Theoretical questions:

[J Brown]

[ This needs a movie treatment ]

What percentage of the internet would be permanently lost if there was a one minute, total electrical blackout worldwide?

Possible considerations;  yes we do have battery backups through-out the world; is that enough to save it all?  Some percentage of the backups would fail.

Any data in transmission might be permanently lost.

Some small percentage of servers would die;  Some percentage of harddrives/flashdrives would never reboot.  

If even one significant/crcuial part of the internet mapping became permanently lost; how castastropic might be its impact. 

Brazil's electrical supply/reliability are apparently problematic.  If it took them 1 month to reboot their system, what impact would it have upon the rest of the world?  If they couldn't restart their electricity for one month; I would assume all of their battery back up would have died at some point.

Anyhow, it could make for a good movie.

[Tim Holloway]

One Measly Minute? Pffft!

Starting from home, the mousetech.com servers can run for about 20
minutes on UPS batteries. I just replaced one, incidentally. The UPS
monitors battery health and screams bloody murder about every 4 hours
(including in the early AM) until you make it happy. I've got monitors
jacked in to cleanly shut down the machines once the batteries get down
to 10%.

Most of the Internet is either TCP or UDP traffic. UDP is for data you
don't CARE if you lose, TCP is restartable. The only real network data
you'll lose is stuff that deals with non-UPS servers and clients and
even then, often it takes more than a minute for them to flake out.
Especially laptops with batteries in them.

As I mentioned in passing to Ralph the other day, I have solar
batteries that run Raspberry Pi's.

In fact, the biggest hole in the transmission scheme is Comcast. Unlike
AT&T landline phones that run out of buildings packed with batteries,
the relay points for local Comcast have no emergency power capabilities
to speak of. However, our backup Internet is to switch to a cellphone
access point and those do seem to carry backup power.

So for a multi-DAY power outage, servers from mousetech.com would be
offline, though internally we'd still be able to talk to sites with a
larger investment in continuous power.

No, the Wikipedia wouldn't get erased, nor alas, even Twitter.

The biggest losses would likely come from systems which had data in RAM
that hadn't flushed out to disk where the machine didn't have enough
power reserves to handle it cleanly.

Remember, the Internet was designed to withstand nuclear war. You can
lost parts of the system - for example the state of Texas (ERCOT) - but
the rest of the world would chug on just fine and a global 1-minute
outage would hardly be enough to annoy most people.

Tim

> --
> You received this message because you are subscribed to the Google
> Groups "JaxLUG" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to jaxlug+un...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/jaxlug/CAEgYwWBrGqycGDercMF7CzABGF4wRxkB6UMry1eL9rPZzvSd-w%40mail.gmail.com
> .

[J Brown]

Tim, thank you for your response;

Is there any scenario you can conceive of that would set the internet back for a period of years?  What is the ultimate disaster?

To view this discussion on the web visit https://groups.google.com/d/msgid/jaxlug/9b8b0eb0103ad9569ae5589d19e5338b604842b4.camel%40gmail.com.

[Tim Holloway]

Well, yes. Rod Serling time:

Imagine, if you will, a world infected with a global pandemic. One so
severe that factories shut down, employees are quarantined (or die),
transport systems are disrupted, making it impossible to manufacture
and deliver new network communications devices.

It would take a decade or more for all the existing network devices to
fail, though if you were to throw in disruptive solar storms to "EMP"
critical nodes, that could accelerate things. To get data from one
place to another post-breakdown you would have to physically commit it
to paper and ship it via low-speed channels like postal services and
carrier pidgeons. At least until all the printers broke down and the
disk/tape/optical archives failed, since they, too depend on
microchips.

You could, of course, re-create the circuits using transistors, but
making reliable transistors is also a thing that few of us can do in a
garage. Vacuum tubes and relays are easier, but bulkier and very power
hungry. Much of the compression and error recovery we're used to would
have to be tossed to keep the complexity down.

We didn't actually see much of this with the actual pandemic because
the necessary technology wasn't affected, but automobile circuits were
HEAVILY impacted, and the Raspberry Pi became an expensive and coveted
commodity. I have a client in Romania using Pi's and we had to scramble
to find alternatives.

It's funny where the technological lines got drawn. I used to think
that surface-mount PCBs required a factory, but now I have the
necessary stuff to do it at home - including Linux-based design and
simulation software, and in fact, I bought an AM/FM/shortwave radio
chip for about $10 that as soon as I can find where I lost it is going
to get soldered to an SMT board to be run by a Pi or Arduino device.

