Damn straight. Your duty as a sysadmin is to ensure that the company can maxamize profits. Whenever you think about doing something, try and decide if it will make the company more money than it will cost.
Epoxy the power dongle into the laptop. Then epoxy the brick to a desk. Their only real recourse is to cut the power cord. Even then, they'll have a laptop with a cut power cord epoxied into the case.
The only thing better than a teenager, is to get two computers and hire two teenagers.
Honestly, why go for an expensive, complicated solution when a simple solution is already at hand.
5 minutes per CD gives about 12 CSs per hour.
That's 25 hours to rip 300 CDs.
$5 per hour comes in at $125. Buy a pizza for lunch over 3 days brings it to just under $200.
If you borrow a laptop or two, there is no reason one guy can't swap out CDs in 3 computers; it's be done in a day. Offer a local teen $150 + pizza for a day's work, and they'll jump at the chance.
So, unless you can come up with something less than $200, you are just shooting yourself in the foot.
>>They focus on making the most popular games work and screw the rest.
No, they work on making you think that the most popular games will work. When you actually try to install, patch, and play these games, you find out otherwise.
>>better than giving it to Microsoft I guess.
If you have a problem with MS, then why buy a game based on that system? A game sale for a MS game is the same as an OS sale for MS. The devs license DirectX, programming tools, and logos. They feed the coffers of the monster.
Me, I just wanna play a fucking game without rebooting. Fuck all that RMS GNU/Bullshit.
Re:How useful? / Machine Requirements
on
Cedega 5.1 Released
·
· Score: 3, Interesting
I tend to agree. I was a subscriber for over a year. Games that were listed as "should work" never worked for me. I used RedHat, Debian, Gentoo, no good. I tried various Nvidia cards across Intel and AMD processors.
We even had geek LAN parties where we tried to get things to work. We eventually got BF1942 to work a little. And Rainbow6 worked quiet well.
But, looking back, I think that the vast majority of people claiming success with WineX were company shills. Either that, or people didn't want to admit that their $5 a month was a complete waste.
Yeah, what he said. And while we are at it, I don't want to play flight simulators unless I can actually be in a cockpit at 15000 feet. And no driving simulators till my seat jumps every time I hit a bump. And no more AI opponents; every character in every game should be played by another player or paid actor.
We should just ignote the fact that there are technical and monitary limitations to everything. Sure, I could sit down and, with off the shelf technology, design an arcade shooter with 360-degree visibility. Of course, it'd cost $50,000 for one box. It'd burn out $1000 worth of bulbs every month. It'd be suseptable to teenagers touching the screens or spilling their sodas.
To top it all off, it'd cost $5 for a few minutes of play.
We use 17" and 19" screens for a reason. The same reason we use keyboards vice voice recognition.
But, I guess in your world everyone drinks unicorn giggles, uses their voice to provide input, and gets output via a nural link.
A former boss of mine had an interesting approach. We are hardware guys. Mostly connecting big telecom multiplexers to bigger multiplexers. Anyway, he'd invite the potential candidate into a room with a bunch of equipment and cabling. Manuals would be scattered around. He'd apoligise about the mess and tell the candidate that he'd be back in a few minutes.
After about 15 minutes, he'd come back and say that they were having a big problem. The candidate could wait or leave.
If the candidate left, he'd be removed from the pool. If he stayed, but didn't tinker with the equipment on the tables, he'd also be removed.
The candidates that stayed and read manuals or tinkered with the equipment were usually hired.
Long, long ago in a college far, far away, I had a Unix class. The classroom consisted of keyboards on desks surrounding the walls. In the middle of the room, there was a teletype.
You would hand-write out the things you needed to do. Then you'd carefully type those into the keyboard. Once you were done, you could go to the teletype and look for the output line that corresponded to your login. And there was your output.
I had Steam installed a few weeks before HL2 was released. I pre-orded HL2 and began to download it a week before release. The day of the release, I signed into Steam and was playing in 5 minutes.
It would have taken longer than that to open the shrinkwrap and install the Vivendi HL2 boxed copy.
When I was schedule to fly to the US for some conference, I clicked the "play HL2 offline" button. I played just fine on the airplane and in the Hotel without Internet access.
