This site is no longer solar powered for now
andrewjvpowell.com
andrewjvpowell.com
When resources are scarce, interesting stuff happens. Nowadays most outages are either due to expired payment, disk full, or provider outage. That's a good thing I guess? Just much less fun.
1200 bps Unix shell access from TENet (Texas Educators’ Network) - thank you, Mr. Horner, my 9th grade US History teacher, for telling that nerdy girl who turned in all her papers in dot matrix print about what she could get her mother, a fellow teacher, to pay all of $5/year for.
Oh, and daring to have us look at Reconstruction’s sudden cut-off, Jim Crow and the fact that we were less than 30 years from its legal end in Texas, the Bonus Army, our rather late entry into WW2, the Japanese-American internment camps, and various shenanigans to do with the Vietnam War - you taught us that it’s possible to both love your country and to accept that it is by no means perfect.
I hope you’re enjoying a good retirement.
I have tiny recollection of this this but thanks to your comment I became acquainted with it again.
Thanks!
I still have ADSL. It's now 5 Mbps, but I don't recall what it was when dial up was common.
Higher speeds of DSL came along with ADSL2 in 2002/2003. There were higher than 1.5mbps before that, but you basically had to be across the street from the CO to get that.
Do you get cell service? We don't and that's actually the bigger hassle. Wi-Fi calling is kind of shit, and the coverage outside the house is appallingly unreliable. iMessage is frustratingly unreliable too. I'll be on Wi-Fi, but it won't have gotten synched up with Verizon and it'll refuse to send a message until it does.
We do have cell service. We're in an older 1950s development that was never rewired, but not far from newer developments.
The biggest problem for me is that downloads keep getting bigger and bigger. The latest AAA games are almost 100 GB and take days to download.
And either way, even that coverage is spotty and speed varies highly on the time of day that you're trying to use it.
All this is fine for a tertiary backup solution (behind Starlink) but not as a primary method of accessing the Internet.
That was the default speed for a long time.
The house's floor slab had a crack that wicked up groundwater when it rained too much, so my carpet would get mushy. Electronics off!
The anti-cheat software I ran (PunkBusters) went crazy after an update and banned anybody who connected. I noticed when my MRTG graphs flatlined.
Girlfriend once yanked out the Ethernet cable when we were having adult time. I didn't bother explaining that one to the players.
Heavy web browsing would cause lag, so I ended up learning how to do QoS via iptables (or its precursor, whose name I forgot).
At the time, the server used an astonishing 25 gigabytes of bandwidth a month. Always raised an eyebrow with the ISP tech when I called in, but they never complained.
Thanks for reminding me I'm getting too old. :V
They had a message up for a while that said they were unavailable because their servers were either underwater, or on the back of a looter's truck that was itself underwater.
It seems nearly everyone goes through a “How hard can it be?” phase where self-hosting seems like it will be a trivial way to save a lot of money. Then you get to learn about all of the things that can, and will, go wrong over time. Great learning experience.
And then there are the rose-colored glasses people who seem to forget all of the troubles they go through with self-hosting. A few companies back we had a guy who insisted on self-hosting some things on premises to save on SaaS costs. The amount of time we spent waiting for him to figure out why the server was misbehaving each week easily cost several times more than any hypothetical savings. Yet when you ask him, he’d claim it was virtually hassle-free. Management liked it because it “saved a lot of money”… as long as you ignore the cost of the time we lost to dealing with it.
I wonder how much we spend babysitting slop (either directly or indirectly through SaaS) vs how much it would cost to invest in engineering around automated and reliable deployment and resilient software?
Highly polished distributed file systems, databases, and orchestration tools that incorporated automatic replication and fail over would be a great start.
Don’t just say “it’s hard.” Of course it is. But is it cheaper to do this or to babysit rickety piles of junk?
Another possible way out is AI sysadmins. I wonder how far we are from AIs that can admin a cluster including upgrades and disaster recovery?
I guess the reason it doesn't make sense is that you're living in your own imagination where nothing is obligated to be sensical!
The current boot image (vmlinuz) on my debian system is 6MB and the initrd is 40MB.
> 11M Apr 6 12:14 vmlinuz-6.1.19-gentoo
> 4.6M Apr 6 12:14 System.map-6.1.19-gentoo
And that's a full desktop QEMU "server" - and i don't use initrd, because i don't need every driver in the tree, just the ones that work for my system. No AI needed. You boot a "kitchen sink" kernel, lspci -k, note down what it says, then just enable those things in the kernel and build it.
