Don't be a tool. Even the WTC directly affected only 0.00006667% of the world's population. Multiply by ten or even a hundred if you want to include families of the deceased.
It sucks royally if you are part of that 0.00006667%, but those are still decent odds even for the most conservative of gamblers.
Try to install one on my doorway, and I'll show you why we have a 2nd Amendment.
Maybe I need to adjust my tin-foil hat, but I'm starting to wonder if maybe Ruby Ridge was a warning for everyone who might think about getting uppity now. Yes, that sounds uber-paranoid even to me, but...
And now, DHS is moving from airlines to Amtrak (Google "TSA Savannah Georgia" for the citation), and according to TFA, wants to expand into stadiums and streets, too. Do I give up my 4th Amendment rights to freedom from unreasonable searches when I walk out my front door, too?
The Bill of Rights sets limits on what the government can and cannot do. You can try to weasel around the Constitution by saying things like, "well...well...well...that doesn't apply here" but I defy you to point me to the text in the Constitution that says so. The only wiggle room there is for the pat-downs and AIT searches is whether or not they constitute a "reasonable" search. Unfortunately, at the moment, GWB and Obama have managed to get the whole nation so afraid of the terrorist boogeyman that people like you are willing to give up anything to be "safe". I only hope we restore a little sanity before the U.S.A. becomes yet another fascist, totalitarian state. Tyranny doesn't happen overnight; it happens little by little, trading one small liberty for the promise of safety or security. We're on that path.
Then there is a very simple solution: everyone who wants to get into an event must strip completely and put their pile of clothes and things into the x-ray machine before they walk through a metal detector.
Unless you have "something that shouldn't be there" stuffed inside the bodily orifice of your choice. Wake up, people: there is NO WAY to stop a sufficiently motivated criminal. Risk is a fact of life; I'd rather live free and accept the marginally greater risk that comes with that freedom than live in a totalitarian nanny state.
I am fortunate enough to live in an area where there is more than enough pristine drinking water nearby, and perhaps you are too. However, in many parts of the country, well water is the only source of water there is. If that well water is contaminated by oil development nearby, from where do you propose the residents get their drinking water? It doesn't just magically appear at the tap, you know...
No one has a right to a particular life style. If these businesses really are messing up ground water (rather than merely being blamed for such) the people harmed by the action can sue for recourse. If that isn't the case, then the residents can always move or treat their water.
at others wonder if they are in the pockets of the businesses they are supposed to be watching
They have to be lenient to businesses else we would have a lot less businesses and a lot more unemployment. The regulations are garbage and unenforceable. Until that fundamental problem gets fixed, I'm not inclined to damage the US further.
Which is why I explicitly stated that sometimes the EPA seems to be way too extreme on one side, but at other times, way too extreme the other. I'm not a radical greenie who wants every square inch of the U.S. turned into a wildlife sanctuary. I happen to rather like driving my truck, riding my motorcycle and working with computers, all of which are products of mining and all of which depend (to varying extents) upon the petrochemical industry. However, I also think that businesses have to recognize that they don't exist in a vacuum. Their operations have an impact upon the surrounding communities, and if they want to continue to operate in those communities, they MUST exercise responsible development. You would think that common sense would dictate that 1) if you want to have consumers to buy your product, you'd better make sure you aren't killing them off while you produce the product in the first place, and 2) if you want to enjoy the wealth that your business has created, you'd better make sure you aren't killing yourself, either. Unfortunately, people often tend to be greedy and short-sighted, and as a result, there is a need for the EPA. Yes, many of the regs may be "garbage" and "unenforceable", but I'd rather have an imperfect EPA than unfettered oil barons trashing the planet one neighborhood at a time. And while I often disagree with many of the environmental groups -- they tend to be just as one-sided and biased as the corporations they target, IMHO -- I think they fill a necessary role, too. If all you ever hear is one side of the argument, how do you find the balanced, middle ground?
You don't have to drink ground water that catches on fire, you know.
