It seemed to have poisoned our department relations with IT when I once visited the server room and I questioned why our server and Oracle database passwords were sitting next to our server.
Are janitors or anybody non-sysadminny allowed in that room? I assume the room is locked and reasonably secure. Sticking the passwords there might not be that bad an idea (though not particularly good) if the physical security is covered.
If a network administrator is unable to secure his own box relatively well (no network PC is ever 100% secure), why the %^&* would I trust him to secure a network?
If a network admin works on a box in an unsecured area, what's to stop me from installing a hardware keylogger for a week or two? They're not likely to notice an extra 4" cord attached to the back of their computer.
Yeah, it means that the combustion chamber is hemispherical - woo. If you go truck shopping, reliability will likely be a bigger issue. As an aside, do ricer rednecks stick HEMI stickers on their 1/4 ton toyota trucks?
Some examples: they can try forcing adult-oriented websites into.xxx 'because of all the poor children'
This is likely illegal, as forcing some group of adult sites (defined by community standards, usually) into a ghetto could be viewed as infringing on 1st ammendment rights.
They can ensure that internet protocols and standards are selected to benefit US business interest.
The will server only to balkanize those standards if the conditions are sufficiently onerous.
They can ensure that domain-name disputes are resolved in a big-business-friendly atmosphere where you don't have a right to your own name even if you paid for the domain, assuming some compancy has decided to sell a product with a similar name, or if someone with the same name happens to be a movie star.
Big business is all over - this is hardly a US-biased activity.
If the PS3 can run PS2 games, will it be able to run them better?
Probably not. PS2 games oeprate with the assumption of a fixed hardware config, and modifying the platform by increasing the speed is likely to have a number of unforseen effects.
Small business owners tend to have a case of megalomania. If they can pet the box, they "own" it. Thus, they'll spend $2,000 on a server rather than $25/mo on a managed solution because they can pet the box, even as they explain about the increased downtime because they don't have a dedicated admin, like their ISP.
Were you expecting sympathy? Anyway, My $2k pricetag was for a low-end server. If we're going to self host (instead of using a managed host for the web side of things like any smart businessperson would), then that can be two smaller $1k boxes with split duties and a semi-sophisticated firewall in front. Not perfect, but still quite adequate. The $2k server has to fit in a 1U form factor to reduce colo charges and sits in a rackmount box, so that costs more too. It also has more memory to avoid future upgrades (it's in a colo facility, after all).
See, there are other people in the world than yourself. And, while it's not hard to put together a Linux/sendmail server that can handle a 20 Mb stream, building one that also runs, oh, say, a web server, WebDAV, SQL, and a few other services useful to a small business may lead you to places where it's not true anymore.
Anybody that runs production hardware like that deserves what they get. There are serious security problems with running all-in-one solutions; if your needs are really so small, get a site-hosting arrangement for $25/mo. I was referring to any company large enough to run their own stuff.
And, since SCSI drives are expensive, you'll typically see a smaller (maybe 20 GB) drive on it on your small business, entry level server that's a year or two old.
If it's entry level, then it's probably IDE, and 80GB is easy for a small server 1.5 years old. Sorry, but your numbers aren't really credible.
Plus, your "20 Mb stream" server doesn't take into account anything at all resource-intensive, such as SpamAssassin, anti-virus, greylisting, or most of the other, processor-intensive functions now in common use. In reality, your baseline "20 Mb stream" server only proves that a modern SCSI drive can read/write data at a rate greater than 20 Mbps.
Any modern disk can do 20Mb/sec. SCSI is no longer necessary for much aside from SAN apps. Regardless of the tasks performed, my point was that no properly configured server should choke on mail. This is a solved problem.
For all you know, the company concerned might have no more than a handful of employees, so a mail server capable of handling 5 million emails in a short space of time would be totally inappropriate.
Let's see - 5M messages at 10k each = 50GB. If it were a small company, they may have only had a 1.5Mb line, so that 50GB would take about 50GB/150K/3600 = 92 hours to complete. Any mail server can handle that, and any competent admin should be able to block the messages within four days!
Of course, a 3rd party hosted mail server could handle the mail a bit faster, so the only question is whether 50GB is an excessive amount. Since I have a 300MB quota, it might be. Then again, maybe not - disk space is cheap, and nuking one message sent to any number of people is pretty straightforward.
And then it spawns more and more processes to process the mail, eating up ram, at which point any other services on the box may be overloaded and deprived of resources.