When I wanted to reinstall, I clicked the "backup this game" button, inserved a DVD, and let it do it's thing. After the reinstall, I clicked the "restore from backup" button and was playing agian in 10 minutes.
Don't just bash Steam when you have no idea what you are talking about. No one has any idea what will happen when Valve goes under. Maybe someone will buy the franchise and keep the servers going. Maybe Valve will release the authenticator code. Anything else is just a conspiracy theory.
BTW, if you think having a physical plastic disc is some guarentee that you'll be able to play a game in 20 years, think agian. I have a "Shogo: MAD" game and a "Jane's Fleet Command" game. Neither will play under WinXP. Sure, I could go back and install Win98, but who wants to hunt down drivers for devices made in 2005 that will operate on a 5+ year old OS.
Raptors are bad ass. I have two of the 36gb version in a RAID0 config. I can sustain over 110MB/sec. At a LAN party, I moved all my "tradable files" to this array and plugged in to the Gig-E switch with my NForce3 Gig-E adaptor.
So, you think that every person should just focus on one issue at a time? The President is perfectly capable of doing two (sometimes more) things at once.
Iraq may or may not have been a mistake. But it happened. If we leave now, it'll be even worse in 20 years. We need to clean up the Middle East (not just Iraq) now. In 50 years, Bagdad has the potential to be the same city as Tokyo or Berlin is today. We can't just leave that mess for our children (or our children's children) to die over. Every year we delay in taking out Muslim extremists (and the governments that support them) is another year for them to become more deadly.
You can't fight a war without running up a deficit.
And then Katrina happened. It's bad shit that could have been prevented, but it happened. We have no choice but to deal with it.
Education sucks. It needs to be addressed.
Outsourcing sucks. Needs to be addressed.
Health care and retirement entitlements need to be addressed.
There are about 5000 burners going full blast on the White House's stove. And not a single one is a back burner.
I think we can all agree that it would have been nice if $event had not happened and $money wasn't required to fix it. It would also be nice to think that next year will be slow and maybe The Man can clear up his inbox.
As a member of the military, I see the anti-war and pro-warrior crowd as a very strange animal. How can you support me without supporting my actions? And it's difficult to claim that I don't support my actions because I've had amply oppertunity to quit.
Some number of years ago, I found myself in charge of a private infrastructure. We had maybe 50 servers and 400 users exchanging sensitive information completely seperate from the main, public network.
Because of the percived importance of uptime on this network, everything required mountians of paperwork. Installing and removing nodes from the domain required three administrators, setting up a new machine required a month on a private VLAN being monitored by a sniffer, memory and hard drives were obselete before they got to the customer.
Anyone who ever worked around an UPS knows how they die. They give plenty of warning. Having an UPS fail is a rediculous way to lose your backbone infrastructure.
My predicessor had done a wonderful job of installing an UPS for every router and switch in the datacenter. Problem is, both power supplies in the routers and switches were connected to the same UPS. In cases where an UPS was about to fail, he unplugged the UPS from the wall and plugged it into, you guessed it, another UPS.
He didn't do it out of ineptitude; it was done because the only option was to clash heads with the IT overlords. They would require studies about how many UPSs failed and if it failed before the MTBF, they'd want us to try and recover money from the manufacturer. They'd want contractors to come in and examine the UPS to bid on a UPS monitor and replacement contract.
In short, asking the overlords was like asking to be turked by a syphalitic bear.
So, some BOFH, overwhelmed by the prospect of repairing the power system, chose another path. He walked over to a failing UPS and simply turned it off. He was the only one with the access to turn it back on, so he had no reason to worry.
Within two hours, all in-progress meetings were cancled. The Supreme Overlords demanded from on high that this lowly tech was to get a blank check and a blank trouble ticket (approved by the Supreme Overlords) to do whatever he needed to do to prevent that from ever happening agian.
Electricians installed two seperate power feeds into every rack.
Each power supply got a seperate UPS.
Old equipment was updated.
Everything was strawberry fields and unicorn giggles after that for the infrastructure department.
Now, to answer your question: You have something that someone wants. Hold it hostage till you get what you need.
