I was thinking the same thing... remove the pigtails and solder over the connectors (short out the radio.) HOWEVER, the wifi gear would still be functional. Enterprising inmates *would* find a way to undo the simple short and re-enable the radio. The only sure way is to burn out the radio(s), AND the ethernet PHY. Doable, but tricky.
There are 3 electric circulation systems that are 100% controllable from the control room. They could have (should have) turned those off to reduce the cooling rate. No one has said IC only was over cooling. Turning off an emergency system in the middle of situation that tripped the system in the first place is, IMO, a Bad Idea. *IF* you lose electrical power -- a now proven false "that can't happen" situation, you'll lose the primary cooling systems and the ability to re-enable the emergency system you've intentionally turned off. When you think something cannot happen, you pretty much guarantee it will happen.
There are plenty of things that could have been handled differently -- and the article lists several. The simplest action here would've been to leave the IC loop active. Had they left the valves open, it's very likely none of this would have happened. That's the first of many places the disaster could have been averted.
And the nuts followed a flawed proceedure and shut off the gravity fed cooling system. Turning off the emergency system that does not require power to opperate, but will require power to re-enable, is simply the worst idea ever. Who ever wrote their proceedures never concidered an event where they lose power, or lose the powered circulation pumps. (which could be a due to power loss, structural damage, control system failure, or the pumps being destroyed.)
True, but the only cars where you'll find such a system is a Tesla or a track/race car. The Prius and it's kind cannot generate that much power from it's electric motor. (and doesn't even try. you can get about 50% braking power from it, but the pads are engaged at that point.
(Unless one uses engine braking, and most people don't know it's there -- the CVT version of "down shifting". But some power is lost from the engine still being engaged.)
What I was referring to was the direct hydrolic link between the driver's brake peddle and the actual brakes. Without power, they will still work... for a limited number of pumps. The "park brake" isn't an "emergency brake" -- it isn't intended to do anything but hold the car in place. That said, an E-Brake will not rapidly stop a car; it will slow it to a stop, however, over a much longer distance. (typically only cabled to the rear brakes.)
I don't have a link, but I recall a trucking fleet adding electrolysis systems to their semi's to feed hydrogen into the engine. They reported a 10% increase in fuel efficiency. At first I thought there's no way that can be true, but when you think about it, that's the same a battery and regen breaks in a hybrid... they're storing excess energy (from the engine, breaking, and gravity) as hydrogen.
That's only true on paper. You have to be *taught* to drive efficiently in an electric/hybrid vehicle. Long slow breaking and gentle acceleration are a must. But you will still lose energy in the process. A moderate, steady speed is the most efficient mode, pretty much for any car, but much more so for electric/hybrid cars as they have somewhere to put excess energy.
Driving my Lexus HS250h, I've found 40+mpg is easy on the highway crusing at 65mph. (70+ starts to pull the mpg's down. 70 is 37-38mpg, 80 is 35-36.) Stop and go city traffic kills it. City only driving falls down to around 35mpg. (38 is doable, but it pisses everybody off -- including me. I can only drive like that when there is *no* traffic around.)
Everything on the car is electically powered. Yes, there is a manual "emergency" provision for breaking, but without power to charge the system, you get about the same out of the volt as ANY powered-off vehicle. Try it with you own car; with the engine off, you'll get 2-3 good cycles out of the breaks before you're screwed.
The problem with the "ask some nice folks" route is "for a bit" is actually SEVERAL HOURS. The Top Gear idiots went tooling around in a few; it ended as you would expect from those nuts.
The Volt does exactly what every other car does... zero charge does not mean "dead". Just like "0 miles remaining" in my car does not mean I'm out of gas. (I actually have ~3gal in the tank at that point which is another 100-150miles!)
Everyone knows Kevin Bacon isn't spam -- he's Bacon after all.:-)
Just because the OP blocked port 25 on his node, doesn't mean it's blocked everywhere. Listing every known exit node may be a little extreme, but it's a fairly safe move. A tor exit node should not be sending email. How does a random server on the internet know it's a real mail server and not tor originating the connection? I say again, if you are crazy enough to run tor on your mail server, you should be prepared for the fallout. (there are more than a few systems that will not accept any connection from a known exit node. mail, web, any connection.) Or put another way, don't come crying to me when your mail server ends up listed in hundreds of blocklists because it's also a tor exit node; the lists block all tor exit nodes, and I'm not making an exception for anyone.
