Posts by Paul Crawford 6133 publicly visible posts • joined 15 Mar 2007
Indeed there are, such as applying AppArmor, having user-writable areas set to no-execute, etc. But also there are always new ways to take over being discovered (even if at a lesser rate than MS gets pwaned). If your ZFS host is on a different OS you can hope said new-takeover flaw is not common to both.
Assuming you don't use the same administrator credentials of course...
One of the attractions of ZFS is the low-overhead snapshots, excellent if your system is hit be ransomware.
But that presumes that the vulnerabilities that allowed it are not able to take over the ZFS host. Having FreeBSD for your NAS and Linux/Windows for your clients allowed that, making the NAS Linux-based removes one aspect of such security.
So chips in a data centre with high L-band attenuation through reinforced concrete have GPS? Nope, not going to work...
Chips have to "phone home" so you can check IP address? Hmm, just have open Internet access routed via a VPN to ${Country_of_choice} then.
In related news pigs have been seen flying!
Hence rebooting is safest, because then you know that everything is using the new stuff.
You also know it will reboot after that has succeeded, useful when you get a forced reboot on power interruptions, watchdog reset, accidental command in wrong terminal, etc...
Oh dear, back to YouTube as a source of research?
You might want to check that Dr's soundness:
https://en.wikipedia.org/wiki/John_Campbell_(YouTuber)#Ivermectin
If that is not enough, then "...in December 2024 shared his faith and belief in Creationism in an interview with Russell Brand".
Except they are taking PhD students off the street (by masked agents, not by going to them at home or study as the police might in order to ask questions) simply for disagreeing with the gov policy. In a country that claims to offer freedom of speech, how is that acceptable?
https://www.nbcnews.com/news/us-news/tufts-university-pushes-release-student-grabbed-street-immigration-off-rcna199460
My feeling is that an ATS is a single point of failure if you only have one, otherwise you also have other single points of failure as well (the generator, the DB that follows the ATS most likely with some incoming breaker, etc). If you are trying to avoid a SPoF then you need dual power routes to each item of equipment, etc, so the ATS is no longer a single path.
Our existing site has no easy way to bypass the ATS because we are cheap, but we do test it regularly (taking site on to generator and off again, so we know it actually switches, the UPS hold up, and no breakers trip). For another client we are including a bypass switch so it can be re-routed in an emergency, or just if maintenance on it needs it powered down.
You need to be sure the HVO is good quality stuff so its free of remnants of oil, etc, that might clog filters but that should not be an issue with major suppliers. There is also Shell GTL which is a similar paraffinic fuel:
https://www.shell.com/business-customers/commercial-fuels/shell-gtl-fuel.html
HVO is a slightly lower density than diesel so your generator's output power and autonomy are both slightly less (3% I think), but that may still be acceptable in most cases.
https://www.equipmentjournal.com/on-the-job/hvo-versus-diesel-dispelling-the-myths/
We found our generators start just as quick, but they are small units (16kVA - 60kVA) so probably not comparable to yours. I can't believe that you would only test months apart! We do it weekly as it fits in with other schedules and my own feeling is I would not expect my car to sit unused for a couple of months and not have issues of seizing up, etc.
BTW low sulphur diesel used in nuclear plant emergency generators is estimated to need replacing every 6 months (it also degrades at >30c storage). HVO (if it stores better) could be a real win for the nuclear industry (No replacing every 6Mo)
We are planning on HVO for our backup generators and it is mostly due to the long storage period claimed (up to 10 years) compared to typical diesel expectation of 6-12 months. The reason is we are looking at several days of run-time for emergencies (arse end of nowhere sort of places so low on the priority list for the grid to fix after a major a storm) but typically they will be tested for 10-20min once a week so fuel will last for a long time. Realistically we will need to refill/cycle the stored stuff maybe every 5 years if HVO lives up to claimed life, just so we keep best part of half a tank of good fuel at all times.
Other factor is the relatively low toxicity and biodegradability of HVO as well, so risks and resulting clean-up cost of a serious clean up on any spillage or leak is far less.
But I have to agree that using it for prime power is bonkers, but so is a lot of what we do as a species :(
I don't care what others use, I'm happy being in that small percent of Linux desktop users.
What I do insist on is if a fried or family member wants me to support them then it is also a Linux desktop they get. So far no real issues, a few minor complaints once about software for kids not being supported, but then I no longer have the "wipe and reinstall after infection in spite of AV" experience that used to be a biannual ritual...