The latest SF protection requirs a reboot. I installed X3 and Silent Hunter 3. The first time I tried to play, it errored because I was running Daemon Tools. I exited DT and tried agian. SF informed me that protection had been applied and then told me to reboot.
If you are going to do a stage 3 or a GRP, you might as well use Debian. Deb has packages for GCC and the kernel compiled for virtually every acrhetecture out there.
I know what you mean about "other things". I love never having to deal with RMP hell agian.
Re:Maybe a grain of salt, but it's what I'd predic
on
Wine vs Windows Benchmarks
·
· Score: 2, Informative
The transition from Debian-based systems to Gentoo will be fairly painless. Just understand that it may be several days before you get a working system. My main computer is a dual-boot for WinXP and Gentoo. During the install, the family had to live without being able to access WinXP for almost 5 days. Of course, this was on a mid-range AthlonXP with 256MB RAM. I have (in boxes at the moment) a new dual-core Athlon. I expect it to take about 48 hours from fdisk to KDE.
There are precompiled packages avalible to speed things up, but where's the fun in that.
Yeah, Debian to Gentoo is quite easy. Just use "emerge" vice "apt-get".
AdBlock and FlashBlock are designed to block ads and flash. Modifying your hosts file to block spyware is a false sense of security. Dangerously false.
If you use Windows, AutoUpdate at least weekly. Nightly may be overkill, but isn't really hurting anything.
Turn on the Firewall and do not allow exceptions unless you know what they are for.
Install and use Opera or Firefox.
Install and update AVG and/or Avast. Norton is overkill for most home users. Why pay $50 for something only marginally better than the free editions?
Install and run SpyBot, MS Ad Scanner, and AdAware.
If you get any Spy/Adware, your only real recourse is to treat it like a virus; nothing can be trusted. You should back up your data and reformat/reinstall. Then scan your backups for malicious programs before you restore them.
An Ad-Blocking Hosts file is a dumb suggestion. If you can modify the Hosts file, what makes you think that a program you launch can't modify the same file?
And before you suggest running as a non-admin user, don't forget that a lot of programs will not run properly unless you have admin rights.
Now, I guess you could put the hosts file on a floppy and write-protect that. Then you can create a symlink to the file on the floppy.
An additional thing about RAID. Always reboot to rebuild your array. If at all possible, come in at like 3am on a Sunday to replace a failed drive. Shutdown and remove the failed drive. Insert a new drive and turn the power back on. Never trust the "hot swap" scam.
Seriously, downtime isn't that hard to get. Especially if you tell them that not granting a reboot window could potentially cost them 5+ hours of unscheduled downtime.
If you work in a place where they refuse to grant downtime, have your boss draft a memo stating that you advised him of the potential problems and he will accept all responsibility in case of an outage.
If he won't do that, then you draft the memo and have a coworker sign statiing that the boss ordered you to hot swap agianst your wishes.
The chances of losing two disks at once are slim. RAID 0+1 will provide great performance and good fault tolerance if you react to problems as they happen.
But I guess it depends on what your users need. If they need raw throughput, RAID 0+1 is better. If they need low latency, then RAID 10 may be the answer. Or maybe both systems would fall within the margin of error of each other.
In any event, once you get into what-if situations, no RAID will be good enough. What if you lose a disk? What about two? Five? Well, what if lightning hits the chasis or the janitor unplugs it to buff the floor?
The best you can do is roll the dice and play the odds. You'll see that I told him to use RAID 0+1. I also told him to use good monitoring setups to mitigate problems. I also suggested a tape backup. Actually, maybe I didn't, but I did tell him to verify his backups work and that he is able to restore from them, so that's kind of the same thing.
When it gets down to it, oppinions are like assholes; everyone has one. And most people only care about their own and don't really want to look at their coworkers'. I guess I'm the same in that respect.
First, make sure you get a SATA2 controller. NCQ is a must for multiuser environments.
Second, whatever controller you buy, grab 3 of them. RAID is great for disk failure, but people rarely think about what they'll do when the controller fails.