Even ubuntu being "quite supportive" is 30-45MB.
The kernel source tree is over a gigabyte, but that has to support everything it supports - arm, risc-v, x86_64, etc.
> 2.2G linux-6.1.19-gentoo/
I mean, no other industry insists that users accept weekly automated updates because their products have so many critical security bugs.
bean counters tend to not evaluate your time as lost money. they are paying you anyways, so sunk cost fallacy or something. as long as they are not having to pay an invoice, they are "saving" money
SaaS mostly exists to extract profits perpetually.
This doesn't mean it's not sometimes worth it, but when you add all the licenses together, there tends to be considerable savings if you're willing to self-host.
In reality both sides of the debate are often wrong.
The $10 option was genuine though, it was for some SaaS tool to monitor endpoints and text if things broke (about 9 years ago when I had basically zero coding experience) - so just one seat, or me trying to hack something together with text/email notifications and a nice UI. Even now I'd probably still prefer using something like that over writing something in-house, although I'd be more happy to look at self-hosting an open source option.
we spent more time trying to make an external tool
do something that could have been built in house in
the same or less time.
There's also a version of the "nobody ever got fired for choosing IBM" fallacy/truism at work, right? If management picks $WELL_KNOWN_PRODUCT_XYZ and things go sideways... well, they don't look foolish. But if they rely on... some random in-house engineer and things don't work out, they look like fools.
(Even though, in both cases, it actually just comes down to the same in-house engineers. Because the in-house engineers still need to do the integration work)
No decent engineer would stay there because there was nothing to build and management would always just contract out everything. Management, which was promoted internally from bad engineers, couldn't shop a good SaaS product or consulting firm and instead were consistently convinced into buying shitty Oracle products.
In-house products were badly done, which led to contracting out engineering, which led to no engineers, which led to getting screwed in the contracting.
Fixing that company would probably involve firing the whole tech department.
Maybe with the exception of Auth? Having some external SaaS startup be in control of your user database can be a bit nerve wracking. Especially if they get acquired by someone terrible.
I'm personally glad that Auth0 wasn't acquired by Oracle. What a nightmare scenario that would be... :D
As a counterpoint to this, at my last house in Melbourne (Australia) I was thinking of doing the same. We have a "National Broadband Network" (NBN) deployed that most residential customers in the country are hooked into.
Shortly after moving into the house, the NBN needed to upgrade the network in the area. So, multi-hour outage. Then the same thing happened about 2 months later.
Thus, any idea of self hosting from my house was dropped and we've kept on using Hetzner. ;)
It works well enough.
I’m not going to argue with your points, but we’ve spent maybe a months worth of man hours figuring out why some SaaS service wasn’t working this year. Only for it to be because some Azure or Office365 or Sharepoint online service wasn’t healthy, and, to be fair, because some of our systems aren’t build very maintainable so that it takes a ridiculous amount of time to figure out what is wrong. Still, it would’ve probably cost us maybe a work-weeks worth of time if we were more competent, which sort of goes against the SaaS idea when the licensing is also approaching the cost of self-hosting and hiring a Sharepoint person. Well, if a Sharepoint person existed of course, but I bet you get my point.
So did my work computer;)
Wow, I haven't seen or heard that name since the 2000s. Good times.
I wonder if people will experiment more?
- https://www.eevblog.com/forum/repair/trick-the-hotspotphone-...
- https://jibout.com/verizon-7730l-mifi-hotspot-battery-bypass...
- https://old.reddit.com/r/Calyx/comments/lorkrv/running_mifi_...
I feel like there's some liability on Netgear's part here: People can't be expected to know they can't leave it connected and a charging circuit should not constantly feed the battery.
It's a shame there aren't more affordable connectivity options for projects like this. Hotspots with batteries tend to be a lot cheaper than battery-less routers and USB dongles. The latter of which isn't even available for 5G.
Rocket sticks come to mind.
I "fixed" it by plugging the charger into an old school mechanical timer. Every six hours, it runs off the battery for 30 minutes. Has been working great for 2 years.
To me this seems like a common misconception. I can’t see it being true - we’d see so many more battery fires out there if this was true, including in this very case (his setup would’ve burnt down within days in that case).
There's tons of options, and the "professional" grade routers aren't much more expensive than that consumer grade AC800s. With the one I have I get a removable SIM, dual SMA antenna connectors for MIMO with the ability to have an external high gain antenna, 5 ethernet ports, and a box that runs a version of ddWRT that I have full control over.