Indeed. One email to a spamtrap, ever, gets an IP listed forever. There are no thresholds, volume limits, or expiration times. And they offer zero proof. The only way to be delisted is their nice money-laudering-esq, quasi-extortion "charity donation" scam. No reputable charitable entity will even speak their name.
There are laws (in the US at least)... slander and defamation. All of the customers now incorrectly listed as "spammers" have a case if they want to push it. (in the US at least, where spamhaus won't even show up.:-))
Once you've listed the spamming block, then your job is done. UNLESS the ISP moves them to another block (evades the listing), you have no reason to escalate. Just because the ISP doesn't do what you want them to isn't a reason to start listing more and more of their customers -- who are not spammers. This isn't helping the anti-spam cause; it is, in fact, undermining the cause.
Things have certainly gone down hill there. They never choked BTI and we were Jeremy Jayne's ISP; every one of his netblocks were listed (often before they were even routed), but the company's ARIN assigned netblocks never were.
Yes. [See also: every large ISP] AOL has their own dumbass rules, as do far too many other ISPs. The real problem, as you've seen is the entire undocumented nature of one's arbitrary rules.
It's their policy to not deal with "spammers". If you've been listed, you are a spammer, and they will not talk to you. Your ISP will have to talk to them. They usually will not tell anyone what got you listed to start with -- if you are a spammer, that information is gold for evading detection in the future. Having dealt with the several times (as the ISP), they're actually good guys. (there are infintely worse anti-spam nuts out there.)
Actually, Spamhaus lists a great many/32's. If they've listed your entire netblock / subnet, you've done *something* to earn it. (what that might be is known only to them. and they aren't sayin')
And to be fair, tor is just too easy to abuse. That said, you shouldn't be running a tor exit node on your mail server.
And their smarthost service is run about as poorly as their rDNS service. The key problems with this... it can introduce a great deal of delay (i.e. overloaded smarthost); your spam score will be at the mercy of everyone else using that relay; and when that relay gets blacklisted because of one of those other customers, you're instantly blacklisted as well.
In a perfect world, it's a simple process of asking the ISP to change the record. HOWEVER, this is not a perfect world. In over 20 years, the only time I've gotten PTRs updated simply and quickly was when I could vi the zone file myself.
For some services, it's a matter of going to your customer control panel. Others, it takes a single email. And then for others, you have to show up in person with a hammer and do it yourself since their staff are complete morons. (cheap comes at a price.) Spammers will generally avoid the latter because they have better things to do with their time.
Translation: it is almost completely useless. Such a policy is far more likely to drop non-spam as actual spam. Spammers rarely have any difficulty finding and/or setting up hosts with working PTRs. People running real mail servers may have a hard time getting PTRs set correctly. (Yes, there are asshats who not only require a PTR, but require it to be a certain format!)
Actually, the hull is charged as part of the protection system. It's a very small charge (less than a volt if I remember correctly) but enough to get current flowing out the zinc annodes. (vs. the hull itself.)
The issue with aluminium/steal junctions is that it creates a bimetal junction, which is a thermocouple and will create a current.
The level of magnetism in most of the magnets was barely enough to keep it attached to the metal shrouding the PC, yet it was sufficient to corrupt her hard...
I have to call bullshit here. If the magnets are so weak they barely stick to steal, then they are not going to hurt a harddrive. Even if you opened the drive and touched the platter, their magnetic field is just not strong enough. To understand this better, find the episode of Mythbusters where they test ways to erase a magnetic strip (i.e. credit card.)
As an aside, but also supporting evidence: studies have shown that server hard drives can be affected by hard drives (in the same machine or in other servers) physically adjacent; causing issues not only via magnetic influence, but also via vibration.
What study (must less studies)? Vibration I would accept. But neighboring magnetics... No. Just. No.