ICs, on the other hand, are much harder. There's a guy in California
making basic IC's in his garage for fun, but even simple logic gate
chips require a lot of fiddly stuff to refine the silicon, a store of
toxic chemicals to dope them and pressure vessels/heaters to manage the
deposition process. To do that at the densities required by many modern
chips means you'd better have a billion dollars or so on your Wal-Mart
gift card and a couple of acres to put everything on.

[J Brown]

very interesting Tim

To view this discussion on the web visit https://groups.google.com/d/msgid/jaxlug/b9dbc25a493a4cdb8b8027cf4d420f7d481f2e10.camel%40gmail.com.

[Russell Grokett]

When I first started working at AT&T in the 1980's, the network was quite hardened against disasters. The equipment buildings have 3 ft thick walls, few windows, diesel generators for a month, even fallout shelters and food for employees...  but by the time I retired a decade ago, hardening was much less. Maybe a week, while many cell towers just a few days if that, mainly due to so many to support. They do have quick response teams for "limited" areas like hurricanes, earthquakes, etc. but for the entire country? No.    

Likely EMP from a space nuke or X-class solar storm would be the nastiest these days. The cell receivers, satellites and electric grid would be hit. See the Canadian X flare of 1989 https://en.wikipedia.org/wiki/March_1989_geomagnetic_storm  and the US nuke EMP test Starfish Prime in 1962.  https://en.wikipedia.org/wiki/Starfish_Prime  both were long before cellular and even most computing.  

There are amateur radio groups that keep transceivers locked in metal boxes to shield them. They would be pulled out in emergencies and powered by lead-acid batteries. (Most generators and even Li-ION batteries have electronics that may or may not survive a really big EMP.)  Only the largest, direct hit X-class or a global nuke war would really be a bad day for all humanity.  Anything less would be a bad day for a mere "fraction" of people, maybe a few hundred million or so. 

Of course, full scale nuke war just bounces the rubble. No need of Internet and things like a can of Spam would be worth more than gold or bitcoin. ;-)      You can buy an empty minuteman missile silo for about the price of a house. Kinda cool, but GF's and wives might disagree.  

See "Nuclear EMP Scenarios" https://apps.dtic.mil/sti/pdfs/AD1097009.pdf for some really "crazy fun" (not) stuff. Maybe Elon needs to push to Mars a bit faster.

...then there's asteroids.  

Russell

[Tim Holloway]

What to do after the Apocalypse is a popular fictional trend. One of
the works I like best is "Alas, Babylon", which was written by a local
resident (Pat Frank) and set in a mythical small town that insofar as
it could credibly exist geographically would have been located close to
my grandfather's house in North Central Florida. Frank's in-laws ran
the legendary Flying Dragon citrus nursery here in town.

On a wider scope - and actually covering the rise and fall of an
electronic/atomic post-apocalyptic world was Walter Miller's Canticle
for Leibowitz. Miller was a resident of Daytona Beach.

The popular fantasy of the ammosexual is that whoever has the most guns
will rule the world when it all collapses. Then there are the gold
hoarders, who think that gold has an Absolute value that fiat currency
doesn't. Of course, the people with the guns expect to take the gold.

In actual fact, I'm beginning to believe that whoever has the biggest
stockpile of AA batteries will rule.

Speaking of absolute value, take about $1.75 of sand and various
pollutants. Mix well, produce a flat-screen TV and sell it for $500.
Then hit it with a hammer. It's worth $1.75 all over again. Context is
everything.

[Tim Holloway]

Over on the Javaranch in one of the more private forums we've had some
Doomsday discussions thanks to all the recent solar activity. IIRC,
there was even a chance of seeing the Aurora Borealis in darker parts
of North Florida recently.

As I understand it, tube-based electronics are supposed to be highly
resistant to EMP and quite a few short/longwave amateurs prefer tube
equipment to this day. I'm actually suspicious that micro-circuits
would be all that susceptible actually. Here's why.

EMP is a concentrated burst of radio energy. Its primary ability to
damage comes from being able to induce lightning-level electrical
currents in circuits not intended to withstand them. Tubes operate at
high voltages and air/vacuum gaps, so they're pretty durable. One of
the worst effects of the 19th-Century Zap was that when you have miles
and miles of copper telegraph wire running across the country you have
one big honkin' set of antennas to harvest and channel that radio
energy. Copper is still common, but fiber optics are also and their
weak spots are at the relay points, not the entire (non-conducting)
fiber.