No, the mail server is a dedicated box, and thee are limits to how many processes it will spawn. What it will do is queue a bunch of messages and work through the backlog. I can build a $3k box (plus the cost of a storage array if needed) that will handle a 20Mbit stream of mail all day long. This isn't rocket science.
The problem that I have with comments like this is that the extremely wealthy pay the majority of US income tax. Therefore, any change will affect them disproportionaly. Look at IRS data. Taypayers in the top 1%, as ranked by income, pay 34% of all federal income tax.
Well of course they do. They own 80% of all property!
Target has a stated corp policy of not checking sigs. They don't want to be responsible for getting duped, so they've entirely washed their hands of it.
Which just means they're wholly liable if the sigs don't match - this is in the merchant agreement that mastercard and visa have.
by that point they had spent $1800 from my checking account. fortunately, my credit union (i swear by credit unions) refunded all but $50 in exchange for me filing a police report.
With a credit card, that would have been a $1800 reduction in available credit, with no risk of bounced checks if your account had $1500 in it. In the case of internet purchases, why would anyone use a debit card?
A VP at a bank should know the risks associated with check cards and be able to qualify for a credit card that doesn't have those risks. It kind of goes with the territory.
Slashdot Mirror
These same cleaners and rent-a-cops would also have access to offices as well. So this is a strawman argument.
Unless the doors are locked with the trashcan left outside, like any proper secured location should do.
It seemed to have poisoned our department relations with IT when I once visited the server room and I questioned why our server and Oracle database passwords were sitting next to our server.
Are janitors or anybody non-sysadminny allowed in that room? I assume the room is locked and reasonably secure. Sticking the passwords there might not be that bad an idea (though not particularly good) if the physical security is covered.
If a network administrator is unable to secure his own box relatively well (no network PC is ever 100% secure), why the %^&* would I trust him to secure a network?
If a network admin works on a box in an unsecured area, what's to stop me from installing a hardware keylogger for a week or two? They're not likely to notice an extra 4" cord attached to the back of their computer.
The fundamental irony in this statement is that it was not a small step for a man, it was a 250,000 mile step for the individual man.
I thought the fundamental irony way that he screwed up the quote, even though he'd practiced it - One small step for man, a giant leap for mankind.
+1 Bizarrecastic
Do I know what the hell HEMI even means?
Yeah, it means that the combustion chamber is hemispherical - woo. If you go truck shopping, reliability will likely be a bigger issue. As an aside, do ricer rednecks stick HEMI stickers on their 1/4 ton toyota trucks?
Some examples: they can try forcing adult-oriented websites into .xxx 'because of all the poor children'
This is likely illegal, as forcing some group of adult sites (defined by community standards, usually) into a ghetto could be viewed as infringing on 1st ammendment rights.
They can ensure that internet protocols and standards are selected to benefit US business interest.
The will server only to balkanize those standards if the conditions are sufficiently onerous.
They can ensure that domain-name disputes are resolved in a big-business-friendly atmosphere where you don't have a right to your own name even if you paid for the domain, assuming some compancy has decided to sell a product with a similar name, or if someone with the same name happens to be a movie star.
Big business is all over - this is hardly a US-biased activity.
If the PS3 can run PS2 games, will it be able to run them better?
Probably not. PS2 games oeprate with the assumption of a fixed hardware config, and modifying the platform by increasing the speed is likely to have a number of unforseen effects.
Energy is needed to generate light signals (touch a light bulb for reference)
Yeah, incandescents aren't very efficient. LEDs (which is what chips would use) are quite a bit better.
Why is it "natural?" Did they route the cabling through the cloths racks?
If you go to a department store, you'll find that all cables are generally routed along the ceiling.
Small business owners tend to have a case of megalomania. If they can pet the box, they "own" it. Thus, they'll spend $2,000 on a server rather than $25/mo on a managed solution because they can pet the box, even as they explain about the increased downtime because they don't have a dedicated admin, like their ISP.
Were you expecting sympathy? Anyway, My $2k pricetag was for a low-end server. If we're going to self host (instead of using a managed host for the web side of things like any smart businessperson would), then that can be two smaller $1k boxes with split duties and a semi-sophisticated firewall in front. Not perfect, but still quite adequate. The $2k server has to fit in a 1U form factor to reduce colo charges and sits in a rackmount box, so that costs more too. It also has more memory to avoid future upgrades (it's in a colo facility, after all).
When black apears white or pigs appear to have sprouted wings, there's usually politics behind it, that's where Critical Thinking separates the herd.