Look at some of the stranger RAID options. If you just use RAID5, you'll be selling yourself short. RAID3 is worth a look. I'd actually suggest you put two controllers in a machine. Run RAID0 on 4 drives on a single controller. Run RAID0 on 4 drives on the other controller. Then use Windows or Linux software RAID to run RAID1 between the two RAID0 drives. Very fast performance and fully fault tollerant.
Keep the OS on a small, slow hard drive seperate from the array. You can do funny things there, but I'd suggest you set it up properly and then use a disk clone utility to create an offline backup to store somewhere.
Arrange for testing in the first few months. Unplug drives from the array and see what happens. Verify that you can restore from tape backup to the array. Veryfy that your cloned OS hard drive can actually get the array online. Extensive testing before you go live. Two tests in the first month after going live. One test every 6 months after that.
If you are using commodity servers, get a spare for everything in there.
If you want high avalibility, look into DRBD. It's like RAID1 over a network.
Monitor the damn thing! My last job someone let the server die. It had RAID5 over 5 drives. One drive had failed and no one noticed. When the second failed, that was the end of it. Learn to use SNMP or get some good monitoring utilities that will notify you of problems. You need to know if a drive fails, if it reports SMART errors, drive temp, proc temp and usage, NIC utilization, drive utilization, system temp, and memory utilization. MRTG+RRD Tool and SNMP will give you pretty charts for all that crap. MotherboardMonitor will also give some nice readouts for Windows. If it's commodity server, look at installing a Crystalfontz LCD so that you can walk by and get a quick status without the need to login.
Another possibility is to get a single Windows box with a shit-ton of RAM. 1GB for general testing and maybe 4GB for production. Also get at least 6 physical drives divided between 3 seperate RAID controllers configured for RAID1. Redundant power supplies would be nice, but only if your environment also provides seperate power feeds in a rack.
Install whatever OS you are comfortable with. Regardless of what Lin or Win fanboys think, you'll get more uptime and beter security with an OS you are used to using.
Next, install VMWare. Create a VMware image for WinXP or 2k3 on the second RAID controller. Create a VMware image for Linux 2.4 Kernel on the third RAID controller.
Install Windows and all its server stuff on one image and install Debian stable on the other image.
Now, at first, this may seem like overkill. But you save a lot of overhead by using a single box. The actual price is comparable to a two servers. The CPU and RAM costs are cut considerably because the two servers will keep a single CPU in constant use. Compare this to dual servers where the CPU may sit idle for the vast majority of the time. Power usage for this setup will be less than dual servers and heat generation will be lower.
That has to be some of the best advice I've ever read online.
A few things I'd like at add:
First, if you stay in management, read about it. You can't be an effective programmer unless you understand your tools. You can't be an effective manager unless you do the same. Commit to reading at least one hour a day from a management book. Start with "7 Habits" and then progress to "Execution: The Discipline of Getting Things Done". After that, just browse similar areas and try to soak in as much as possible.
Second, do some traditional baselining. How many outages do you work a week? How many lines of code or milestones have you reached? Are you on or ahead of schedule? Are you under budget? Work on taking geeky information and presenting it to your boss in a money-based manner. Show him how you need overtime to patch the WMF exploit or the company could potentially lose hundreds of hours of computer time.
Third, look at streamlining the processes. Don't think of it as a "do more with less" scenario. Just look at your teams and make sure they do their thing really well. In fact, you should make your teams focus on obtaining excellence. You will have to realign the values of your people so that they actually have the desire to do the best job possible.
As do WoWs. That still does not prevent instances where 50 or more people take part in a raid. How will D&D adapt in a landscape where 50+ people expect a 5+ hour raid culminating with a single bad guy?
Slashdot Mirror
Damn straight. Your duty as a sysadmin is to ensure that the company can maxamize profits. Whenever you think about doing something, try and decide if it will make the company more money than it will cost.
Epoxy the power dongle into the laptop. Then epoxy the brick to a desk. Their only real recourse is to cut the power cord. Even then, they'll have a laptop with a cut power cord epoxied into the case.
The only thing better than a teenager, is to get two computers and hire two teenagers.
Honestly, why go for an expensive, complicated solution when a simple solution is already at hand.
5 minutes per CD gives about 12 CSs per hour.
That's 25 hours to rip 300 CDs.
