, 2021, at 15:17 , Warren Kumari  wrote:
> 
> 
> 
> On Wed, Sep 1, 2021 at 2:28 PM mailto:b...@theworld.com>> 
> wrote:
> 
> Every time I've read a thread about using TVs for monitors several
> people who'd tried would say don't do it.
> 
> And everytime I see an email thread about the difference or not between 
> monitors and TVs I'm taken over by an all consuming rage...
> I have a **monitor** I purchased it from Dell, and it clearly said 
> "monitor" on the box, it identifies itself somewhere display settings as a 
> "monitor", and even says "monitor" in small letters somewhere on the back 
> It's a MONITOR dagnabit... but, for some unfathomable reason it has some tiny 
> little speakers in it, and every time I connect it via HDMI to my Mac laptop, 
> the machine decides to completely ignore the fact that I've told it that I 
> want to use a specific sound output, and starts playing all audio though the 
> monitors speakers. Oh, and because this is HDMI, and Apple apparently follows 
> the HDMI spec, the Mac volume controls won't work ("This device has no audio 
> level control" or something...) and I have to go scrummaging around in some 
> horrendous on-screen monitor menu to make it less obnoxiously loud...

Yes, it’s not clear why Apple doesn’t implement more of the HDMI spec and send 
it CEC commands to control the volume when it’s connected to an HDMI device 
with sound output.

Interestingly, my Apple TV does implement that part of the spec and my Amp that 
it is connected to dutifully obeys and everything works as expected… Display on 
the monitor (TV if you prefer), sound from the 7.1 speakers through the amp as 
expected, and control of the playback through the Apple TV all from the single 
elegant Apple TV Remote. So clearly, Apple has mastered the skills necessary to 
make this possible. Why they don’t bring them to MacOS yet remains a mystery to 
me.

> All attempts to get this less stupid result in Apple pointing at the HDMI 
> spec and saying that if a device advertises audio capabilites they list it as 
> an output device, and Dell pointing out that they simply advirtise the fact 
> that the device has a speaker, and, well, shrug, not thier issue if things 
> try and use it.

Listing it as an output device doesn’t require them to auto switch to that 
output device upon connection… You might want to point out to Apple that an 
ability to override this less than desirable behavior would be sufficient to 
cure your issue without violating the HDMI spec.

It pains me to say this, but Dell is right. The HDMI spec doesn’t allow for 
them to have a (useful) implementation of a speaker (or speakers) in an HDMI 
monitor that can some how say “I have a speaker, but don’t use it unless the 
user specifically tells you to.”. OTOH, Dell could (and I’ve seen monitors and 
even televisions that do) add a user control to “Disable HDMI audio 
negotiations” or something to that effect.

> There used to be a good webpage that had some instructions along the lines of:
> Step 1: Open 
> /System/Library/Extensions/AMDRadeonX6000HWServices.kext/Contents/PlugIns/AMDRadeonX6300HWLibs.kext
>  in a hex editor
> Step 2: Change the byte at offset 931 to 0xED, offset 12323 to 0xFD, offset 
> 94 to 0x00 and offset 42 to 0x03. 
> Step 3: ???
> Step 4: The HDMI capabilities parser no longer understands the audio 
> capability message, and so the Mac will never try to use HDMI audio ever 
> again well, until you upgrade... oh, this is perfectly safe, trust us, 
> nothing could possibly go wrong here...
> 
> Unfortunately this was only for a specific version of a specific kext on a 
> specific model of Macbook, but it did work... 

I suppose, if you’re willing to never have the ability to use HDMI Audio Output 
from your laptop (which wouldn’t work well for me).

I will say that it’s annoying to have to do it each time you connect to the 
monitor, but it is relatively trivial to change the audio output back after the 
monitor and laptop finish their whole HDMI negotiation and the various auto 
switches have finished screwing up your system settings.

System Preferences->Audio->Output — Select the output you want instead of the 
HDMI monitor.

> All I want is to be able to reliably inform my computer that the thingie on 
> my desk is "just" a monitor and not a TV/HiFi system/similar... is that too 
> much to ask!?!!?!!?!??!! 

I’m reminded of a certain advertising slogan…
“Dude! You got [stuck with] a Dell.”

> (Actually, this used to annoy me enough that I purchased one of bunnie 
> Huang's NeTV (https://www.bunniestudios.com/blog/?cat=17 
> ) devices, which allows taking in 
> HDMI, munging it and sending it out (e.g to do text overlays). My plan was to 
> repurpose it as a straight data passthrough, but overriding the HDMI profile 
> info, but as with most of these sorts of projects I got sidetracked into 
> playing with the build environment in

Re: Carbon Monoxide warnings - keep generators outside 20ft away from doors and windows

[Jay Hennigan]

On 9/1/21 15:07, Haudy Kazemi via NANOG wrote:
Several articles have mentioned 8 transmission lines were lost to the 
hurricane (a single big event event). A casual reader might think 8 
lines would offer an 8-way level of redundancy.


My WAG is the reality of load vs capacity is more like a N-1 or N-2 
redundancy, but that's really just a WAG. It is unclear to me if all 8 
lines were damaged by the storm, or if some failed/tripped when loads 
shifted onto the remaining lines after the first failure occurred 
(cascading failure).



Does anyone know, or has anyone seen, details?


Several pictures showing massive steel "cat-style" towers on the ground, 
as well as news reports that eight transmission lines are down.


Also saw a report that they will be isolating the storm area from the 
grid and bringing in temporary power plant(s), restoring critical load 
first.


--
Jay Hennigan - j...@west.net
Network Engineering - CCIE #7880
503 897-8550 - WB6RDV

Re: The great Netflix vpn debacle! (geofeeds)

[Warren Kumari]

On Wed, Sep 1, 2021 at 2:28 PM  wrote:

>
> Every time I've read a thread about using TVs for monitors several
> people who'd tried would say don't do it.


And everytime I see an email thread about the difference or not between
monitors and TVs I'm taken over by an all consuming rage...
I have a **monitor** I purchased it from Dell, and it clearly said
"monitor" on the box, it identifies itself somewhere display settings as a
"monitor", and even says "monitor" in small letters somewhere on the
back It's a MONITOR dagnabit... but, for some unfathomable reason it
has some tiny little speakers in it, and every time I connect it via HDMI
to my Mac laptop, the machine decides to completely ignore the fact that
I've told it that I want to use a specific sound output, and starts playing
all audio though the monitors speakers. Oh, and because this is HDMI, and
Apple apparently follows the HDMI spec, the Mac volume controls won't work
("This device has no audio level control" or something...) and I have to go
scrummaging around in some horrendous on-screen monitor menu to make it
less obnoxiously loud...

All attempts to get this less stupid result in Apple pointing at the HDMI
spec and saying that if a device advertises audio capabilites they list it
as an output device, and Dell pointing out that they simply advirtise the
fact that the device has a speaker, and, well, shrug, not thier issue if
things try and use it.

There used to be a good webpage that had some instructions along the lines
of:
Step 1:
Open 
/System/Library/Extensions/AMDRadeonX6000HWServices.kext/Contents/PlugIns/AMDRadeonX6300HWLibs.kext
in a hex editor
Step 2: Change the byte at offset 931 to 0xED, offset 12323 to 0xFD, offset
94 to 0x00 and offset 42 to 0x03.
Step 3: ???
Step 4: The HDMI capabilities parser no longer understands the audio
capability message, and so the Mac will never try to use HDMI audio ever
again well, until you upgrade... oh, this is perfectly safe, trust us,
nothing could possibly go wrong here...

Unfortunately this was only for a specific version of a specific kext on a
specific model of Macbook, but it did work...

All I want is to be able to reliably inform my computer that the thingie on
my desk is "just" a monitor and not a TV/HiFi system/similar... is that too
much to ask!?!!?!!?!??!! 

(Actually, this used to annoy me enough that I purchased one of bunnie
Huang's NeTV (https://www.bunniestudios.com/blog/?cat=17) devices, which
allows taking in HDMI, munging it and sending it out (e.g to do text
overlays). My plan was to repurpose it as a straight data passthrough, but
overriding the HDMI profile info, but as with most of these sorts of
projects I got sidetracked into playing with the build environment instead,
and now the hardware is buried under a pile of other abandoned projects
somewhere on my workbench)

Thank you all, I feel much better now...
W



> I think the gist was that
> the image processors in the TVs would fuzz text or something like
> that. That it was usable but they were unhappy with their attempts, it
> was tiring on the eyes.
>
> Maybe that's changed or maybe people happy with this don't do a lot of
> text? Or maybe there are settings involved they weren't aware of, or
> some TVs (other than superficial specs like 4K vs 720p) are better for
> this than others so some will say they're happy and others not so
> much?
>
> Or maybe the unhappy ones were all trolls/sockpuppets from companies
> manufacturing/selling $500+ 24" **GAMING** monitors.
>
> On September 1, 2021 at 09:48 nanog@nanog.org (Owen DeLong via NANOG)
> wrote:
>  >
>  >
>  > > On Aug 31, 2021, at 18:01 , Michael Thomas  wrote:
>  > >
>  > >
>  > > On 8/31/21 4:40 PM, Owen DeLong via NANOG wrote:
>  > >> On the other hand, the last time I went looking for a 27” monitor, I
> ended up buying a 44” smart television because it was a cheaper HDMI 4K
> monitor than the 27” alternatives that weren’t televisions. (It also ended
> up being cheaper than the 27” televisions which didn’t do 4K only 1080p,
> but I digress).
>  > >
>  > > Back when 4k just came out and they were really expensive, I found a
> "TV" by an obscure brand called Seiki which was super cheap. It was a 39"
> model. It's just a monitor to me, but I have gotten really used to its size
> and not needing two different monitors (and the gfx card to support it).
> What's distressing is that I was looking at what would happen if I needed
> to replace it and there is this gigantic gap where there are 30" monitors
> (= expensive) and 50" TV's which are relatively cheap. The problem is that
> 40" is sort of Goldielocks with 4k where 50" is way too big and 30" is too
> small. Thankfully it's going on 10 years old and still working fine.
>  >
>  > Costco stocks several 44” 4K TV models (like the one I got) that are
> relatively cheap. It’s a little larger than your 40” goldilocks, but I
> think still within range.
>  >
>  > Owen
>  >
>
> --
> -Barry Shein
>
> Soft

Re: Carbon Monoxide warnings - keep generators outside 20ft away from doors and windows

[Haudy Kazemi via NANOG]

Several articles have mentioned 8 transmission lines were lost to the
hurricane (a single big event event). A casual reader might think 8 lines
would offer an 8-way level of redundancy.