Microcircuits are low voltage, but actually remarkably durable these
days, which is good since I used to fry more components than I like
just by getting 2 wires in the wrong place. I almost never toast modern
stuff.

More importantly, a target the size of a grain of sand makes a very
poor radio antenna for any frequency likely to be EMP-intense. So in
short, as long as the circuit isn't attached to any significant wiring
(including power supply leads!) then I'm not too worried. If a part
blows, I'd dig another out of the drawer and see how that worked. It
might even be safer NOT to keep them in metal boxes, as in an EMP
environment, a Faraday Cage would likely see significant eddy currents.

A lot of scare has been published about EMP, but one of the biggest
worriers about stuff like that is the military and not to be a
conspiracy theorist, but I don't think that they'd consider it in their
interest to be too precise about the actual vulnerabilities or lack
thereof.

We no longer harden against a Soviet nuclear attack. In fact, we've
about reached the point where a lot of people would welcome Russian
invasion with open arms, and I don't mean the Atheistic Satanist Commie
Libruls. I think once we reached M.A.D. levels a lot of people didn't
see any point in bunker construction anymore.

[J Brown]

Solar flares and car batteries?

Is it true that if [ let's say ]  Duval County took a big enough direct hit from a solar flare that it could cook our car batteries - basically "totalling" our cars?  Is it true you have about 10 minutes to uncable the batteries?   Anybody know?

To view this discussion on the web visit https://groups.google.com/d/msgid/jaxlug/c1a424b3b1d2a879d83005019fa04aa9c1e0d588.camel%40gmail.com.

[J Brown]

A big enough flare would reverse the polarity and at that point you're screwed?

[Tim Holloway]

A lead-acid automobile battery is internally a set of cells each of
which forms something more or less resembling a loop.

If you hit one with a microwave-oven level of radiation, it would
probably boil the electrolyte and blow the thing up. Then again, if you
get hit with microwave-level radiation on a global scale, you'd
probably blow up yourself.

I doubt that specifically you'd reverse the vehicle polarity with a
pulse. For one thing, that's not really in keeping with the overall
wiring pattern unless you get a very specifically polarized wave coming
in. However, EMP is well-accepted as a way to halt a vehicle in short
order. It would interfere with the spark timing, and likely blow out
the timing computer. This might actually be less of a problem for the
old-time mechanical units (assuming the coil doesn't short), but either
way expect the engine to stall at a minimum.

Modern cars contain multiple data buses, all of which can become
antennas for EMP feeding low-power chips. Plus, of course the power
circuits for head and tail lights and so forth. I'm not big on random
electrical explosions. For example, the bridge of the USS Enterprise,
since they're control circuits and thus likely to be very low
voltage/current. When was the last time a graphics card blew up for
you? But I could definitely see sealed-beam headlights exploding and
don't even THINK about LEDs. Likely other odds and ends, such as having
the starter motor relay pull in while the engine was running, but I've
only theoretical speculation. You could also potentially arc-weld
moving parts together.

If you have a Tesla, odds are that the motor windings would fry. Best
bet is a Stanley Steamer, I think.

[J Brown]

Tim, I was under the impression that the older the vehicle; the more likely it is to survive - as you suggested.   What's a Stanley Steamer going for these days?  smile

To view this discussion on the web visit https://groups.google.com/d/msgid/jaxlug/8ef7009b10c8dd890ca015fea5c6f4558b3400fb.camel%40gmail.com.

[Tim Holloway]

Within limits. Obviously if there are no control electronics or
electrics, that limits what an EMP can do. On the other hand, as I said
before, one reason why the 19th-Century Zap was so serious was that it
induced and propagated over long lengths of telegraph wire. Modern
communications often don't go for more than a few miles before the
copper changes to fiber or radio or vice versa, so you'd have to zap
multiple disparate segments.

In fact, I tried to get ADSL Internet several years back. The splitter
is still on my wall, but the service failed because there's fiber
between my land line wiring and the local switching office and ADSL
required a solid copper path all the way.

Fun fact: Electric automobiles pre-date petroleum-product ones by many,
many years. In fact, they're almost as ancient as steam locomotive rail
systems. People were playing with electric cars more than 30 years
before Nicola Tesla was born.