We need a +6 or +7 for posts like this. Either that or I've been here too long.
"MiB" is the abbreviation for "mebibyte", which is 2^20 bytes
hehehe, maybebytes... I'll stick with megabytes, TYVM.
So, to disambiguate the prefixes while not disturbing the traditional meanings, the IEC coined a new set of binary prefixes
Too bad they didn't get much community buyin.
See, there are other people in the world than yourself. And, while it's not hard to put together a Linux/sendmail server that can handle a 20 Mb stream, building one that also runs, oh, say, a web server, WebDAV, SQL, and a few other services useful to a small business may lead you to places where it's not true anymore.
Anybody that runs production hardware like that deserves what they get. There are serious security problems with running all-in-one solutions; if your needs are really so small, get a site-hosting arrangement for $25/mo. I was referring to any company large enough to run their own stuff.
And, since SCSI drives are expensive, you'll typically see a smaller (maybe 20 GB) drive on it on your small business, entry level server that's a year or two old.
If it's entry level, then it's probably IDE, and 80GB is easy for a small server 1.5 years old. Sorry, but your numbers aren't really credible.
Plus, your "20 Mb stream" server doesn't take into account anything at all resource-intensive, such as SpamAssassin, anti-virus, greylisting, or most of the other, processor-intensive functions now in common use. In reality, your baseline "20 Mb stream" server only proves that a modern SCSI drive can read/write data at a rate greater than 20 Mbps.
Any modern disk can do 20Mb/sec. SCSI is no longer necessary for much aside from SAN apps. Regardless of the tasks performed, my point was that no properly configured server should choke on mail. This is a solved problem.
For all you know, the company concerned might have no more than a handful of employees, so a mail server capable of handling 5 million emails in a short space of time would be totally inappropriate.
Let's see - 5M messages at 10k each = 50GB. If it were a small company, they may have only had a 1.5Mb line, so that 50GB would take about 50GB/150K/3600 = 92 hours to complete. Any mail server can handle that, and any competent admin should be able to block the messages within four days!
Of course, a 3rd party hosted mail server could handle the mail a bit faster, so the only question is whether 50GB is an excessive amount. Since I have a 300MB quota, it might be. Then again, maybe not - disk space is cheap, and nuking one message sent to any number of people is pretty straightforward.
And then it spawns more and more processes to process the mail, eating up ram, at which point any other services on the box may be overloaded and deprived of resources.
No, the mail server is a dedicated box, and thee are limits to how many processes it will spawn. What it will do is queue a bunch of messages and work through the backlog. I can build a $3k box (plus the cost of a storage array if needed) that will handle a 20Mbit stream of mail all day long. This isn't rocket science.
Check the URL, the Top 1% gets only 17% of the income, yet pays 34% of the tax.
Property, asshole. They own 80% of the property. Dividends are not considered regular income.
The problem that I have with comments like this is that the extremely wealthy pay the majority of US income tax. Therefore, any change will affect them disproportionaly. Look at IRS data. Taypayers in the top 1%, as ranked by income, pay 34% of all federal income tax.
Well of course they do. They own 80% of all property!
It's a simple matter to go to the embassy, renounce your US citizenship and surrender your US passport if they're not providing you any services.
I was under the impression that the State Dept didn't acknowledge any such activity.
Shouldn't you just tax a person in his place of employment if he is a remote worker?
The reasonable thing would be to tax them in their place of presence.
Target has a stated corp policy of not checking sigs. They don't want to be responsible for getting duped, so they've entirely washed their hands of it.
Which just means they're wholly liable if the sigs don't match - this is in the merchant agreement that mastercard and visa have.
by that point they had spent $1800 from my checking account. fortunately, my credit union (i swear by credit unions) refunded all but $50 in exchange for me filing a police report.
With a credit card, that would have been a $1800 reduction in available credit, with no risk of bounced checks if your account had $1500 in it. In the case of internet purchases, why would anyone use a debit card?
Oh? And why would you say that?
A VP at a bank should know the risks associated with check cards and be able to qualify for a credit card that doesn't have those risks. It kind of goes with the territory.
Have you ever _used_ a credit card? The clerks don't look, don't care.
I'm in sattle - about half of the merchants check my id when I use a card.
apparently ended up giving my Visa CheckCard number to the wrong person.
Wait, you're a VP at a bank, and you used a check card online? Not to question your intelligence, but that's not a very bright thing to do.