My WAG is the reality of load vs capacity is more like a N-1 or N-2
redundancy, but that's really just a WAG. It is unclear to me if all 8
lines were damaged by the storm, or if some failed/tripped when loads
shifted onto the remaining lines after the first failure occurred
(cascading failure).

Does anyone know, or has anyone seen, details?



On Wed, Sep 1, 2021, 16:12 Sean Donelan  wrote:

> One person has died and at least 27 people are being treated for carbon
> monoxide poisoning from portable generators.
>
> Officials are reminding people to operate portable generators only
> outside, 20 feet away from homes, doors and windows.  Not in carports,
> garages, basements.
>
> To restore power, Entergy has "islanded" (disconnected) the City of New
> Orleans from the regional grid and started a local power plant. The
> transmission lines were toppled during the storm, but the cables were
> still connected to the terminals. Islanding the city makes sense, but I
> don't remember a power company islanding large parts of the grid before.
> Public officials are now saying it may be 30+ days to fully restore power.
>
> Entergy has implemented restoration priority, which means hospitals,
> public safety and critical infrastructure will be restored first. Along
> with some incidental customers on the same circuits.
>
>
> Customers out of service
>
> Louisiana - 987,588
> Mississippi - 31,516
> Florida - 21,867
> California - 21,339
> Pennsylvania - 10,415
>
> Reminder, Puerto Rico still has not fully recovered from hurricanes in
> 2017.  Puerto Rico still has rolling blackouts.  And yes, Puerto Rico is
> an island, so its electric grid is naturally an island.
>
> The major wireless providers have activated their open roaming agreements,
> allowing customers to roam on any working infrastructure from other
> service providers.  They are also waiving overages and many other feeds in
> the affected region.  Check your service provider's website for details.
>
> AT&T says 82 percent of its network in service in Louisiana.
>
> First responders say the AT&T FIRSTNET failed (again) during the
> hurricane.
>
> T-Mobile says 70 percent of its network in service in Louisiana.
>
> Verizon says it has "gaps in coverage" but its network remains resilient.
> I don't know what that means.
>
> I haven't found reports from cable companies in the region.
>

NANOG Embraces Hybrid FAQs + Upcoming Webinar

[Nanog News]

Upcoming Webinar - Fireside Chat with CENIC Network Interconnection Engineer

*The fireside chat will include several topics including:*

   - The path taken to a tech career
   - What is CENIC
   - Opportunities for volunteer or educational work

*Speaker:* Nick Plunkett, Network Interconnection Engineer - Internet
Services at CENIC

Nick is a Network Interconnection Engineer - Internet Services at CENIC,
where he has been working with both internal and external colleagues to
expand the interconnection network of the California Research and Education
Network (CalREN).

*REGISTER NOW  *
NANOG Embraces Hybrid Meetings

*Meetings will now be offered in-person + virtually — *NANOG has made a
permanent shift to becoming an hybrid event platform, offering equally
robust programming from the comfort of your laptop and/or in-person venues!

We are excited to announce NANOG 83
as our first hybrid event. NANOG 83
will take place virtually + in-person in Minneapolis, MN from Nov. 1 - 3.
Register now.  

*Have questions? *We have answers! Check out our FAQ sheet.

*READ FAQs* 

Re: The great Netflix vpn debacle! (geofeeds)

[Haudy Kazemi via NANOG]

Some TVs may also try to rescale the inputs, or enhance/process the image
in ways that can improve perceived video quality. Things like increasing
frame rates of sources that are lower frame rates (thus the 120 Hz and 240
Hz TVs that attempt to make 24, 30, and 60 FPS sources look better), or
deinterlacing 1080i ATSC sources.

Some of this image processing may not work well in specific monitor use
cases.

I have had generally good results with using a TV as an HTPC monitor.  Only
issues I've run into over the years are

1.) a 1080p Sony TV with a VGA input that could not handle 1920x1080 (using
HDMI worked)
and
2.) a 720p Toshiba that could not show the BIOS screen of the attached
computer (I think this was either an unsupported resolution issue, or a
timing issue where the TV couldn't wake up fast enough from the 'signal
lost' message to display a brand new signal input).

YMMV.


VPNs: there is a race going on between streaming services who want to block
VPNs, and VPN services who have customers who want to be able to watch
streams (whether in or out of their regions). Some VPN customers buy VPN
services because they do not trust their ISP to not do stuff like selling
browsing histories.

I think ISPs are getting caught in the middle, maybe when they have IP
ranges near or in the middle of ranges that are suspected by IP reputation
companies as being used by VPN services. I'd guess the problem is more
likely to affect smaller ISPs, and not the
Comcast/Cox/Charter/Spectrum/CenturyLinks of the world. There are also
'distributed VPN' services that let people share their connections with
others.

We are also seeing fragmentation in the cable/streaming service space,
similar to what happened in the cable/Dish Network/DirecTV wars. Add it all
up, some customers may throw up their hands in annoyance at the various
platforms and then revert to other means of obtaining the content they seek.



On Wed, Sep 1, 2021, 15:13 Owen DeLong via NANOG  wrote:

>
>
> > On Sep 1, 2021, at 11:25 , b...@theworld.com wrote:
> >
> >
> > Every time I've read a thread about using TVs for monitors several
> > people who'd tried would say don't do it. I think the gist was that
> > the image processors in the TVs would fuzz text or something like
> > that. That it was usable but they were unhappy with their attempts, it
> > was tiring on the eyes.
>
> That was definitely true of 480 TVs and older 1080p units, but modern sets
> are almost designed to be monitors first and everything else second.
>
> > Maybe that's changed or maybe people happy with this don't do a lot of
> > text? Or maybe there are settings involved they weren't aware of, or
> > some TVs (other than superficial specs like 4K vs 720p) are better for
> > this than others so some will say they're happy and others not so
> > much?
>
> There are some tradeoffs… For example, sitting normal computer monitor
> distance from a 44” 4K screen, you can damn near see the individual pixels
> and that can make text look fuzzy, especially if your GPU or OS are stupid
> enough to use a technique called anti-aliasing on text (which is the most
> probable source of the fuzziness in your originally quoted complaint).
>
> Older TVs would try to smooth some aspects of the analog signal they were
> using through anti-aliasing pixels that occurred on the edge of a change in
> the color signal to “smooth” the image. (The extent of this action was what
> was controlled by the “Sharpness” knob back in the analog days).
>
> Turning off this capability (Sharpness to the left most or lowest setting)
> would
> often improve things greatly.
>
> > Or maybe the unhappy ones were all trolls/sockpuppets from companies
> > manufacturing/selling $500+ 24" **GAMING** monitors.
>
> Possible, but unlikely.
>
> Owen
>
> >
> > On September 1, 2021 at 09:48 nanog@nanog.org (Owen DeLong via NANOG)
> wrote:
> >>
> >>
> >>> On Aug 31, 2021, at 18:01 , Michael Thomas  wrote:
> >>>
> >>>
> >>> On 8/31/21 4:40 PM, Owen DeLong via NANOG wrote:
>  On the other hand, the last time I went looking for a 27” monitor, I
> ended up buying a 44” smart television because it was a cheaper HDMI 4K
> monitor than the 27” alternatives that weren’t televisions. (It also ended
> up being cheaper than the 27” televisions which didn’t do 4K only 1080p,
> but I digress).
> >>>
> >>> Back when 4k just came out and they were really expensive, I found a
> "TV" by an obscure brand called Seiki which was super cheap. It was a 39"
> model. It's just a monitor to me, but I have gotten really used to its size
> and not needing two different monitors (and the gfx card to support it).
> What's distressing is that I was looking at what would happen if I needed
> to replace it and there is this gigantic gap where there are 30" monitors
> (= expensive) and 50" TV's which are relatively cheap. The problem is that
> 40" is sort of Goldielocks with 4k where 50" is way too big and 30" is too
> small. Thankfully it's going on 10 years old and stil

Carbon Monoxide warnings - keep generators outside 20ft away from doors and windows

[Sean Donelan]

One person has died and at least 27 people are being treated for carbon 
monoxide poisoning from portable generators.


Officials are reminding people to operate portable generators only 
outside, 20 feet away from homes, doors and windows.  Not in carports, 
garages, basements.


To restore power, Entergy has "islanded" (disconnected) the City of New 
Orleans from the regional grid and started a local power plant. The 
transmission lines were toppled during the storm, but the cables were 
still connected to the terminals. Islanding the city makes sense, but I 
don't remember a power company islanding large parts of the grid before. 
Public officials are now saying it may be 30+ days to fully restore power.