I am fortunate enough to live in an area where there is more than enough pristine drinking water nearby, and perhaps you are too. However, in many parts of the country, well water is the only source of water there is. If that well water is contaminated by oil development nearby, from where do you propose the residents get their drinking water? It doesn't just magically appear at the tap, you know...
Here's my take. Ground water contamination (and soil stability) is a minor, local problem. Our civilization becoming crippled because some envirofreaks want to halt progress, is a serious, global problem.
Yeah...unless it's your ground water and your soil stability that you are worried about, then it's a major local problem. My house, and the land it's on cost me something like five to six years of my annual salary. If some oil company starting pumping chemicals into the ground that caused my well water to be undrinkable, and the soil that my house is built upon to become unstable, it would financially ruin me. You might be okay with that happening to me, but I seriously doubt you'd be so flippant with it if it was your house and property that suddenly became worthless. That kind of short-sighted attitude is callous, unjust and, quite frankly, idiotic. While the EPA is far from perfect -- I sometimes roll my eyes at excessive regulation and government interference, and at others wonder if they are in the pockets of the businesses they are supposed to be watching -- there is no doubt in my mind that we are far better off for having at least some semblance of environmental restraint upon pure free market capitalism.
The problem is that when everyone is trying to "prove" their side of a debate, and the truth be damned, it gets a little hard to separate the truth from the fluff.
Is Gasland accurate, or is it sensationalist hype to sell one side of the argument? Is the EPA really unbiased, or has lobbying put pressure on Congress to approve the fracking, and has Congress in turn put pressure on the EPA to minimize the potential danger? Who is funding the studies, and what pressure were the researchers under to prove one side of the argument?
I took a statistics course in the '90s in college, and my stats instructor had served as an "expert witness" in a number of court cases. As he explained it, he would be hired by one of the parties in a lawsuit, and they would invariably ask, "What do the statistics say about <some phenomenon>" and he would frankly reply, "What do you want them to say?" The point he was making is that by selecting your sample set appropriately, you can "prove" whatever point you want to prove.
We can't put every business on the chopping block at the first cry of foul play. We should also be careful to differentiate between hype, hysteria, and the facts. And there is a difference between negligence and honest mistakes - these businesses stand to suffer a great deal for any of their mistakes. It just so happens that a lot rides on these types of businesses, unlike one that simply makes toys. Of course they will be ridiculed if something goes wrong, and we would be remissed if we didn't demand they do better. But, they do constantly improve. The gas companies know very well what the media and gov't will do to them if someone or something gets hurt, and I think you know that factors into their operations decisions.
For the most part, I agree with you. The bottom line is that, unless western society as a whole is willing to take a huge step back into the Dark Ages or we suddenly discover some miraculous new energy source, we will need oil for the foreseeable future. I also agree that a lot of media hype/hysteria about the big, bad oil companies is exactly that: sensationalist hype feeding hysteria to sell news stories. However, experience has shown that companies in general -- and oil companies are no exception -- are often willing to take unreasonable risks to cut costs, sometimes with disastrous consequences. If you don't believe that, then I have two words for you: Deepwater Horizon.
Sometimes industries are crucified in the media because they well and truly deserve it.
There is a big difference between the laws of economics and the physical laws like gravity. Congress cannot affect physical laws. As you said, no matter what laws Congress passes, the law of gravity will never, ever change. However, Congress certainly can pass laws that impose additional costs on the free market, and that will most certainly affect economic decisions, albeit not always in the way they intended (for example, Prohibition).
Yes, economics is a major player in keeping petroleum the top energy source for the U.S. right now. Oil companies and motor companies have a huge investment in that infrastructure, so changing to solar/wind/Mr. Fusion overnight would be prohibitively expensive. However, if you think that business is not lobbying to keep their investments profitable, you are delusional. Even the agriculture business has (successfully) lobbied Congress to create a economic and political climate to produce energy from energy sources that would be inefficient in a truly free market, i.e., ethanol from corn -- which takes more petroleum to produce than it replaces(!).