Slashdot Mirror
I was thinking the same thing... remove the pigtails and solder over the connectors (short out the radio.) HOWEVER, the wifi gear would still be functional. Enterprising inmates *would* find a way to undo the simple short and re-enable the radio. The only sure way is to burn out the radio(s), AND the ethernet PHY. Doable, but tricky.
There are 3 electric circulation systems that are 100% controllable from the control room. They could have (should have) turned those off to reduce the cooling rate. No one has said IC only was over cooling. Turning off an emergency system in the middle of situation that tripped the system in the first place is, IMO, a Bad Idea. *IF* you lose electrical power -- a now proven false "that can't happen" situation, you'll lose the primary cooling systems and the ability to re-enable the emergency system you've intentionally turned off. When you think something cannot happen, you pretty much guarantee it will happen.
There are plenty of things that could have been handled differently -- and the article lists several. The simplest action here would've been to leave the IC loop active. Had they left the valves open, it's very likely none of this would have happened. That's the first of many places the disaster could have been averted.
And the nuts followed a flawed proceedure and shut off the gravity fed cooling system. Turning off the emergency system that does not require power to opperate, but will require power to re-enable, is simply the worst idea ever. Who ever wrote their proceedures never concidered an event where they lose power, or lose the powered circulation pumps. (which could be a due to power loss, structural damage, control system failure, or the pumps being destroyed.)
True, but the only cars where you'll find such a system is a Tesla or a track/race car. The Prius and it's kind cannot generate that much power from it's electric motor. (and doesn't even try. you can get about 50% braking power from it, but the pads are engaged at that point.
(Unless one uses engine braking, and most people don't know it's there -- the CVT version of "down shifting". But some power is lost from the engine still being engaged.)
What I was referring to was the direct hydrolic link between the driver's brake peddle and the actual brakes. Without power, they will still work... for a limited number of pumps. The "park brake" isn't an "emergency brake" -- it isn't intended to do anything but hold the car in place. That said, an E-Brake will not rapidly stop a car; it will slow it to a stop, however, over a much longer distance. (typically only cabled to the rear brakes.)
Long breaking is essential to getting energy back. Aggressive/hard breaking uses the break pads, which just creates heat; you get none of that back.
I don't have a link, but I recall a trucking fleet adding electrolysis systems to their semi's to feed hydrogen into the engine. They reported a 10% increase in fuel efficiency. At first I thought there's no way that can be true, but when you think about it, that's the same a battery and regen breaks in a hybrid... they're storing excess energy (from the engine, breaking, and gravity) as hydrogen.
That's only true on paper. You have to be *taught* to drive efficiently in an electric/hybrid vehicle. Long slow breaking and gentle acceleration are a must. But you will still lose energy in the process. A moderate, steady speed is the most efficient mode, pretty much for any car, but much more so for electric/hybrid cars as they have somewhere to put excess energy.
Driving my Lexus HS250h, I've found 40+mpg is easy on the highway crusing at 65mph. (70+ starts to pull the mpg's down. 70 is 37-38mpg, 80 is 35-36.) Stop and go city traffic kills it. City only driving falls down to around 35mpg. (38 is doable, but it pisses everybody off -- including me. I can only drive like that when there is *no* traffic around.)
Everything on the car is electically powered. Yes, there is a manual "emergency" provision for breaking, but without power to charge the system, you get about the same out of the volt as ANY powered-off vehicle. Try it with you own car; with the engine off, you'll get 2-3 good cycles out of the breaks before you're screwed.
The problem with the "ask some nice folks" route is "for a bit" is actually SEVERAL HOURS. The Top Gear idiots went tooling around in a few; it ended as you would expect from those nuts.
The Volt does exactly what every other car does... zero charge does not mean "dead". Just like "0 miles remaining" in my car does not mean I'm out of gas. (I actually have ~3gal in the tank at that point which is another 100-150miles!)
Everyone knows Kevin Bacon isn't spam -- he's Bacon after all. :-)
Just because the OP blocked port 25 on his node, doesn't mean it's blocked everywhere. Listing every known exit node may be a little extreme, but it's a fairly safe move. A tor exit node should not be sending email. How does a random server on the internet know it's a real mail server and not tor originating the connection? I say again, if you are crazy enough to run tor on your mail server, you should be prepared for the fallout. (there are more than a few systems that will not accept any connection from a known exit node. mail, web, any connection.) Or put another way, don't come crying to me when your mail server ends up listed in hundreds of blocklists because it's also a tor exit node; the lists block all tor exit nodes, and I'm not making an exception for anyone.