Entergy has implemented restoration priority, which means hospitals, 
public safety and critical infrastructure will be restored first. Along 
with some incidental customers on the same circuits.



Customers out of service

Louisiana - 987,588
Mississippi - 31,516
Florida - 21,867
California - 21,339
Pennsylvania - 10,415

Reminder, Puerto Rico still has not fully recovered from hurricanes in 
2017.  Puerto Rico still has rolling blackouts.  And yes, Puerto Rico is 
an island, so its electric grid is naturally an island.


The major wireless providers have activated their open roaming agreements, 
allowing customers to roam on any working infrastructure from other 
service providers.  They are also waiving overages and many other feeds in 
the affected region.  Check your service provider's website for details.


AT&T says 82 percent of its network in service in Louisiana.

First responders say the AT&T FIRSTNET failed (again) during the 
hurricane.


T-Mobile says 70 percent of its network in service in Louisiana.

Verizon says it has "gaps in coverage" but its network remains resilient. 
I don't know what that means.


I haven't found reports from cable companies in the region.

Re: An update on the AfriNIC situation

[Randy Bush]

again, do not be distracted by the rather obvious DoS on this list.  our
administrative infra is being attacked.  defend it by putting your money
where your mouth is.

https://www.tespok.co.ke/?page_id=14001

i did and will again.

randy

Re: An update on the AfriNIC situation

[Noah]

On Wed, 1 Sep 2021, 22:46 Owen DeLong via NANOG,  wrote:

>
>
> > On Sep 1, 2021, at 04:48 , Mark Tinka  wrote:
> >
> >
> >
> > On 9/1/21 00:56, Owen DeLong via NANOG wrote:
> >
> >> Not to put too fine a point on this, but what human cost?
> >>
> >> There were exactly 3 employees that AFRINIC wasn’t able to pay in July,
> including
> >> the CEO (who is one of the major protagonists in creating this problem
> in the first
> >> place). I don’t know who the other two were.
> >>
> >> Everyone else got paid for July.
> >>
> >> AFRINIC has received clearance of enough money to cover their normal
> expenses
> >> for August and September. As such, there shouldn’t be any problems with
> salaries
> >> or “human cost” in those months. Hopefully given that reprieve, cooler
> heads at
> >> AFRINIC can prevail and some form of settlement can be achieved before
> they run
> >> out of money from that reprieve.
> >
> > This is rich!
> >
> > Of course, one should expect people to be mentally settled and do good
> work when they have no security or clarity about whether they will get paid
> at the end of each month.
> >
> > These aren't robots, mate. People have real thoughts and real feelings.
> >
> > Stress is not intangible... it releases cortisol, which delivers a
> number of physiological side effects that work against good health.
> Financial insecurity is just about the worst stress anyone has to deal with.
> >
> > None of us would sleep well if we knew that our source of income is not
> guaranteed, despite what some may say about "As such, there shouldn't be
> any problems with salaries..."
> >
> > Mark.
>
> Well… I don’t see you calling out the stress and cortisol actions for the
> various staff and customers of Cloud Innovation or Larus in your postings
> when AFRINIC first issued an existential threat to their business based on
> made-up policies that don’t actually exist.
>


M see below.


> As such, I’d argue that AFRINIC attacked a much larger population first.


You are not being honest Owen.

See below an extract from AFRNIC CEO message on the termination of CIL.

>



   -

In order not to disrupt Internet connectivity of the relevant users
   especially in the current context of the COVID-19 pandemic, all affected
   users will exceptionally be granted a grace period of 90 days to consider
   other available options in their best interests. Consequently, the actual
   reclamation of the relevant number resources will occur following the
   expiry of the grace period.


Eddy Kayihura
Chief Executive Officer,
African Network Information Centre (AFRINIC)




So how did AFRINIC attack the so called larger population you are claiming
about?

All Cloud Innovation Ltd clients had been given 90 days to work things out
with AFRINIC.

Cheers,
Noah

Re: An update on the AfriNIC situation

[Owen DeLong via NANOG]

> On Sep 1, 2021, at 13:30 , Tom Beecher  wrote:
> 
>  They attacked a member on the basis of violations of rules that don’t 
> actually exist.
> 
> You continually refer to AFRINIC's actions as an 'attack'. However, that 
> would seem to be an open question of law , which AFRINIC cannot litigate 
> because they're have no access to their money. 

Apparently this isn’t an issue for them as they are, in fact, litigating it and 
have managed to hire the most expensive law firm in Mauritius to do so.

Until they went looking for other excuses, their attack was based primarily on 
accusations that out of region use was not permitted.

All of the resources in question were issued prior to activation of the Soft 
Landing (5.4) policy, so 5.4.6.2 does not apply.

If you can find some other place in the CPM that prohibits such, then by all 
means, let me know. If you cannot, then I think the openness of that question 
is rather limited.

Owen

> 
> On Wed, Sep 1, 2021 at 3:40 PM Owen DeLong  > wrote:
> 
> 
>> On Sep 1, 2021, at 04:21 , Tom Beecher > > wrote:
>> 
>> AFRINIC has received clearance of enough money to cover their normal expenses
>> for August and September. As such, there shouldn’t be any problems with 
>> salaries
>> or “human cost” in those months. Hopefully given that reprieve, cooler heads 
>> at
>> AFRINIC can prevail and some form of settlement can be achieved before they 
>> run
>> out of money from that reprieve.
>> 
>> It's good that people are still being paid.
>> 
>> That being said, while some may have the opinion that AFRINIC's actions have 
>> been 'objectionable' , others have the opinion that their actions were 
>> justified and proper. Does it not concern you at all that AFRINIC may be 
>> forced into a 'settlement' because they cannot access their funds due to a 
>> very dubious claim of damages? 
> 
> I don’t think AFRINIC is being forced into anything. They attacked a member 
> on the basis of violations of rules that don’t actually exist. The company 
> retaliated by obtaining an injunction. AFRINIC managed to get the injunction 
> temporarily lifted on a technicality (largely of AFRINIC’s making) and took 
> an opportunity to make an existential attack on the member during that brief 
> window of opportunity. AFRINIC then deliberately dragged their feet on 
> complying with a court order restoring the injunction.
> 
> As such, no, I’m really not concerned about AFRINIC’s ability to avoid 
> settling and I utterly reject the idea that the claim of damages is at all 
> dubious.
> 
>> There are enough challenges with the internet in Africa to work through 
>> already. We shouldn't encourage more difficulties by endorsing strongarm 
>> tactics that prevent issues from being properly adjudicated in courts.
> 
> Agreed. Please review the recent conduct of the AFRINIC board in detail in 
> this context.
> 
> Owen
> 
>> 
>> On Tue, Aug 31, 2021 at 6:59 PM Owen DeLong via NANOG > > wrote:
>> >> I regret the true human cost that Mark pointed out, yet I am fascinated
>> >> by the case and the arguments on both sides. The court will have their
>> >> work cut out for them.
>> > 
>> > That human cost came not from disagreement on the policies and
>> > contract provisions, but from a vengeful action of financial bullying.
>> 
>> Not to put too fine a point on this, but what human cost?
>> 
>> There were exactly 3 employees that AFRINIC wasn’t able to pay in July, 
>> including
>> the CEO (who is one of the major protagonists in creating this problem in 
>> the first
>> place). I don’t know who the other two were.
>> 
>> Everyone else got paid for July.
>> 
>> AFRINIC has received clearance of enough money to cover their normal expenses
>> for August and September. As such, there shouldn’t be any problems with 
>> salaries
>> or “human cost” in those months. Hopefully given that reprieve, cooler heads 
>> at
>> AFRINIC can prevail and some form of settlement can be achieved before they 
>> run
>> out of money from that reprieve.
>> 
>> > I saw my quota of questionable court decisions to automatically agree
>> > with whatever is decided in this case, even if CI loses, but the
>> > arguments from both sides will indeed be very interesting and useful
>> > to close out loopholes in the system.
>> 
>> Only if they are ever able to be made public, which is a little iffy given 
>> the Mauritian
>> court system. It may well be that only the final ruling is able to be made 
>> public.
>> 
>> Owen
>> 
> 

Re: Happy 40th anniversary RFC 791!

[John Curran]

On 1 Sep 2021, at 2:42 PM, Mel Beckman  wrote:
> 
> For anyone unaware, Jon Postel, a good friend and mentor to many of us at the 
> dawn of the Internet, was the primary editor of this landmark document. 
> 
> Those were the days we thought ARPAnet would never be allowed to go 
> commercial. Thanks to Jon’s tireless campaigning (among others), not to 
> mention meticulous documentation, it did. 
> 
> He was taken from us too soon.
> 
> https://www.ietfjournal.org/in-memory-of-jon-postel/

Indeed.   Also worth reading is Vint’s “I remember IANA” RFC 
> – 
calling out Jon’s enormous contribution to our industry & honored by the ISOC 
service award bearing his name.

FYI,
/John

Re: An update on the AfriNIC situation

[Tom Beecher]

>
>  They attacked a member on the basis of violations of rules that don’t
> actually exist.
>

You continually refer to AFRINIC's actions as an 'attack'. However, that
would seem to be an open question of law , which AFRINIC cannot litigate
because they're have no access to their money.