In other words, you'll just call anyone who disagrees with you names rather than posting an insightful/informative mod explaining what they got wrong? You may very well be right that they don't know what they are talking about and that they are just being sensationalist, but since you didn't deign to rebut anything they said, I'll never know for sure.
IMHO, you did not only lower yourself to their level but dropped even lower by being condescending without contributing anything useful to the discussion.
I lived in Japan for a while as a kid, so while I won't pretend to even be marginally fluent in the language, I got the impression while living there that Japanese had something at least somewhat similar to American swear words. For example, I once was sitting in a restaurant with a Japanese friend of my family's. In the restaurant was a parrot (myna bird? something like that, anyway) to which restaurant patrons had taught some choice phrases and/or words. Every time the bird was speak, the Japanese people in the restaurant would laugh. Not wanting to miss out on what was obviously a very funny joke, I asked the family friend what the bird was saying, but he declined to tell me, saying something to the effect of "It is not polite," or "it is not appropriate." I was something like six or eight at the time, so take the anecdote for however much or however little it's worth <shrug>
Most people skilled in a language (yes, even perl) can figure out what a piece of code actually does given enough time...
Unfortunately, in the case of perl, human beings do not have a sufficiently long lifespan:D
I kid, I kid (mostly...)! However, even though I'm on the perl side of a perl vs. python debate where I work, I have to admit, I'd much rather try to figure out what someone else's python code did than try to figure out what someone else's perl code did. There are just too many ways to do the same thing in perl, and some of those ways can get pretty darned arcane.
...and there is what you do when every time there is a problem your phone starts ringing with people screaming at you because they can't work and why haven't you fixed the server yet and how long will it take and what about their important meeting at 2pm etc.
I solved that problem by putting my phone on forward during an outage:) There are only two of us sys admins where I work, and there are two desktop support people. When the poo hits the fan, the desktop support guys run interference while we admins work on the problem. YMMV.
The strange thing is, you've got some practices that are touted as "best practice" while at the same time being castigated as a security risk. So what are normal people (as normal as geeks can be) to do?
The answer lies in risk management: if you want a perfectly secure computer, remove the hard drive, encase the computer in solid concrete and sink it in the middle of the Marianas Trench. That's probably about as secure as you can make a computer. Unfortunately, it's not terribly useful. So a "normal geek" should seek to find the best balance between "secure" and "usable".
Maybe the group in the discussion you referenced (and unfortunately, I missed that one...) needs more security than I do in my environment, and therefore eliminating sudo is a good choice for them. I set up my personal servers to use SSH trusts with a pass phrase to log in and then to require sudo to run anything that needs root permissions. This accomplishes several things:
Remote connections can only come from hosts that I have explicitly allowed (that's why I set it up in the first place -- I was tired of seeing all of the brute-force SSH username/password guessing games in my logs).
To sign in, you have to know the SSH trust passphrase, so even if you have access to a "trusted" host, there is still an authentication mechanism required
Once you have logged in to the server, you must know the password for the login account (*NOT* the same as the SSH trust passphrase) to use sudo. So even though you've authenticated, you have to authenticate again with a second password to do anything very useful.
Finally, since you are using sudo, I can see what users have done in my log files, and I will get an e-mail if someone tries to run sudo with an incorrect password. Consequently, unless you already know both the SSH trust passphrase AND the login account password, I'll get an e-mail as soon as you try to run anything with sudo, reducing the time an attacker has to do anything nefarious.
I'm sure it's not a perfect solution, but it seems to work reasonably well for me.
Point taken. Next question: do you want a regular user to run tcpdump? Granted, most networks are switched, so the potential security hole is rather small in a typical business scenario, but if you happen to be on a network segment connected with a hub (or on a shared server, for example), it is something to consider. You'd be surprised what I've seen customers broadcasting in the clear over the Internet feeds my employer provides to them...