$5 per hour comes in at $125. Buy a pizza for lunch over 3 days brings it to just under $200.
If you borrow a laptop or two, there is no reason one guy can't swap out CDs in 3 computers; it's be done in a day. Offer a local teen $150 + pizza for a day's work, and they'll jump at the chance.
So, unless you can come up with something less than $200, you are just shooting yourself in the foot.
>>They focus on making the most popular games work and screw the rest.
No, they work on making you think that the most popular games will work. When you actually try to install, patch, and play these games, you find out otherwise.
>>better than giving it to Microsoft I guess.
If you have a problem with MS, then why buy a game based on that system? A game sale for a MS game is the same as an OS sale for MS. The devs license DirectX, programming tools, and logos. They feed the coffers of the monster.
Me, I just wanna play a fucking game without rebooting. Fuck all that RMS GNU/Bullshit.
I tend to agree. I was a subscriber for over a year. Games that were listed as "should work" never worked for me. I used RedHat, Debian, Gentoo, no good. I tried various Nvidia cards across Intel and AMD processors.
We even had geek LAN parties where we tried to get things to work. We eventually got BF1942 to work a little. And Rainbow6 worked quiet well.
But, looking back, I think that the vast majority of people claiming success with WineX were company shills. Either that, or people didn't want to admit that their $5 a month was a complete waste.
Yeah, what he said. And while we are at it, I don't want to play flight simulators unless I can actually be in a cockpit at 15000 feet. And no driving simulators till my seat jumps every time I hit a bump. And no more AI opponents; every character in every game should be played by another player or paid actor.
We should just ignote the fact that there are technical and monitary limitations to everything. Sure, I could sit down and, with off the shelf technology, design an arcade shooter with 360-degree visibility. Of course, it'd cost $50,000 for one box. It'd burn out $1000 worth of bulbs every month. It'd be suseptable to teenagers touching the screens or spilling their sodas.
To top it all off, it'd cost $5 for a few minutes of play.
We use 17" and 19" screens for a reason. The same reason we use keyboards vice voice recognition.
But, I guess in your world everyone drinks unicorn giggles, uses their voice to provide input, and gets output via a nural link.
A former boss of mine had an interesting approach. We are hardware guys. Mostly connecting big telecom multiplexers to bigger multiplexers. Anyway, he'd invite the potential candidate into a room with a bunch of equipment and cabling. Manuals would be scattered around. He'd apoligise about the mess and tell the candidate that he'd be back in a few minutes.
After about 15 minutes, he'd come back and say that they were having a big problem. The candidate could wait or leave.
If the candidate left, he'd be removed from the pool. If he stayed, but didn't tinker with the equipment on the tables, he'd also be removed.
The candidates that stayed and read manuals or tinkered with the equipment were usually hired.
Long, long ago in a college far, far away, I had a Unix class. The classroom consisted of keyboards on desks surrounding the walls. In the middle of the room, there was a teletype.
You would hand-write out the things you needed to do. Then you'd carefully type those into the keyboard. Once you were done, you could go to the teletype and look for the output line that corresponded to your login. And there was your output.
I had Steam installed a few weeks before HL2 was released. I pre-orded HL2 and began to download it a week before release. The day of the release, I signed into Steam and was playing in 5 minutes.
It would have taken longer than that to open the shrinkwrap and install the Vivendi HL2 boxed copy.
When I was schedule to fly to the US for some conference, I clicked the "play HL2 offline" button. I played just fine on the airplane and in the Hotel without Internet access.
When I wanted to reinstall, I clicked the "backup this game" button, inserved a DVD, and let it do it's thing. After the reinstall, I clicked the "restore from backup" button and was playing agian in 10 minutes.
Don't just bash Steam when you have no idea what you are talking about. No one has any idea what will happen when Valve goes under. Maybe someone will buy the franchise and keep the servers going. Maybe Valve will release the authenticator code. Anything else is just a conspiracy theory.
BTW, if you think having a physical plastic disc is some guarentee that you'll be able to play a game in 20 years, think agian. I have a "Shogo: MAD" game and a "Jane's Fleet Command" game. Neither will play under WinXP. Sure, I could go back and install Win98, but who wants to hunt down drivers for devices made in 2005 that will operate on a 5+ year old OS.