On Wed, Sep 1, 2021 at 3:40 PM Owen DeLong  wrote:

>
>
> On Sep 1, 2021, at 04:21 , Tom Beecher  wrote:
>
> AFRINIC has received clearance of enough money to cover their normal
>> expenses
>> for August and September. As such, there shouldn’t be any problems with
>> salaries
>> or “human cost” in those months. Hopefully given that reprieve, cooler
>> heads at
>> AFRINIC can prevail and some form of settlement can be achieved before
>> they run
>> out of money from that reprieve.
>>
>
> It's good that people are still being paid.
>
> That being said, while some may have the opinion that AFRINIC's actions
> have been 'objectionable' , others have the opinion that their actions were
> justified and proper. Does it not concern you at all that AFRINIC may be
> forced into a 'settlement' because they cannot access their funds due to a
> very dubious claim of damages?
>
>
> I don’t think AFRINIC is being forced into anything. They attacked a
> member on the basis of violations of rules that don’t actually exist. The
> company retaliated by obtaining an injunction. AFRINIC managed to get the
> injunction temporarily lifted on a technicality (largely of AFRINIC’s
> making) and took an opportunity to make an existential attack on the member
> during that brief window of opportunity. AFRINIC then deliberately dragged
> their feet on complying with a court order restoring the injunction.
>
> As such, no, I’m really not concerned about AFRINIC’s ability to avoid
> settling and I utterly reject the idea that the claim of damages is at all
> dubious.
>
> There are enough challenges with the internet in Africa to work through
> already. We shouldn't encourage more difficulties by endorsing strongarm
> tactics that prevent issues from being properly adjudicated in courts.
>
>
> Agreed. Please review the recent conduct of the AFRINIC board in detail in
> this context.
>
> Owen
>
>
> On Tue, Aug 31, 2021 at 6:59 PM Owen DeLong via NANOG 
> wrote:
>
>> >> I regret the true human cost that Mark pointed out, yet I am fascinated
>> >> by the case and the arguments on both sides. The court will have their
>> >> work cut out for them.
>> >
>> > That human cost came not from disagreement on the policies and
>> > contract provisions, but from a vengeful action of financial bullying.
>>
>> Not to put too fine a point on this, but what human cost?
>>
>> There were exactly 3 employees that AFRINIC wasn’t able to pay in July,
>> including
>> the CEO (who is one of the major protagonists in creating this problem in
>> the first
>> place). I don’t know who the other two were.
>>
>> Everyone else got paid for July.
>>
>> AFRINIC has received clearance of enough money to cover their normal
>> expenses
>> for August and September. As such, there shouldn’t be any problems with
>> salaries
>> or “human cost” in those months. Hopefully given that reprieve, cooler
>> heads at
>> AFRINIC can prevail and some form of settlement can be achieved before
>> they run
>> out of money from that reprieve.
>>
>> > I saw my quota of questionable court decisions to automatically agree
>> > with whatever is decided in this case, even if CI loses, but the
>> > arguments from both sides will indeed be very interesting and useful
>> > to close out loopholes in the system.
>>
>> Only if they are ever able to be made public, which is a little iffy
>> given the Mauritian
>> court system. It may well be that only the final ruling is able to be
>> made public.
>>
>> Owen
>>
>>
>

Re: The great Netflix vpn debacle! (geofeeds)

[Owen DeLong via NANOG]

> On Sep 1, 2021, at 11:25 , b...@theworld.com wrote:
> 
> 
> Every time I've read a thread about using TVs for monitors several
> people who'd tried would say don't do it. I think the gist was that
> the image processors in the TVs would fuzz text or something like
> that. That it was usable but they were unhappy with their attempts, it
> was tiring on the eyes.

That was definitely true of 480 TVs and older 1080p units, but modern sets
are almost designed to be monitors first and everything else second.

> Maybe that's changed or maybe people happy with this don't do a lot of
> text? Or maybe there are settings involved they weren't aware of, or
> some TVs (other than superficial specs like 4K vs 720p) are better for
> this than others so some will say they're happy and others not so
> much?

There are some tradeoffs… For example, sitting normal computer monitor
distance from a 44” 4K screen, you can damn near see the individual pixels
and that can make text look fuzzy, especially if your GPU or OS are stupid
enough to use a technique called anti-aliasing on text (which is the most
probable source of the fuzziness in your originally quoted complaint).

Older TVs would try to smooth some aspects of the analog signal they were
using through anti-aliasing pixels that occurred on the edge of a change in
the color signal to “smooth” the image. (The extent of this action was what
was controlled by the “Sharpness” knob back in the analog days).

Turning off this capability (Sharpness to the left most or lowest setting) would
often improve things greatly.

> Or maybe the unhappy ones were all trolls/sockpuppets from companies
> manufacturing/selling $500+ 24" **GAMING** monitors.

Possible, but unlikely.

Owen

> 
> On September 1, 2021 at 09:48 nanog@nanog.org (Owen DeLong via NANOG) wrote:
>> 
>> 
>>> On Aug 31, 2021, at 18:01 , Michael Thomas  wrote:
>>> 
>>> 
>>> On 8/31/21 4:40 PM, Owen DeLong via NANOG wrote:
 On the other hand, the last time I went looking for a 27” monitor, I ended 
 up buying a 44” smart television because it was a cheaper HDMI 4K monitor 
 than the 27” alternatives that weren’t televisions. (It also ended up 
 being cheaper than the 27” televisions which didn’t do 4K only 1080p, but 
 I digress).
>>> 
>>> Back when 4k just came out and they were really expensive, I found a "TV" 
>>> by an obscure brand called Seiki which was super cheap. It was a 39" model. 
>>> It's just a monitor to me, but I have gotten really used to its size and 
>>> not needing two different monitors (and the gfx card to support it). What's 
>>> distressing is that I was looking at what would happen if I needed to 
>>> replace it and there is this gigantic gap where there are 30" monitors (= 
>>> expensive) and 50" TV's which are relatively cheap. The problem is that 40" 
>>> is sort of Goldielocks with 4k where 50" is way too big and 30" is too 
>>> small. Thankfully it's going on 10 years old and still working fine.
>> 
>> Costco stocks several 44” 4K TV models (like the one I got) that are 
>> relatively cheap. It’s a little larger than your 40” goldilocks, but I think 
>> still within range.
>> 
>> Owen
>> 
> 
> -- 
>-Barry Shein
> 
> Software Tool & Die| b...@theworld.com | 
> http://www.TheWorld.com
> Purveyors to the Trade | Voice: +1 617-STD-WRLD   | 800-THE-WRLD
> The World: Since 1989  | A Public Information Utility | *oo*

Re: The great Netflix vpn debacle! (geofeeds)

[Owen DeLong via NANOG]

Where possible vote with your dollars by selecting providers that do.

Where there are multiple providers and none support v6, make it clear to all 
that the first one to support
v6 will get your business and that subsequently, the best v6 support will win.

Where there are not multiple providers, lobby your regulators to eliminate 
vertical integration (stop allowing
those that own the natural monopoly in layer 1 to leverage that into a monopoly 
over higher layer services).

Owen


> On Sep 1, 2021, at 10:59 , Nimrod Levy  wrote:
> 
> All this chatter about IPv6 support on devices is fun and all, but there are 
> providers still not on board. 
> They operate in my neighborhood and they know who they are...
> 
> Nimrod
> 
> 

Re: Happy 40th anniversary RFC 791!

[Michael Thomas]

On 9/1/21 12:26 PM, Mel Beckman wrote:

I still have a slew on Lantronix terminal servers :)


A few years back I was shocked to hear that the original OS that I wrote 
-- called whimsically Punix for Puny Unix -- which was used by Lantronix 
was still being sold. I mean, that's over 30 years ago. I doubt our IP 
and TCP drivers changed much in that time. One of the cute things I did 
was pump out most of the character IO in the null job so that it didn't 
have to do interrupt based IO for the most part, which kept the costs 
low (= no DMA silicon, cheaper processors).


It still amazes me that we built the internet basically without the 
internet. I came >.< close to driving up to ISI to attend an IETF 
meeting around 1987 or maybe 88 to complain about the shitty LPR 
interface on Unix and what could be done about it. I'm not sure whether 
it would have fallen on receptive ears, but I was totally naive about 
how new this all was.


Mike

Re: An update on the AfriNIC situation

[Owen DeLong via NANOG]

> On Sep 1, 2021, at 04:48 , Mark Tinka  wrote:
> 
> 
> 
> On 9/1/21 00:56, Owen DeLong via NANOG wrote:
> 
>> Not to put too fine a point on this, but what human cost?
>> 
>> There were exactly 3 employees that AFRINIC wasn’t able to pay in July, 
>> including
>> the CEO (who is one of the major protagonists in creating this problem in 
>> the first
>> place). I don’t know who the other two were.
>> 
>> Everyone else got paid for July.
>> 
>> AFRINIC has received clearance of enough money to cover their normal expenses
>> for August and September. As such, there shouldn’t be any problems with 
>> salaries
>> or “human cost” in those months. Hopefully given that reprieve, cooler heads 
>> at
>> AFRINIC can prevail and some form of settlement can be achieved before they 
>> run
>> out of money from that reprieve.
> 
> This is rich!
> 
> Of course, one should expect people to be mentally settled and do good work 
> when they have no security or clarity about whether they will get paid at the 
> end of each month.
> 
> These aren't robots, mate. People have real thoughts and real feelings.
> 
> Stress is not intangible... it releases cortisol, which delivers a number of 
> physiological side effects that work against good health. Financial 
> insecurity is just about the worst stress anyone has to deal with.
> 
> None of us would sleep well if we knew that our source of income is not 
> guaranteed, despite what some may say about "As such, there shouldn't be any 
> problems with salaries..."
> 
> Mark.