Indeed. One email to a spamtrap, ever, gets an IP listed forever. There are no thresholds, volume limits, or expiration times. And they offer zero proof. The only way to be delisted is their nice money-laudering-esq, quasi-extortion "charity donation" scam. No reputable charitable entity will even speak their name.
There are laws (in the US at least)... slander and defamation. All of the customers now incorrectly listed as "spammers" have a case if they want to push it. (in the US at least, where spamhaus won't even show up. :-))
Once you've listed the spamming block, then your job is done. UNLESS the ISP moves them to another block (evades the listing), you have no reason to escalate. Just because the ISP doesn't do what you want them to isn't a reason to start listing more and more of their customers -- who are not spammers. This isn't helping the anti-spam cause; it is, in fact, undermining the cause.
Things have certainly gone down hill there. They never choked BTI and we were Jeremy Jayne's ISP; every one of his netblocks were listed (often before they were even routed), but the company's ARIN assigned netblocks never were.
Half right... it means someone with a clue found someone with a clue at their ISP. The ISP half of that clue sandwich is a rare thing these days.
Right. That stops all the DKIM signed spam coming from Yahoo! Oh, wait, it doesn't; in fact, it would allow all of it through.
Yes. [See also: every large ISP] AOL has their own dumbass rules, as do far too many other ISPs. The real problem, as you've seen is the entire undocumented nature of one's arbitrary rules.
It's their policy to not deal with "spammers". If you've been listed, you are a spammer, and they will not talk to you. Your ISP will have to talk to them. They usually will not tell anyone what got you listed to start with -- if you are a spammer, that information is gold for evading detection in the future. Having dealt with the several times (as the ISP), they're actually good guys. (there are infintely worse anti-spam nuts out there.)
Actually, Spamhaus lists a great many /32's. If they've listed your entire netblock / subnet, you've done *something* to earn it. (what that might be is known only to them. and they aren't sayin')
And to be fair, tor is just too easy to abuse. That said, you shouldn't be running a tor exit node on your mail server.
And their smarthost service is run about as poorly as their rDNS service. The key problems with this... it can introduce a great deal of delay (i.e. overloaded smarthost); your spam score will be at the mercy of everyone else using that relay; and when that relay gets blacklisted because of one of those other customers, you're instantly blacklisted as well.
You do realize the ISP owns the f'ing address. That means *THEY* control what PTRs are assigned. They are the only ones that can change them.
In a perfect world, it's a simple process of asking the ISP to change the record. HOWEVER, this is not a perfect world. In over 20 years, the only time I've gotten PTRs updated simply and quickly was when I could vi the zone file myself.
For some services, it's a matter of going to your customer control panel. Others, it takes a single email. And then for others, you have to show up in person with a hammer and do it yourself since their staff are complete morons. (cheap comes at a price.) Spammers will generally avoid the latter because they have better things to do with their time.
Translation: it is almost completely useless. Such a policy is far more likely to drop non-spam as actual spam. Spammers rarely have any difficulty finding and/or setting up hosts with working PTRs. People running real mail servers may have a hard time getting PTRs set correctly. (Yes, there are asshats who not only require a PTR, but require it to be a certain format!)
Actually, the hull is charged as part of the protection system. It's a very small charge (less than a volt if I remember correctly) but enough to get current flowing out the zinc annodes. (vs. the hull itself.)
The issue with aluminium/steal junctions is that it creates a bimetal junction, which is a thermocouple and will create a current.
I have to call bullshit here. If the magnets are so weak they barely stick to steal, then they are not going to hurt a harddrive. Even if you opened the drive and touched the platter, their magnetic field is just not strong enough. To understand this better, find the episode of Mythbusters where they test ways to erase a magnetic strip (i.e. credit card.)
What study (must less studies)? Vibration I would accept. But neighboring magnetics... No. Just. No.