The answer is right there in his second sentence: "I've found that the pre-emptive reboot will trigger hidden system problems..." The purpose is to verify that the hardware is still working as it should.
I mentioned elsewhere in the comments that I personally have seen a server keep running with a failed hard drive (non-RAID, obviously) for several months. In this particular case, everything the server needed to keep running was available in RAM and NFS mount, but the server obviously couldn't boot up again after a power outage took it off-line. This caused an unplanned outage in the middle of a production day, when a scheduled reboot in the middle of an outage window would have been much less catastrophic.
Rebooting is bad. It booted the first time, Why would it not boot the second?
Please tell me you aren't seriously asking that question?
Here are a few examples:
1) Because the RAID controller battery died, and you didn't know it because you never, ever reboot your box. Consequently, when your server dies while you are on vacation in Hawaii, you end up talking an entry-level desktop support monkey through your RAID setup via telephone;
2) Because when your vendor SSH'd into the box, the moron deleted/var (or/boot) while troubleshooting (no joke, this actually happened to me once);
3) Because hardware ages, and problems that may lie dormant can be exposed when the server is rebooted. I've seen Dell USFF desktops work fine until they were rebooted, but then fail to come back up because some caps on the motherboard were leaking. The desktop would have eventually failed, but the reboot revealed the problem earlier. Better to discover that your server is on its last legs during a maintenance window than in the middle of the production day.
4) Again, because hardware ages, and problems that may lie dormant can be exposed when the server is rebooted. We had another server lose a hard drive (no RAID). All of its services were running from a RAM drive and writing to NFS mounts, so we had no idea the HDD had died until a power outage took it off-line, and the server didn't come back up afterwards.
5) You do update your server from time to time, don't you? Do you know that all of your configs, etc., are still valid for the updated services, kernels, etc.? I once had to administer a VMWare server that required work every time we updated the Linux kernel for VMWare to start. If you updated the kernel, didn't reboot the server and run the proper tweaks, then subsequently had a power failure, the server would start, but none of the guest OS's would. Rebooting the server after updates proved that the guest OS's would be available after a power failure.
YMMV, so do what works in your environment, but IMHO, periodic reboots are a good way of verifying that your server will come back on-line if something were to take it off the air at night, while you are on vacation, etc.
In fact, none of the *nix sysadmins I know would dream of rebooting the box to clear a problem except as a last resort.
It's rare that I reboot a *Nix box to clear a problem, but I have had a problem with some Linux-based routers that I use where SNMP hangs and can't be killed until the box is rebooted. Since we use SNMP to keep tabs on the routers (they are providing access to our anchor-tenant customer, located about 500 miles away from my desk), if SNMP quits working, it's typically worth a reboot to fix...although I'll wait until off-hours to do so.
I haven't personally been bit by rebooting a Linux box and making everything worse... But I've seen enough other people get bit, and I've read enough horror stories on-line.
I have, and again, it was on those same Linux-based routers*. There have been a couple of times when sending a reboot command to a router that had a sufficiently hosed snmpd that the box failed to shut down. Unfortunately, it tends to fail at a point where SSH has already ended, so I lose access to the router. Fortunately, it tends to fail at a point before ospfd and forwarding has shut down, so while I can neither monitor nor manage the router, it hasn't quit passing customer traffic. So as far as getting bitten goes, it's not terribly bad, but it is annoying.
*In fairness, these routers have proven to be extremely reliable, powerful and flexible. Our customer tends to have far more problems with the Cisco routers they connect to our Linux-based routers than we have with our routers. On 70 or so deployed routers, I have an snmpd problem maybe once every two or three months. Perhaps twice a year, I'll actually have to reboot a router to clear snmpd, and I think I've had three or four of those reboots require me to dispatch someone to physically power off the router (in five years).
Slashdot Mirror
Don't be a tool. Even the WTC directly affected only 0.00006667% of the world's population. Multiply by ten or even a hundred if you want to include families of the deceased.
It sucks royally if you are part of that 0.00006667%, but those are still decent odds even for the most conservative of gamblers.