Well… I don’t see you calling out the stress and cortisol actions for the 
various staff and customers of Cloud Innovation or Larus in your postings when 
AFRINIC first issued an existential threat to their business based on made-up 
policies that don’t actually exist.

As such, I’d argue that AFRINIC attacked a much larger population first. The 
fact that (at least so far), AFRINIC’s attacks have been less successful than 
Cloud Innovation’s counter-attacks doesn’t change the fact that if you want to 
call it out that way, there’s a greater human cost on the other side.

Owen

Re: An update on the AfriNIC situation

[Owen DeLong via NANOG]

> On Sep 1, 2021, at 04:21 , Tom Beecher  wrote:
> 
> AFRINIC has received clearance of enough money to cover their normal expenses
> for August and September. As such, there shouldn’t be any problems with 
> salaries
> or “human cost” in those months. Hopefully given that reprieve, cooler heads 
> at
> AFRINIC can prevail and some form of settlement can be achieved before they 
> run
> out of money from that reprieve.
> 
> It's good that people are still being paid.
> 
> That being said, while some may have the opinion that AFRINIC's actions have 
> been 'objectionable' , others have the opinion that their actions were 
> justified and proper. Does it not concern you at all that AFRINIC may be 
> forced into a 'settlement' because they cannot access their funds due to a 
> very dubious claim of damages? 

I don’t think AFRINIC is being forced into anything. They attacked a member on 
the basis of violations of rules that don’t actually exist. The company 
retaliated by obtaining an injunction. AFRINIC managed to get the injunction 
temporarily lifted on a technicality (largely of AFRINIC’s making) and took an 
opportunity to make an existential attack on the member during that brief 
window of opportunity. AFRINIC then deliberately dragged their feet on 
complying with a court order restoring the injunction.

As such, no, I’m really not concerned about AFRINIC’s ability to avoid settling 
and I utterly reject the idea that the claim of damages is at all dubious.

> There are enough challenges with the internet in Africa to work through 
> already. We shouldn't encourage more difficulties by endorsing strongarm 
> tactics that prevent issues from being properly adjudicated in courts.

Agreed. Please review the recent conduct of the AFRINIC board in detail in this 
context.

Owen

> 
> On Tue, Aug 31, 2021 at 6:59 PM Owen DeLong via NANOG  > wrote:
> >> I regret the true human cost that Mark pointed out, yet I am fascinated
> >> by the case and the arguments on both sides. The court will have their
> >> work cut out for them.
> > 
> > That human cost came not from disagreement on the policies and
> > contract provisions, but from a vengeful action of financial bullying.
> 
> Not to put too fine a point on this, but what human cost?
> 
> There were exactly 3 employees that AFRINIC wasn’t able to pay in July, 
> including
> the CEO (who is one of the major protagonists in creating this problem in the 
> first
> place). I don’t know who the other two were.
> 
> Everyone else got paid for July.
> 
> AFRINIC has received clearance of enough money to cover their normal expenses
> for August and September. As such, there shouldn’t be any problems with 
> salaries
> or “human cost” in those months. Hopefully given that reprieve, cooler heads 
> at
> AFRINIC can prevail and some form of settlement can be achieved before they 
> run
> out of money from that reprieve.
> 
> > I saw my quota of questionable court decisions to automatically agree
> > with whatever is decided in this case, even if CI loses, but the
> > arguments from both sides will indeed be very interesting and useful
> > to close out loopholes in the system.
> 
> Only if they are ever able to be made public, which is a little iffy given 
> the Mauritian
> court system. It may well be that only the final ruling is able to be made 
> public.
> 
> Owen
> 

Re: Telecommunications network drafting software

[Mel Beckman]

I’ve found PowerPoint amazingly versatile for network design. It’s primary 
advantage over dedicated tools like Visio is that you can mix in photos, 
talking points, and migration checklists.  Plus everyone can read and edit 
PowerPoint.

You can use the stencils and icons from many graphics libraries. No, it won’t 
redraw lines as you drag objects (at least not well), but I find the other 
advantages persuasive.

-mel via cell

On Sep 1, 2021, at 12:13 PM, Michael Hallgren  wrote:


TikZ ?
mh

De : Etienne-Victor Depasquale via NANOG 
Envoyé : mercredi 1 septembre 2021 20:30
À : NANOG
Objet : Telecommunications network drafting software

Hello folks,

Would you care to share some pointers to drafting software which you use to 
draw up architectural drafts (for telecoms networks, including cable operators' 
networks) ?

I've found Visio to be a bit weak in this respect, even after adding third 
party stencils.

One product I'm exploring is ConceptDraw's "diagram" product, but I'd like to 
hear about anything you can share with me.

Cheers,

Etienne

--
Ing. Etienne-Victor Depasquale
Assistant Lecturer
Department of Communications & Computer Engineering
Faculty of Information & Communication Technology
University of Malta
Web. https://www.um.edu.mt/profile/etiennedepasquale

Re: Happy 40th anniversary RFC 791!

[Mel Beckman]

I still have a slew on Lantronix terminal servers :)

-mel via cell

> On Sep 1, 2021, at 11:57 AM, Michael Thomas  wrote:
> 
> 
>> On 9/1/21 11:42 AM, Mel Beckman wrote:
>> For anyone unaware, Jon Postel, a good friend and mentor to many of us at 
>> the dawn of the Internet, was the primary editor of this landmark document.
>> 
>> Those were the days we thought ARPAnet would never be allowed to go 
>> commercial. Thanks to Jon’s tireless campaigning (among others), not to 
>> mention meticulous documentation, it did.
>> 
>> He was taken from us too soon.
>> 
>> https://www.ietfjournal.org/in-memory-of-jon-postel/
>> 
> RFC 791 and 793 are remarkably easy to read and follow what's going on. And 
> that's not just hindsight speaking. We implemented both a few years later 
> purely from the RFC's with no connection to anybody else and when we actually 
> had kit that was connected to the Internet it just worked. One of the most 
> remarkable things in my life was that I wrote debugger for the OS I wrote for 
> the Lantronix terminal servers which I ended up remotely debugging in New 
> Zealand. That was just mind blowing. Of course you'd call that a backdoor 
> these days but they were very happy that I could figure out what was going on.
> 
> I never had the pleasure of meeting Jon, though I saw him at IETF meetings.
> 
> Mike
> 

Re: Telecommunications network drafting software

[Eric Kuhnke]

For logical diagrams of networks, on MacOS, I recommend Omnigraffle.


On Wed, Sep 1, 2021 at 2:36 PM Etienne-Victor Depasquale via NANOG <
nanog@nanog.org> wrote:

> Hello folks,
>
> Would you care to share some pointers to drafting software which you use
> to draw up architectural drafts (for telecoms networks, including cable
> operators' networks) ?
>
> I've found Visio to be a bit weak in this respect, even after adding third
> party stencils.
>
> One product I'm exploring is ConceptDraw's "diagram" product, but I'd like
> to hear about anything you can share with me.
>
> Cheers,
>
> Etienne
>
> --
> Ing. Etienne-Victor Depasquale
> Assistant Lecturer
> Department of Communications & Computer Engineering
> Faculty of Information & Communication Technology
> University of Malta
> Web. https://www.um.edu.mt/profile/etiennedepasquale
>

Re: The great Netflix vpn debacle! (geofeeds)

[Michael Thomas]

On 9/1/21 11:49 AM, Matthew Huff wrote:

IPv6 tunnels work great for network geeks, but rather poorly for home users 
with streaming, gaming etc...It's not necessarily the performance, it's either 
the geolocation, latency, or the very issue that started this thread - VPN 
banning.

Remember, the streaming services couldn't care less about geolocation or VPN 
banning, it's the contractual obligations with the content providers. The 
content providers care about vpn banning because it gets around geolocation, 
which interferes with their business models (different release schedules to 
different regions, etc..)

Been there, done that...Stuck on Fios with no IPv6. Ran into rather 
"interesting" problems with various streaming services with IPv6 configured.

Well, my point is that a properly pre-configured home router could 
probably make this plug and play. Openwrt can probably do what I'm 
thinking. Streaming should not be a problem but gaming/latency 
definitely is.


I frankly don't understand why these home router vendors don't just 
adopt Openwrt and the like instead of maintaining their own code. They 
are extremely cost sensitive so you'd think that it would be a big win 
(yes, I know some do but, say, Linksys doesn't and their software is 
complete shit and I know this first hand). Why can't I have router 
distos just like Linux distos where somebody with clue does the work to 
customize distos with various features. My ISP could then just point at 
the ones they like too. It's really sad that home routers are completely 
treated like black boxes where people and their devices have no problem 
customizing them to their taste. My suspicion is this all a 
self-fulfilling prophecy.


Mike

Re: Telecommunications network drafting software

[Michael Hallgren]

TikZ ? 
mh

De : Etienne-Victor Depasquale via NANOG 
Envoyé : mercredi 1 septembre 2021 20:30
À : NANOG
Objet : Telecommunications network drafting software

Hello folks,

Would you care to share some pointers to drafting software which you use to 
draw up architectural drafts (for telecoms networks, including cable operators' 
networks) ?

I've found Visio to be a bit weak in this respect, even after adding third 
party stencils.

One product I'm exploring is ConceptDraw's "diagram" product, but I'd like to 
hear about anything you can share with me.