The AC800s is $200, my modem was $350.
GL.iNet GL-AR750S-Ext
Wiflyer WE826-T
Installing OpenWRT requires soldering wires to access the serial port, so I haven't bothered yet (the native OS works fine for what I need), but as I understand everything is supported.
I wrote a script that scrapes the hotspot web interface to turn off a smart plug when it's at 90% and turn back on at 60%.
The cheap ZTE devices in particular seem especially bad at keeping the thermals within a safe range. Even with a 100% full battery they'll let the electrolyte boil if left plugged in for their charging logic to do as it wishes in a hot environment.
What's so frustrating is practically every consumer electronics device seems to now have some form of LiPo pouch cells in them. I wish it were normal for manufacturers to offer chunkier variants using NiMH AA/AAA cells to consumers. Instead I've been resorting to "industrial" stuff having no battery at all like the Gl-Inet X300B [0] I eventually ended up with.
For a while I used a hacked together pack of four NiMH AAs with a battery board stolen from the LiPo on a ZTE/AT&T hotspot that kept swelling its pack. The battery board kept the device happy enough to believe it had the proper LiPo connected. The now external AAs stayed charged more-or-less, and didn't care how hot the ambient temps got through summer. But it was a sprawling mess of wires, soldered AAs arranged like an 80s-era RC-12L saddle-pack, kept ~together with Duck tape.
When aggressively overcharged they may rupture, but they don't go up in flames like Li-ion does.
They had enough voltage to let the thing turn on and weren't going to catch fire, problem solved. Left it that way for years.
One brand I use is ATP for my ruggedized systems[0]. Yes, they may be an order of magnitude more expensive than even a high endurance from SanDisk or Samsung, but I started using these in 2015 or 2016 and found they hold in harsh terrain, tolerate brown outs (these devices are solar powered and run off of super capacitors), and JustWorks(tm).
[0]: https://www.digikey.com/en/products/detail/atp-electronics-i...
What are the statistics on lithium-ion battery failures? Are they dangerous full stop, dangerous in certain scenarios (e.g. heat), or do we only hear about the failures when 99.9999… of the time they are fine as they are everywhere?
Are all lithium-ion batteries created the same?
I am honestly amazed the whole setup lasted for so long.
Battery management aside, those portable 12v fridges are great at using a fan to heat or cool
Regarding temps, the modem itself, I seem to remember, never reported a temp higher than 55 degrees C. I don't live in a very warm place though...
Think about it in terms of "A 33kWH battery roughly equals 1 gallon of gasoline". This is the most paranoid possible way to look at it, but it does help to establish some perspective. Energy is energy and if you don't really think it through, you may wind up with much worse than LTE modem/router fires.
Tesla power walls are an example of something I could never install in my home due to this simple math. I would not be able to sleep at night knowing there is a chance they could go up while bolted to my home. I've got one of those EcoFlow batteries (2kWH) for emergencies, but I keep it on a concrete slab (protected from weather) outside my home.
Go look at how a datacenter does their batteries if you want to get a sense of this very same paranoia at scale. All the DC-scale UPS systems I've seen typically consist of gigantic, low voltage cells installed in a manner expressly designed to deal with the worst case.
I think the density of the energy is what is most dangerous in my mind. A whole room full of lead acid batteries might not be very portable, but its a hell of a lot easier to inspect room-scale components, prove they are safe, and deal with emergencies before they turn into catastrophes.
Nah, that's the wrong way to think about it. Oxidation (well, burning) of lithium releases far more energy than the electric charge of the battery. Electric (dis)charge merely gives the initial heat to break structural integrity and set lithium on fire.
And even then the problem is not in the amount of energy, but in that it's almost impossible to extinguish a lithium fire. And the fumes from battery are thick and toxic.
> I think the density of the energy is what is most dangerous in my mind. A whole room full of lead acid batteries might not be very portable...
Nah again. Lead-acid just does not burn so spectacularly. And about the energy density - filled fuel tank of a regular car has much higher energy density than Tesla battery, but is significantly harder to ignite, easier to extinguish, and when burning, releases the energy slower.
I've never heard of a house blowing up like this, but it has happened with submarines.
Always figured there was some minor defect or something that caused the battery to start emitting hydrogen.