Well said. Makes me wish I had mod points.
Try to install one on my doorway, and I'll show you why we have a 2nd Amendment.
Maybe I need to adjust my tin-foil hat, but I'm starting to wonder if maybe Ruby Ridge was a warning for everyone who might think about getting uppity now. Yes, that sounds uber-paranoid even to me, but...
"Maybe" doesn't work for me. Find out for sure, and then "maybe" I'll agree to be scanned. But I wouldn't count on it.
Yeah, you are right.
And now, DHS is moving from airlines to Amtrak (Google "TSA Savannah Georgia" for the citation), and according to TFA, wants to expand into stadiums and streets, too. Do I give up my 4th Amendment rights to freedom from unreasonable searches when I walk out my front door, too?
The Bill of Rights sets limits on what the government can and cannot do. You can try to weasel around the Constitution by saying things like, "well...well...well...that doesn't apply here" but I defy you to point me to the text in the Constitution that says so. The only wiggle room there is for the pat-downs and AIT searches is whether or not they constitute a "reasonable" search. Unfortunately, at the moment, GWB and Obama have managed to get the whole nation so afraid of the terrorist boogeyman that people like you are willing to give up anything to be "safe". I only hope we restore a little sanity before the U.S.A. becomes yet another fascist, totalitarian state. Tyranny doesn't happen overnight; it happens little by little, trading one small liberty for the promise of safety or security. We're on that path.
Then there is a very simple solution: everyone who wants to get into an event must strip completely and put their pile of clothes and things into the x-ray machine before they walk through a metal detector.
Unless you have "something that shouldn't be there" stuffed inside the bodily orifice of your choice. Wake up, people: there is NO WAY to stop a sufficiently motivated criminal. Risk is a fact of life; I'd rather live free and accept the marginally greater risk that comes with that freedom than live in a totalitarian nanny state.
Give it a try in Arizona, and let me know how well that works for you...
Not to mention that, IIRC, when Social Security was first rolled out, it was to be an optional program. How did that work out for us Yanks?
I am fortunate enough to live in an area where there is more than enough pristine drinking water nearby, and perhaps you are too. However, in many parts of the country, well water is the only source of water there is. If that well water is contaminated by oil development nearby, from where do you propose the residents get their drinking water? It doesn't just magically appear at the tap, you know...
No one has a right to a particular life style. If these businesses really are messing up ground water (rather than merely being blamed for such) the people harmed by the action can sue for recourse. If that isn't the case, then the residents can always move or treat their water.
"No business has the right to a particular income stream. If the business really isn't messing up the ground water (rather than merely claiming such), then they can sue for recourse. If that isn't the case, then the business can always move." Do you still feel the same way about the issue now, or are you just a hypocrite? Furthermore: "We hold these truths to be self-evident, that all men are...endowed by their Creator with certain unalienable Rights, that among these are Life, Liberty, and the pursuit of Happiness..." If your business operation is poisoning my well water and thereby depriving me of my "unalienable Rights" then I daresay you are quite mistaken.
at others wonder if they are in the pockets of the businesses they are supposed to be watching
They have to be lenient to businesses else we would have a lot less businesses and a lot more unemployment. The regulations are garbage and unenforceable. Until that fundamental problem gets fixed, I'm not inclined to damage the US further.
Which is why I explicitly stated that sometimes the EPA seems to be way too extreme on one side, but at other times, way too extreme the other. I'm not a radical greenie who wants every square inch of the U.S. turned into a wildlife sanctuary. I happen to rather like driving my truck, riding my motorcycle and working with computers, all of which are products of mining and all of which depend (to varying extents) upon the petrochemical industry. However, I also think that businesses have to recognize that they don't exist in a vacuum. Their operations have an impact upon the surrounding communities, and if they want to continue to operate in those communities, they MUST exercise responsible development. You would think that common sense would dictate that 1) if you want to have consumers to buy your product, you'd better make sure you aren't killing them off while you produce the product in the first place, and 2) if you want to enjoy the wealth that your business has created, you'd better make sure you aren't killing yourself, either. Unfortunately, people often tend to be greedy and short-sighted, and as a result, there is a need for the EPA. Yes, many of the regs may be "garbage" and "unenforceable", but I'd rather have an imperfect EPA than unfettered oil barons trashing the planet one neighborhood at a time. And while I often disagree with many of the environmental groups -- they tend to be just as one-sided and biased as the corporations they target, IMHO -- I think they fill a necessary role, too. If all you ever hear is one side of the argument, how do you find the balanced, middle ground?