Cheers,

Etienne

-- 
Ing. Etienne-Victor Depasquale
Assistant Lecturer
Department of Communications & Computer Engineering
Faculty of Information & Communication Technology
University of Malta
Web. https://www.um.edu.mt/profile/etiennedepasquale

Re: The great Netflix vpn debacle! (geofeeds)

[Nimrod Levy]

On Wed, Sep 1, 2021 at 2:26 PM Michael Thomas  wrote:

>
> On 9/1/21 10:59 AM, Nimrod Levy wrote:
> > All this chatter about IPv6 support on devices is fun and all, but
> > there are providers still not on board.
> > They operate in my neighborhood and they know who they are...
> >
> This is about inside your premise before any NAT's enter the picture.
> What would be nice is if home routers offered up v6 as the default way
> to number and v6 tunnels past ISP's that don't have v6. Home routers
> could make that all rather seamless where users wouldn't need to know
> that was happening. It's really a pity that home routers are a race to
> the bottom where everything else with networking is expected to evolve
> over time.
>

I can't disagree about the quality of CPE, but I don't think that adding
tunnels by default is appropriate. We tried that with 6to4 and while that
worked, it didn't work well. Where would the far end of the tunnel
terminate? Who wants to build and manage that infrastructure? I'd rather
have the ISPs focus on deploying native IPv6 connectivity or at the very
worst, on-net 6rd. But I can tell you from experience that 6rd will only
take you so far before you figure out that you really needed native in the
first place.

Even more so, tunnels don't solve the problem that started this thread in
the first place. Netfilx (and probably others) consider IPv6 tunnel brokers
to be VPN providers and deny those connections. I stopped using a tunnel at
home for that very reason.

I think it's 100% appropriate for a CPE to not offer IPv6 on the inside
interfaces if it doesn't have a v6 upstream connection. What would the
point be?


> Mike
>

Re: Telecommunications network drafting software

[Fletcher Kittredge]

For physical plant (fiber) design, analysis and inventory, Vetro is really
useful.


On Wed, Sep 1, 2021 at 2:37 PM Etienne-Victor Depasquale via NANOG <
nanog@nanog.org> wrote:

> Hello folks,
>
> Would you care to share some pointers to drafting software which you use
> to draw up architectural drafts (for telecoms networks, including cable
> operators' networks) ?
>
> I've found Visio to be a bit weak in this respect, even after adding third
> party stencils.
>
> One product I'm exploring is ConceptDraw's "diagram" product, but I'd like
> to hear about anything you can share with me.
>
> Cheers,
>
> Etienne
>
> --
> Ing. Etienne-Victor Depasquale
> Assistant Lecturer
> Department of Communications & Computer Engineering
> Faculty of Information & Communication Technology
> University of Malta
> Web. https://www.um.edu.mt/profile/etiennedepasquale
>


-- 
Fletcher Kittredge
GWI
207-602-1134
www.gwi.net

Re: Happy 40th anniversary RFC 791!

[Michael Thomas]

On 9/1/21 11:42 AM, Mel Beckman wrote:

For anyone unaware, Jon Postel, a good friend and mentor to many of us at the 
dawn of the Internet, was the primary editor of this landmark document.

Those were the days we thought ARPAnet would never be allowed to go commercial. 
Thanks to Jon’s tireless campaigning (among others), not to mention meticulous 
documentation, it did.

He was taken from us too soon.

https://www.ietfjournal.org/in-memory-of-jon-postel/

RFC 791 and 793 are remarkably easy to read and follow what's going on. 
And that's not just hindsight speaking. We implemented both a few years 
later purely from the RFC's with no connection to anybody else and when 
we actually had kit that was connected to the Internet it just worked. 
One of the most remarkable things in my life was that I wrote debugger 
for the OS I wrote for the Lantronix terminal servers which I ended up 
remotely debugging in New Zealand. That was just mind blowing. Of course 
you'd call that a backdoor these days but they were very happy that I 
could figure out what was going on.


I never had the pleasure of meeting Jon, though I saw him at IETF meetings.

Mike

RE: The great Netflix vpn debacle! (geofeeds)

[Matthew Huff]

IPv6 tunnels work great for network geeks, but rather poorly for home users 
with streaming, gaming etc...It's not necessarily the performance, it's either 
the geolocation, latency, or the very issue that started this thread - VPN 
banning.

Remember, the streaming services couldn't care less about geolocation or VPN 
banning, it's the contractual obligations with the content providers. The 
content providers care about vpn banning because it gets around geolocation, 
which interferes with their business models (different release schedules to 
different regions, etc..)

Been there, done that...Stuck on Fios with no IPv6. Ran into rather 
"interesting" problems with various streaming services with IPv6 configured.


Matthew Huff | Director of Technical Operations | OTA Management LLC

Office: 914-460-4039
mh...@ox.com | www.ox.com
...

-Original Message-
From: NANOG  On Behalf Of Michael Thomas
Sent: Wednesday, September 1, 2021 2:26 PM
To: Nimrod Levy ; Owen DeLong 
Cc: nanog@nanog.org
Subject: Re: The great Netflix vpn debacle! (geofeeds)


On 9/1/21 10:59 AM, Nimrod Levy wrote:
> All this chatter about IPv6 support on devices is fun and all, but 
> there are providers still not on board.
> They operate in my neighborhood and they know who they are...
>
This is about inside your premise before any NAT's enter the picture. 
What would be nice is if home routers offered up v6 as the default way 
to number and v6 tunnels past ISP's that don't have v6. Home routers 
could make that all rather seamless where users wouldn't need to know 
that was happening. It's really a pity that home routers are a race to 
the bottom where everything else with networking is expected to evolve 
over time.

Mike

Re: Happy 40th anniversary RFC 791!

[Mel Beckman]

For anyone unaware, Jon Postel, a good friend and mentor to many of us at the 
dawn of the Internet, was the primary editor of this landmark document. 

Those were the days we thought ARPAnet would never be allowed to go commercial. 
Thanks to Jon’s tireless campaigning (among others), not to mention meticulous 
documentation, it did. 

He was taken from us too soon.

https://www.ietfjournal.org/in-memory-of-jon-postel/

-mel

> On Sep 1, 2021, at 11:20 AM, Michael Thomas  wrote:
> 
> 
> aka IPv4. The RFC doesn't have the exact date it was published, but the 
> internet as we know it was being born. What a journey it's been.
> 
> https://datatracker.ietf.org/doc/html/rfc791
> 
> Mike
> 

Re: The great Netflix vpn debacle! (geofeeds)

[Tom Beecher]

Televisions generally have a way smaller pixel density than a computer
monitor. It is very noticeable.

On Wed, Sep 1, 2021 at 2:27 PM  wrote:

>
> Every time I've read a thread about using TVs for monitors several
> people who'd tried would say don't do it. I think the gist was that
> the image processors in the TVs would fuzz text or something like
> that. That it was usable but they were unhappy with their attempts, it
> was tiring on the eyes.
>
> Maybe that's changed or maybe people happy with this don't do a lot of
> text? Or maybe there are settings involved they weren't aware of, or
> some TVs (other than superficial specs like 4K vs 720p) are better for
> this than others so some will say they're happy and others not so
> much?
>
> Or maybe the unhappy ones were all trolls/sockpuppets from companies
> manufacturing/selling $500+ 24" **GAMING** monitors.
>
> On September 1, 2021 at 09:48 nanog@nanog.org (Owen DeLong via NANOG)
> wrote:
>  >
>  >
>  > > On Aug 31, 2021, at 18:01 , Michael Thomas  wrote:
>  > >
>  > >
>  > > On 8/31/21 4:40 PM, Owen DeLong via NANOG wrote:
>  > >> On the other hand, the last time I went looking for a 27” monitor, I
> ended up buying a 44” smart television because it was a cheaper HDMI 4K
> monitor than the 27” alternatives that weren’t televisions. (It also ended
> up being cheaper than the 27” televisions which didn’t do 4K only 1080p,
> but I digress).
>  > >
>  > > Back when 4k just came out and they were really expensive, I found a
> "TV" by an obscure brand called Seiki which was super cheap. It was a 39"
> model. It's just a monitor to me, but I have gotten really used to its size
> and not needing two different monitors (and the gfx card to support it).
> What's distressing is that I was looking at what would happen if I needed
> to replace it and there is this gigantic gap where there are 30" monitors
> (= expensive) and 50" TV's which are relatively cheap. The problem is that
> 40" is sort of Goldielocks with 4k where 50" is way too big and 30" is too
> small. Thankfully it's going on 10 years old and still working fine.
>  >
>  > Costco stocks several 44” 4K TV models (like the one I got) that are
> relatively cheap. It’s a little larger than your 40” goldilocks, but I
> think still within range.
>  >
>  > Owen
>  >
>
> --
> -Barry Shein
>
> Software Tool & Die| b...@theworld.com |
> http://www.TheWorld.com
> Purveyors to the Trade | Voice: +1 617-STD-WRLD   | 800-THE-WRLD
> The World: Since 1989  | A Public Information Utility | *oo*
>

Telecommunications network drafting software

[Etienne-Victor Depasquale via NANOG]

Hello folks,

Would you care to share some pointers to drafting software which you use to
draw up architectural drafts (for telecoms networks, including cable
operators' networks) ?

I've found Visio to be a bit weak in this respect, even after adding third
party stencils.

One product I'm exploring is ConceptDraw's "diagram" product, but I'd like
to hear about anything you can share with me.

Cheers,

Etienne

-- 
Ing. Etienne-Victor Depasquale
Assistant Lecturer
Department of Communications & Computer Engineering
Faculty of Information & Communication Technology
University of Malta
Web. https://www.um.edu.mt/profile/etiennedepasquale

Re: The great Netflix vpn debacle! (geofeeds)

[Michael Thomas]

On 9/1/21 11:25 AM, b...@theworld.com wrote:

Every time I've read a thread about using TVs for monitors several
people who'd tried would say don't do it. I think the gist was that
the image processors in the TVs would fuzz text or something like
that. That it was usable but they were unhappy with their attempts, it
was tiring on the eyes.