> VAXen, my children, just don't belong some places. In my business, I am frequently called by small sites and startups having VAX problems. So when a friend of mine in an Extremely Large Financial Institution (ELFI) called me one day to ask for help, I was intrigued because this outfit is a really major VAX user--they have several large herds of VAXen--and plenty of sharp VAXherds to take care of them.
> (large skip)
> So now the whole data center was dead, sort of. The fire alarm system had it's own battery backup and was still alive. The lead acid storage batteries of the immortal power system had been discharging at a furious rate keeping all those big blue boxes running and there was a significant amount of sulfuric acid vapor. Nothing actually caught fire but the smoke detectors were convinced it had.
> The fire alarm klaxon went off and the siren warning of imminent halon gas release was screaming. We started to panic but the data center manager shouted over the din, ``Don't worry, the halon system failed its acceptance test last week. It's disabled and nothing will happen.''
> He was half right, the primary halon system indeed failed to discharge. But the secondary halon system observed that the primary had conked and instantly did its duty, which was to deal with Dire Disasters. It had twice the capacity and six times the discharge rate.
> (and it continues)
If you wanna get more paranoid: https://www.reddit.com/r/spicypillows/
Jokes aside, hard to say, but it's mostly scenarios.
Some batteries have proper venting (or puffing), protection circuits and sensors to prevent thermal runaway, and they are very unlikely to fail catastrophically, they would stop working before, generally.
Now those take more space, so such features are often offloaded onto the PCB designers to make them fit, and you may trust them more (or less) than the battery manufacturer.
More or less the same can be said about the charging circuit, and how it might handle low voltage or trickling/stop charge.
Personally I've seen flashlights that burn through the 18650 without respecting the minimum voltage whatsoever, and that would put 5V to charge. There a proper battery will refuse to charge or just die, but an unprotected one might catch fire.
As anecdotal data, after dozens and dozens of devices, I've only owned two batteries that became spicy (without catching fire), a Chinese-brand laptop and a cheap quadcopter battery.
No
They have different chemistries with different characteristics. Some are safer than others
See here https://en.m.wikipedia.org/wiki/Comparison_of_commercial_bat...
Beyond the actual chemistry there's regulators of various quality and strategies that can be found in many modern battery packs.
I'm not a professional in battery manufacturing but I'm sure there's somebody around here who is that can give more details
[1]: https://louwrentius.com/my-solar-powered-blog-is-now-on-lith...
But, this is not really a failure of solar power system, just failure of a Netgear device. It seems that it used it's battery as a high current source for when it needed more peak power for wireless transmission, thus the instability and failures when just plugged to USB charger. I don't think such constant drain and constant recharge was really healthy for the battery. I wonder if it was/is stated in manual that it was supposed to be used charged and unplugged and not really as full time usage device.
I'm looking for some kind of tablet that is wall-mountable but doesn't contain a Lithium battery. Some other kind of safe battery (NiMH?) that can last through momentary brownouts and power cuts of a few minutes would be nice.
The tablets are wall-mounted in multiple places elsewhere in the apartment. I don't want them to go into a full reboot cycle if power is lost only for a couple of seconds or even minutes.
Maybe there was a fault in the circuitry to begin with.
https://hackaday.com/2021/02/12/pcb-mods-silence-voltage-war...
I believe the RPi and the car battery they were using are still fine.
[1] https://www.netgear.com/images/datasheet/mobile/ac800s_optus...
But you have two tiny potential bombs inserted into your head.
> The battery is specifically designed to withstand the deep ear environment for months at a time
Have Phonak/Lyric achieved the impossible building a battery that never runs out of power? How does that work?
This really isn't on them at all, a properly designed Li-ion system will be absolutely fine left plugged in 24/7. Netgear either doesn't understand how to design a safe battery powered device, or this was a rare case of a bad battery just going off regardless of how you treat it.
Of course the router isn't as nice of a software environment as a pi but that's part of the fun, no? If you want easy just use GitHub pages or square space or whatever.
Where they differ is how the cells are organized, safety measures taken outside of the cells themselves and things like cooling, mechanical construction of the assembly and so on. And even big manufacturers get things wrong, notably: Bosch. They have a pannier mounted e-bike battery pack that has serious problems with water ingestion, boards frying and balancing wires catching fire.
Utility-scale battery packs are climate-managed and are in real actual use right now. O'ahu has sunsetted its coal power generation and added utility-scale batteries: https://insideevs.com/news/614726/tesla-megapacks-hawaii-arr...
I have absolutely no idea how you got that from what I wrote.
(/sarcasm, obviously, I have no idea how those things are legal)