how about instead of a fictitious "mother earth", accepting faults in MO and AK have been shaking OK for a very long time
Seriously? Faults in Alaska have been shaking Oklahoma?!?! Oh...you meant AR. Never mind.
You don't have to drink ground water that catches on fire, you know.
I am fortunate enough to live in an area where there is more than enough pristine drinking water nearby, and perhaps you are too. However, in many parts of the country, well water is the only source of water there is. If that well water is contaminated by oil development nearby, from where do you propose the residents get their drinking water? It doesn't just magically appear at the tap, you know...
Here's my take. Ground water contamination (and soil stability) is a minor, local problem. Our civilization becoming crippled because some envirofreaks want to halt progress, is a serious, global problem.
Yeah...unless it's your ground water and your soil stability that you are worried about, then it's a major local problem. My house, and the land it's on cost me something like five to six years of my annual salary. If some oil company starting pumping chemicals into the ground that caused my well water to be undrinkable, and the soil that my house is built upon to become unstable, it would financially ruin me. You might be okay with that happening to me, but I seriously doubt you'd be so flippant with it if it was your house and property that suddenly became worthless. That kind of short-sighted attitude is callous, unjust and, quite frankly, idiotic. While the EPA is far from perfect -- I sometimes roll my eyes at excessive regulation and government interference, and at others wonder if they are in the pockets of the businesses they are supposed to be watching -- there is no doubt in my mind that we are far better off for having at least some semblance of environmental restraint upon pure free market capitalism.
The problem is that when everyone is trying to "prove" their side of a debate, and the truth be damned, it gets a little hard to separate the truth from the fluff.
Is Gasland accurate, or is it sensationalist hype to sell one side of the argument? Is the EPA really unbiased, or has lobbying put pressure on Congress to approve the fracking, and has Congress in turn put pressure on the EPA to minimize the potential danger? Who is funding the studies, and what pressure were the researchers under to prove one side of the argument?
I took a statistics course in the '90s in college, and my stats instructor had served as an "expert witness" in a number of court cases. As he explained it, he would be hired by one of the parties in a lawsuit, and they would invariably ask, "What do the statistics say about <some phenomenon>" and he would frankly reply, "What do you want them to say?" The point he was making is that by selecting your sample set appropriately, you can "prove" whatever point you want to prove.
We can't put every business on the chopping block at the first cry of foul play. We should also be careful to differentiate between hype, hysteria, and the facts. And there is a difference between negligence and honest mistakes - these businesses stand to suffer a great deal for any of their mistakes. It just so happens that a lot rides on these types of businesses, unlike one that simply makes toys. Of course they will be ridiculed if something goes wrong, and we would be remissed if we didn't demand they do better. But, they do constantly improve. The gas companies know very well what the media and gov't will do to them if someone or something gets hurt, and I think you know that factors into their operations decisions.
For the most part, I agree with you. The bottom line is that, unless western society as a whole is willing to take a huge step back into the Dark Ages or we suddenly discover some miraculous new energy source, we will need oil for the foreseeable future. I also agree that a lot of media hype/hysteria about the big, bad oil companies is exactly that: sensationalist hype feeding hysteria to sell news stories. However, experience has shown that companies in general -- and oil companies are no exception -- are often willing to take unreasonable risks to cut costs, sometimes with disastrous consequences. If you don't believe that, then I have two words for you: Deepwater Horizon.