Maybe that's changed or maybe people happy with this don't do a lot of
text? Or maybe there are settings involved they weren't aware of, or
some TVs (other than superficial specs like 4K vs 720p) are better for
this than others so some will say they're happy and others not so
much?


It's been a while but there was a setting for mine that I had to futz 
with so that didn't happen. You're right that you should definitely check.


Mike

Re: The great Netflix vpn debacle! (geofeeds)

[Michael Thomas]

On 9/1/21 10:59 AM, Nimrod Levy wrote:
All this chatter about IPv6 support on devices is fun and all, but 
there are providers still not on board.

They operate in my neighborhood and they know who they are...

This is about inside your premise before any NAT's enter the picture. 
What would be nice is if home routers offered up v6 as the default way 
to number and v6 tunnels past ISP's that don't have v6. Home routers 
could make that all rather seamless where users wouldn't need to know 
that was happening. It's really a pity that home routers are a race to 
the bottom where everything else with networking is expected to evolve 
over time.


Mike

Re: The great Netflix vpn debacle! (geofeeds)

[bzs]

Every time I've read a thread about using TVs for monitors several
people who'd tried would say don't do it. I think the gist was that
the image processors in the TVs would fuzz text or something like
that. That it was usable but they were unhappy with their attempts, it
was tiring on the eyes.

Maybe that's changed or maybe people happy with this don't do a lot of
text? Or maybe there are settings involved they weren't aware of, or
some TVs (other than superficial specs like 4K vs 720p) are better for
this than others so some will say they're happy and others not so
much?

Or maybe the unhappy ones were all trolls/sockpuppets from companies
manufacturing/selling $500+ 24" **GAMING** monitors.

On September 1, 2021 at 09:48 nanog@nanog.org (Owen DeLong via NANOG) wrote:
 > 
 > 
 > > On Aug 31, 2021, at 18:01 , Michael Thomas  wrote:
 > > 
 > > 
 > > On 8/31/21 4:40 PM, Owen DeLong via NANOG wrote:
 > >> On the other hand, the last time I went looking for a 27” monitor, I 
 > >> ended up buying a 44” smart television because it was a cheaper HDMI 4K 
 > >> monitor than the 27” alternatives that weren’t televisions. (It also 
 > >> ended up being cheaper than the 27” televisions which didn’t do 4K only 
 > >> 1080p, but I digress).
 > > 
 > > Back when 4k just came out and they were really expensive, I found a "TV" 
 > > by an obscure brand called Seiki which was super cheap. It was a 39" 
 > > model. It's just a monitor to me, but I have gotten really used to its 
 > > size and not needing two different monitors (and the gfx card to support 
 > > it). What's distressing is that I was looking at what would happen if I 
 > > needed to replace it and there is this gigantic gap where there are 30" 
 > > monitors (= expensive) and 50" TV's which are relatively cheap. The 
 > > problem is that 40" is sort of Goldielocks with 4k where 50" is way too 
 > > big and 30" is too small. Thankfully it's going on 10 years old and still 
 > > working fine.
 > 
 > Costco stocks several 44” 4K TV models (like the one I got) that are 
 > relatively cheap. It’s a little larger than your 40” goldilocks, but I think 
 > still within range.
 > 
 > Owen
 > 

-- 
-Barry Shein

Software Tool & Die| b...@theworld.com | http://www.TheWorld.com
Purveyors to the Trade | Voice: +1 617-STD-WRLD   | 800-THE-WRLD
The World: Since 1989  | A Public Information Utility | *oo*

Happy 40th anniversary RFC 791!

[Michael Thomas]

aka IPv4. The RFC doesn't have the exact date it was published, but the 
internet as we know it was being born. What a journey it's been.


https://datatracker.ietf.org/doc/html/rfc791

Mike

Re: The great Netflix vpn debacle! (geofeeds)

[Nimrod Levy]

All this chatter about IPv6 support on devices is fun and all, but there
are providers still not on board.
They operate in my neighborhood and they know who they are...

Nimrod

Re: The great Netflix vpn debacle! (geofeeds)

[Owen DeLong via NANOG]

> On Aug 31, 2021, at 18:01 , Michael Thomas  wrote:
> 
> 
> On 8/31/21 4:40 PM, Owen DeLong via NANOG wrote:
>> On the other hand, the last time I went looking for a 27” monitor, I ended 
>> up buying a 44” smart television because it was a cheaper HDMI 4K monitor 
>> than the 27” alternatives that weren’t televisions. (It also ended up being 
>> cheaper than the 27” televisions which didn’t do 4K only 1080p, but I 
>> digress).
> 
> Back when 4k just came out and they were really expensive, I found a "TV" by 
> an obscure brand called Seiki which was super cheap. It was a 39" model. It's 
> just a monitor to me, but I have gotten really used to its size and not 
> needing two different monitors (and the gfx card to support it). What's 
> distressing is that I was looking at what would happen if I needed to replace 
> it and there is this gigantic gap where there are 30" monitors (= expensive) 
> and 50" TV's which are relatively cheap. The problem is that 40" is sort of 
> Goldielocks with 4k where 50" is way too big and 30" is too small. Thankfully 
> it's going on 10 years old and still working fine.

Costco stocks several 44” 4K TV models (like the one I got) that are relatively 
cheap. It’s a little larger than your 40” goldilocks, but I think still within 
range.

Owen

Re: The great Netflix vpn debacle! (geofeeds)

[Owen DeLong via NANOG]

> On Aug 31, 2021, at 17:51 , Michael Thomas  wrote:
> 
> 
> On 8/31/21 5:13 PM, Jay Hennigan wrote:
>> On 8/31/21 16:32, Jeroen Massar via NANOG wrote:
>> 
>>> Fun part being that it is hard to get a Dumb TV... though that is primarily 
>>> simply because of all the tracking non-sense in them that makes them 
>>> 'cheaper'... (still wonder how well that tracking stuff complies with GDPR, 
>>> I am thinking it does not ... Schrems anyone? :) )
>> 
>> Just get a "smart" TV, don't connect it to the Internet, and use its HDMI 
>> ports for your cable box, Apple TV, etc. and/or antenna input for local 
>> off-air reception.
>> 
> 
> Yeah, until TV manufacturers actually start incorporating, oh say, Google tv 
> (which is just a form of Android) they are always going to be inferior. 
> Having the TV just be a monitor is a feature, not a bug. It's a lot cheaper 
> to upgrade a $50 hdmi based dongle than the whole TV, doubly so since 
> manufacturers have a bad reputation  for not supporting upgrades beyond the 
> sell date. I have no idea whether any of the external ones support v6 though.

Apple TV supports IPv6, but does not allow the user to set a static IPv6 
address and it uses rotating privacy addresses, so the security implications 
are “interesting”. OTOH, it does appear to support DHCPv6 and if you set M+O, 
it looks like you can collect the DUID and give it a fixed DHCP address.

Android and by extension Google’s HDMI dongles/devices have some IPv6 support, 
but of course don’t work with DHCPv6 because of Lorenzo’s religious problems.

> One thing that might be nice is for routers to internally number using v6 in 
> preference to v4 and NAT that (if needed). Then you can easily tell what is 
> still a laggard. My wifi cams might be poorly supported, but they don't need 
> to interoperate with much on the Internet.

I actually have had an idea for a long time of producing a router-on-a-stick 
kind of device which would be a small linux SBC with two ethernet ports and 
some LEDs.

The OS would go on a micro-SD card and it would literally be a single-device 
NAT64 setup so that the IPv4-only device on the downstream side could work with 
the IPv6-only LAN (which might further have a NAT64 gateway to deal with the 
IPv4-only legacy portions of the world outside.

Ideally, the upstream ethernet port would be PoE to power the device (and the 
device would be sold with a small, cheap PoE injector in case needed).

> Mike, Google TV has been pretty nice since the Amazon feud finally ended 
> though I hate that the protocol is still pretty proprietary

To the best of my knowledge, the FireTV and its ilk still can’t spell IPv6.

Owen

Re: The great Netflix vpn debacle! (geofeeds)

[Lyden, John C]

> But I also agree there are limits to what needs IoTing. I don't live in a
> house large enough that I can't go see if the box needs cleaning within
> about 20s. I also sure as hell don't need a notification on my phone that
> one of them just made a deposit.

*ding!* 1 New Notification: "clean my filth, human servant."

My wife has been campaigning for one for about 2 months now. I'm unsure how 
it'll do with 4 of varying age from 1 to 16 though. Still tempting.

John C. Lyden
Manager of Network Infrastructure
Infrastructure Services, Division of Information Resources & Technology 
Rowan University 

Re: An update on the AfriNIC situation

[Valerie Wittkop]

REMINDER from the admins… 

Please ensure your posts are topical, do not go against the Usage Guidelines 
, and 
provide for an  exchange of technical information and the discussion of 
specific implementation issues that require cooperation among network service 
providers.

The Mailing List is not an appropriate platform to resolve personal issues, 
engage in disputes, or file complaints.

Also, please ask yourself these three questions before posting -
Does my email have operational/technical content?
Would I be interested in reading this email?
Would 12,864 other Internet engineers want to read this?

And in case you missed our first message among the many messages. In this 
thread: 

This thread has had a few twists and turns, as well as ups and downs. So I am 
taking a moment on behalf of the admins to remind everyone here of the Usage 
Guidelines 
.