Longer term those hardware specialities will be a problem for Windows as well when new drivers for the recent incarnation of Windows cease to be developed. Being able to put those machines on an isolated VLAN, etc, and using the IoT versions of Windows so less demands on networking and other crap would allow some use and safety once the OS goes out of support.
At the end of the day you pay your money/privacy/sanity and you take your choice. For me it is only a few special CAD packages for old days I need, and they run in VMs OK (including older versions of Office) so I use Linux as my first choice these days.
Linux still sucks, but the taste of donkey balls is far less...
Depends on the length, if the loss of several kW is over 100m or so you would hardly notice the heating at any one point. For short cables the usable capacity is normally set by the thermal heating effects, but for long cables it is the voltage drop that usually* does it.
[*] the 3rd limit is being able to disconnect under fault conditions fast enough for safety, that is more complicated as it depends on the characteristics of the protection device (fuse, circuit breaker, etc) and the supply (fault impedance / earthing type).
Or even simpler as 3&E or T&E
In the USA it is usually 2-phase split supply so 120V - 0V - 120V plus earth/ground. Big loads use the 240V from the outer set, smaller loads are 120V off either half.
In the UK it is rare for a domestic property to have 3-phase, and when it does it is usually either a very big place with a few single-phase boards on different floors/wings, or there are a couple of big loads like fast EV charger, or a big water heater or heat-pump. So the 3&E cable is almost always seen in small sizes (1mm^2 or 1.5mm^2) and used for two-way light switches, even though it is rated for 400V/230V 3-phase use
With few exceptions (simple heaters or motors, for example) 3P equipment also needs a neutral as it will have either an imbalanced load or a few single phase loads in addition to a big 3P one (for example controller logic on 230V and a 3P motor system using the 400V delta) so when you do see it wired it is normally 4 core plus an earth (could be a 5th conductor, or the steel armour of some cables).
There has been work on relativistic magnetrons with GW peak output powers:
https://pubs.aip.org/aip/mre/article/4/6/067201/253062/Review-of-the-relativistic-magnetron
"In 1985, Benford et al. developed a 6.9 GW, 4.5 GHz relativistic magnetron. In the late 1980s, and early 1990s, they studied phase locking and high-repetition-rate operation. They also developed the ORION HPM test source for the UK in the early 1990s. These advances at Physics International are reviewed below."
The problem is not solved by grabbing Greenland. As one reg writer in the past so succinctly put it, the difference between ore and dirt is the economics of extraction.
The USA (and other countries) real problem is not a lack of high-concentration "dirt" but the lack of refining capacity, and for the rare earth metals that is hard because they are all so similar in chemical terms. That well-developed refining capacity is what China has to give it dominance (along with a disregard for environmental factors relating to it, but I bet the USA would be much the same).
The argument for Galileo was not about normal peace-time situations, but in some sort of conflict region when other big players decide to deny use of their own systems / degrade accuracy for that region as they have skin in the conflict's game.
In the past we assumed the USA would be on the side of 'the west' & democracy, while it was considered obvious that Russia and China would not cooperate even if just out of principle. Now all bets are off for GPS.
Yes, in fact the whole program was supposed to be a commercial service but nobody ever expected that to fly. It was just the EU wonks trying to put a market spin on something they knew they needed for strategic / military reasons but didn't have the wider support to push it for that alone.
Now of course it seems like a very wise investment...
...and that is keeping power on. I wonder who the UPS vendor is?
From bitter experiences we had some Dell-branded APC 5kVA UPS and they were useless, often failing when tested and failed-hard, internal bypass also broken. Out of 5 in total they were all dead within 2 years of operation. APC now owned by Schneider, so I wonder if that crap has appeared elsewhere?
I think the plan had been to include FPGA cores within the CPU to allow ultra-fast offloading of some types of calculation where it might not be a 'standard' addition (like AES, etc) for bespoke DSP use, etc.
But one way or another, having got something Intel failed to make much out of it, selling at a loss a few years later. Just like most other things they have had/tried and dropped.
You have a very valid point, but doing so means appointing competent people to senior positions, those with a knowledge of both the work to be done and how to manage a commercial entity.
Not just lickspittles who happen to tow the party line and have bugger-all in the way of useful competence...
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