Sometimes industries are crucified in the media because they well and truly deserve it.
Ummm....sort of.
There is a big difference between the laws of economics and the physical laws like gravity. Congress cannot affect physical laws. As you said, no matter what laws Congress passes, the law of gravity will never, ever change. However, Congress certainly can pass laws that impose additional costs on the free market, and that will most certainly affect economic decisions, albeit not always in the way they intended (for example, Prohibition).
Yes, economics is a major player in keeping petroleum the top energy source for the U.S. right now. Oil companies and motor companies have a huge investment in that infrastructure, so changing to solar/wind/Mr. Fusion overnight would be prohibitively expensive. However, if you think that business is not lobbying to keep their investments profitable, you are delusional. Even the agriculture business has (successfully) lobbied Congress to create a economic and political climate to produce energy from energy sources that would be inefficient in a truly free market, i.e., ethanol from corn -- which takes more petroleum to produce than it replaces(!).
In other words, you'll just call anyone who disagrees with you names rather than posting an insightful/informative mod explaining what they got wrong? You may very well be right that they don't know what they are talking about and that they are just being sensationalist, but since you didn't deign to rebut anything they said, I'll never know for sure.
IMHO, you did not only lower yourself to their level but dropped even lower by being condescending without contributing anything useful to the discussion.
Ummm...maybe.
I lived in Japan for a while as a kid, so while I won't pretend to even be marginally fluent in the language, I got the impression while living there that Japanese had something at least somewhat similar to American swear words. For example, I once was sitting in a restaurant with a Japanese friend of my family's. In the restaurant was a parrot (myna bird? something like that, anyway) to which restaurant patrons had taught some choice phrases and/or words. Every time the bird was speak, the Japanese people in the restaurant would laugh. Not wanting to miss out on what was obviously a very funny joke, I asked the family friend what the bird was saying, but he declined to tell me, saying something to the effect of "It is not polite," or "it is not appropriate." I was something like six or eight at the time, so take the anecdote for however much or however little it's worth <shrug>
Most people skilled in a language (yes, even perl) can figure out what a piece of code actually does given enough time...
Unfortunately, in the case of perl, human beings do not have a sufficiently long lifespan :D
I kid, I kid (mostly...)! However, even though I'm on the perl side of a perl vs. python debate where I work, I have to admit, I'd much rather try to figure out what someone else's python code did than try to figure out what someone else's perl code did. There are just too many ways to do the same thing in perl, and some of those ways can get pretty darned arcane.
...and there is what you do when every time there is a problem your phone starts ringing with people screaming at you because they can't work and why haven't you fixed the server yet and how long will it take and what about their important meeting at 2pm etc.
I solved that problem by putting my phone on forward during an outage :) There are only two of us sys admins where I work, and there are two desktop support people. When the poo hits the fan, the desktop support guys run interference while we admins work on the problem. YMMV.
The strange thing is, you've got some practices that are touted as "best practice" while at the same time being castigated as a security risk. So what are normal people (as normal as geeks can be) to do?
The answer lies in risk management: if you want a perfectly secure computer, remove the hard drive, encase the computer in solid concrete and sink it in the middle of the Marianas Trench. That's probably about as secure as you can make a computer. Unfortunately, it's not terribly useful. So a "normal geek" should seek to find the best balance between "secure" and "usable".
Maybe the group in the discussion you referenced (and unfortunately, I missed that one...) needs more security than I do in my environment, and therefore eliminating sudo is a good choice for them. I set up my personal servers to use SSH trusts with a pass phrase to log in and then to require sudo to run anything that needs root permissions. This accomplishes several things:
I'm sure it's not a perfect solution, but it seems to work reasonably well for me.
Point taken. Next question: do you want a regular user to run tcpdump? Granted, most networks are switched, so the potential security hole is rather small in a typical business scenario, but if you happen to be on a network segment connected with a hub (or on a shared server, for example), it is something to consider. You'd be surprised what I've seen customers broadcasting in the clear over the Internet feeds my employer provides to them...