First and foremost - please report to adm...@nanog.org 
 a concern about violation of the guidelines. Using 
the list to argue about violations is not appropriate. 

Secondly - there have been entirely too many messages in this thread that are 
in violation of guideline number 6, specifically with comments that are 
defamatory, abusive, or lack respect for other participants. 

Finally - the NANOG mail list is not the place to conduct a trial. (See 
guidelines number 8.)

Should you have any questions/concerns about this reminder, please send a 
message to adm...@nanog.org  

Valerie Wittkop
Program Director
vwitt...@nanog.org | +1 734-730-0225 (mobile) | www.nanog.org
NANOG | 305 E. Eisenhower Pkwy, Suite 100 | Ann Arbor, MI 48108, USA
ASN 19230

Re: An update on the AfriNIC situation

[Brielle]

On Sep 1, 2021, at 5:56 AM, Mark Tinka  wrote:
> 
> 
> 
>> On 9/1/21 13:21, Tom Beecher wrote:
>> 
>> 
>> There are enough challenges with the internet in Africa to work through 
>> already. We shouldn't encourage more difficulties by endorsing strongarm 
>> tactics that prevent issues from being properly adjudicated in courts.
> 
> One would think...
> 
> There are many African-based and African-led organizations that would go 
> toe-to-toe with AFRINIC over any number of legal disagreements. But there is 
> not a single African-based and African-led organization that would ever cut 
> off AFRINIC's supply just for their own benefit, or to prove a point. None.
> 
> I can't even say that it would impose less stress on us, Africans, if it was 
> one of our own shooting ourselves in the foot in this way, because we do not 
> have a culture of going against each other in the way that we are currently 
> witnessing. It's not in our DNA.
> 
> That a non-African entity and its posse are stumping their feet on the 
> African continent in this way, and making poor people even poorer, is as 
> diseased as it is rotten.

As an American, holding legacy ARIN resources, there are times where I disagree 
with how they do things or how they’ve operated in the past….

However, I’d never be that daft to file a lawsuit against them with the obvious 
intention of trying to financially ruin/bankrupt them to get what I want, even 
if I was 100% within the legal right.

I’d be internet enemy #1.  I’d expect to be the target of nasty shit directed 
my way.  I’d expect my netblocks to be globally black holed to all hell.  I’d 
deserve every bit of it.

People have long memories for things like this.  

I still harbor resentment towards NetSol over their hijacking of all 
non-existing domains to their site in the 2000s…

Re: An update on the AfriNIC situation

[Mark Tinka]

On 9/1/21 13:21, Tom Beecher wrote:



There are enough challenges with the internet in Africa to work 
through already. We shouldn't encourage more difficulties by endorsing 
strongarm tactics that prevent issues from being properly adjudicated 
in courts.


One would think...

There are many African-based and African-led organizations that would go 
toe-to-toe with AFRINIC over any number of legal disagreements. But 
there is not a single African-based and African-led organization that 
would ever cut off AFRINIC's supply just for their own benefit, or to 
prove a point. None.


I can't even say that it would impose less stress on us, Africans, if it 
was one of our own shooting ourselves in the foot in this way, because 
we do not have a culture of going against each other in the way that we 
are currently witnessing. It's not in our DNA.


That a non-African entity and its posse are stumping their feet on the 
African continent in this way, and making poor people even poorer, is as 
diseased as it is rotten.


Mark.

Re: An update on the AfriNIC situation

[Mark Tinka]

On 9/1/21 00:56, Owen DeLong via NANOG wrote:


Not to put too fine a point on this, but what human cost?

There were exactly 3 employees that AFRINIC wasn’t able to pay in July, 
including
the CEO (who is one of the major protagonists in creating this problem in the 
first
place). I don’t know who the other two were.

Everyone else got paid for July.

AFRINIC has received clearance of enough money to cover their normal expenses
for August and September. As such, there shouldn’t be any problems with salaries
or “human cost” in those months. Hopefully given that reprieve, cooler heads at
AFRINIC can prevail and some form of settlement can be achieved before they run
out of money from that reprieve.


This is rich!

Of course, one should expect people to be mentally settled and do good 
work when they have no security or clarity about whether they will get 
paid at the end of each month.


These aren't robots, mate. People have real thoughts and real feelings.

Stress is not intangible... it releases cortisol, which delivers a 
number of physiological side effects that work against good health. 
Financial insecurity is just about the worst stress anyone has to deal with.


None of us would sleep well if we knew that our source of income is not 
guaranteed, despite what some may say about "As such, there shouldn't be 
any problems with salaries..."


Mark.

Re: An update on the AfriNIC situation

[Mark Tinka]

On 8/31/21 22:55, Sabri Berisha wrote:



I regret the true human cost that Mark pointed out, yet I am fascinated
by the case and the arguments on both sides. The court will have their
work cut out for them.


The human cost has nothing to do with the wording of allocation 
language. That was just pure posturing.


Mark.

Re: An update on the AfriNIC situation

[Mark Tinka]

On 8/31/21 22:37, Rubens Kuhl wrote:


I can try helping with that: in underserved regions it's not unusual
for network services for that population to be physically hosted out
of the region. For instance, if you have a hosting service that only
accepts South African rands and your language options are Afrikaans
and Zulu, you can credibly argue to AfriNIC that you are targeting its
service region and are eligible for AfriNIC number resources.

But you would need to be upfront with that, including mentioning that
your upstreams are not from Africa and your installations won't be in
Africa.
Otherwise you applied for number resources under false pretenses, and
will bear the risk of such.


I believe CI had a small footprint in South Africa in some rack. 
Certainly nothing that would bother a /29.


Mark.

Re: An update on the AfriNIC situation

[Mark Tinka]

On 8/31/21 22:28, Sabri Berisha wrote:



It's easy to argue that CI is in full compliance with that since their
assignment supports connectivity between users in Africa and their clients'
services. In that case, only IP space used outside of Africa not advertised
to the internet would be in violation.


AFRINIC's intention with that language is that those IP addresses need 
to be assigned for use by eyeballs and operations in Africa. Of course, 
the Internet is a two-way street, but the IP addresses from which the 
traffic is sourced, in this case, need to be used inside the continent 
of Africa.


Mark.

Re: The great Netflix vpn debacle! (geofeeds)

[Tom Beecher]

100% the Litter Robot is amazing. ( Except for my older cat, she's pushing
19, had to build a ramp for her. )

But I also agree there are limits to what needs IoTing. I don't live in a
house large enough that I can't go see if the box needs cleaning within
about 20s. I also sure as hell don't need a notification on my phone that
one of them just made a deposit.

On Wed, Sep 1, 2021 at 12:05 AM Jay Hennigan  wrote:

> On 8/31/21 20:18, J. Hellenthal wrote:
> > Also don't get a smart litterbox... ;-)
> >
> > Yeah that's a thing and connects to the local Wi-Fi. Kinda want to DMZ
> that mutha and wait for a script kiddie to turn one of my cats upside
> down...
> >
> > dubs litter-robot.com
>
> I have one, the cat loves it and it's very easy to clean. No need to
> enable the wi-fi. Front panel indicators are more than sufficient.
> "Wi-fi enabled" on things that don't need wi-fi is a marketing gimmick
> that's way over-used.
>
> Rule 37.024 subsection 7: Cats are always on-topic.
>
> --
> Jay Hennigan - j...@west.net
> Network Engineering - CCIE #7880
> 503 897-8550 - WB6RDV
>

Re: An update on the AfriNIC situation

[Tom Beecher]

>
> AFRINIC has received clearance of enough money to cover their normal
> expenses
> for August and September. As such, there shouldn’t be any problems with
> salaries
> or “human cost” in those months. Hopefully given that reprieve, cooler
> heads at
> AFRINIC can prevail and some form of settlement can be achieved before
> they run
> out of money from that reprieve.
>

It's good that people are still being paid.

That being said, while some may have the opinion that AFRINIC's actions
have been 'objectionable' , others have the opinion that their actions were
justified and proper. Does it not concern you at all that AFRINIC may be
forced into a 'settlement' because they cannot access their funds due to a
very dubious claim of damages?

There are enough challenges with the internet in Africa to work through
already. We shouldn't encourage more difficulties by endorsing strongarm
tactics that prevent issues from being properly adjudicated in courts.

On Tue, Aug 31, 2021 at 6:59 PM Owen DeLong via NANOG 
wrote:

> >> I regret the true human cost that Mark pointed out, yet I am fascinated
> >> by the case and the arguments on both sides. The court will have their
> >> work cut out for them.
> >
> > That human cost came not from disagreement on the policies and
> > contract provisions, but from a vengeful action of financial bullying.
>
> Not to put too fine a point on this, but what human cost?
>
> There were exactly 3 employees that AFRINIC wasn’t able to pay in July,
> including
> the CEO (who is one of the major protagonists in creating this problem in
> the first
> place). I don’t know who the other two were.
>
> Everyone else got paid for July.
>
> AFRINIC has received clearance of enough money to cover their normal
> expenses
> for August and September. As such, there shouldn’t be any problems with
> salaries
> or “human cost” in those months. Hopefully given that reprieve, cooler
> heads at
> AFRINIC can prevail and some form of settlement can be achieved before
> they run
> out of money from that reprieve.
>
> > I saw my quota of questionable court decisions to automatically agree
> > with whatever is decided in this case, even if CI loses, but the
> > arguments from both sides will indeed be very interesting and useful
> > to close out loopholes in the system.
>
> Only if they are ever able to be made public, which is a little iffy given
> the Mauritian
> court system. It may well be that only the final ruling is able to be made
> public.
>
> Owen
>
>