My boss is indeed a tech head, and not a typical PHB. So, yeah, I'd probably keep him informed because if I don't he'll figure it out on his own.
The answer is right there in his second sentence: "I've found that the pre-emptive reboot will trigger hidden system problems..." The purpose is to verify that the hardware is still working as it should.
I mentioned elsewhere in the comments that I personally have seen a server keep running with a failed hard drive (non-RAID, obviously) for several months. In this particular case, everything the server needed to keep running was available in RAM and NFS mount, but the server obviously couldn't boot up again after a power outage took it off-line. This caused an unplanned outage in the middle of a production day, when a scheduled reboot in the middle of an outage window would have been much less catastrophic.
Rebooting is bad. It booted the first time, Why would it not boot the second?
Please tell me you aren't seriously asking that question?
/var (or /boot) while troubleshooting (no joke, this actually happened to me once);
Here are a few examples:
1) Because the RAID controller battery died, and you didn't know it because you never, ever reboot your box. Consequently, when your server dies while you are on vacation in Hawaii, you end up talking an entry-level desktop support monkey through your RAID setup via telephone;
2) Because when your vendor SSH'd into the box, the moron deleted
3) Because hardware ages, and problems that may lie dormant can be exposed when the server is rebooted. I've seen Dell USFF desktops work fine until they were rebooted, but then fail to come back up because some caps on the motherboard were leaking. The desktop would have eventually failed, but the reboot revealed the problem earlier. Better to discover that your server is on its last legs during a maintenance window than in the middle of the production day.
4) Again, because hardware ages, and problems that may lie dormant can be exposed when the server is rebooted. We had another server lose a hard drive (no RAID). All of its services were running from a RAM drive and writing to NFS mounts, so we had no idea the HDD had died until a power outage took it off-line, and the server didn't come back up afterwards.
5) You do update your server from time to time, don't you? Do you know that all of your configs, etc., are still valid for the updated services, kernels, etc.? I once had to administer a VMWare server that required work every time we updated the Linux kernel for VMWare to start. If you updated the kernel, didn't reboot the server and run the proper tweaks, then subsequently had a power failure, the server would start, but none of the guest OS's would. Rebooting the server after updates proved that the guest OS's would be available after a power failure.
YMMV, so do what works in your environment, but IMHO, periodic reboots are a good way of verifying that your server will come back on-line if something were to take it off the air at night, while you are on vacation, etc.
In fact, none of the *nix sysadmins I know would dream of rebooting the box to clear a problem except as a last resort.
It's rare that I reboot a *Nix box to clear a problem, but I have had a problem with some Linux-based routers that I use where SNMP hangs and can't be killed until the box is rebooted. Since we use SNMP to keep tabs on the routers (they are providing access to our anchor-tenant customer, located about 500 miles away from my desk), if SNMP quits working, it's typically worth a reboot to fix...although I'll wait until off-hours to do so.
I haven't personally been bit by rebooting a Linux box and making everything worse... But I've seen enough other people get bit, and I've read enough horror stories on-line.
I have, and again, it was on those same Linux-based routers*. There have been a couple of times when sending a reboot command to a router that had a sufficiently hosed snmpd that the box failed to shut down. Unfortunately, it tends to fail at a point where SSH has already ended, so I lose access to the router. Fortunately, it tends to fail at a point before ospfd and forwarding has shut down, so while I can neither monitor nor manage the router, it hasn't quit passing customer traffic. So as far as getting bitten goes, it's not terribly bad, but it is annoying.
*In fairness, these routers have proven to be extremely reliable, powerful and flexible. Our customer tends to have far more problems with the Cisco routers they connect to our Linux-based routers than we have with our routers. On 70 or so deployed routers, I have an snmpd problem maybe once every two or three months. Perhaps twice a year, I'll actually have to reboot a router to clear snmpd, and I think I've had three or four of those reboots require me to dispatch someone to physically power off the router (in five years).