Raptors are bad ass. I have two of the 36gb version in a RAID0 config. I can sustain over 110MB/sec. At a LAN party, I moved all my "tradable files" to this array and plugged in to the Gig-E switch with my NForce3 Gig-E adaptor.
That link was saturated the entire night.
So, you think that every person should just focus on one issue at a time? The President is perfectly capable of doing two (sometimes more) things at once.
Iraq may or may not have been a mistake. But it happened. If we leave now, it'll be even worse in 20 years. We need to clean up the Middle East (not just Iraq) now. In 50 years, Bagdad has the potential to be the same city as Tokyo or Berlin is today. We can't just leave that mess for our children (or our children's children) to die over. Every year we delay in taking out Muslim extremists (and the governments that support them) is another year for them to become more deadly.
You can't fight a war without running up a deficit.
And then Katrina happened. It's bad shit that could have been prevented, but it happened. We have no choice but to deal with it.
Education sucks. It needs to be addressed.
Outsourcing sucks. Needs to be addressed.
Health care and retirement entitlements need to be addressed.
There are about 5000 burners going full blast on the White House's stove. And not a single one is a back burner.
I think we can all agree that it would have been nice if $event had not happened and $money wasn't required to fix it. It would also be nice to think that next year will be slow and maybe The Man can clear up his inbox.
Only time will tell.
As a member of the military, I see the anti-war and pro-warrior crowd as a very strange animal. How can you support me without supporting my actions? And it's difficult to claim that I don't support my actions because I've had amply oppertunity to quit.
Some number of years ago, I found myself in charge of a private infrastructure. We had maybe 50 servers and 400 users exchanging sensitive information completely seperate from the main, public network.
Because of the percived importance of uptime on this network, everything required mountians of paperwork. Installing and removing nodes from the domain required three administrators, setting up a new machine required a month on a private VLAN being monitored by a sniffer, memory and hard drives were obselete before they got to the customer.
Anyone who ever worked around an UPS knows how they die. They give plenty of warning. Having an UPS fail is a rediculous way to lose your backbone infrastructure.
My predicessor had done a wonderful job of installing an UPS for every router and switch in the datacenter. Problem is, both power supplies in the routers and switches were connected to the same UPS. In cases where an UPS was about to fail, he unplugged the UPS from the wall and plugged it into, you guessed it, another UPS.
He didn't do it out of ineptitude; it was done because the only option was to clash heads with the IT overlords. They would require studies about how many UPSs failed and if it failed before the MTBF, they'd want us to try and recover money from the manufacturer. They'd want contractors to come in and examine the UPS to bid on a UPS monitor and replacement contract.
In short, asking the overlords was like asking to be turked by a syphalitic bear.
So, some BOFH, overwhelmed by the prospect of repairing the power system, chose another path. He walked over to a failing UPS and simply turned it off. He was the only one with the access to turn it back on, so he had no reason to worry.
Within two hours, all in-progress meetings were cancled. The Supreme Overlords demanded from on high that this lowly tech was to get a blank check and a blank trouble ticket (approved by the Supreme Overlords) to do whatever he needed to do to prevent that from ever happening agian.
Electricians installed two seperate power feeds into every rack.
Each power supply got a seperate UPS.
Old equipment was updated.
Everything was strawberry fields and unicorn giggles after that for the infrastructure department.
Now, to answer your question: You have something that someone wants. Hold it hostage till you get what you need.
The latest SF protection requirs a reboot. I installed X3 and Silent Hunter 3. The first time I tried to play, it errored because I was running Daemon Tools. I exited DT and tried agian. SF informed me that protection had been applied and then told me to reboot.
To me, forcing a reboot is "unstable".
If you are going to do a stage 3 or a GRP, you might as well use Debian. Deb has packages for GCC and the kernel compiled for virtually every acrhetecture out there.
I know what you mean about "other things". I love never having to deal with RMP hell agian.
The transition from Debian-based systems to Gentoo will be fairly painless. Just understand that it may be several days before you get a working system. My main computer is a dual-boot for WinXP and Gentoo. During the install, the family had to live without being able to access WinXP for almost 5 days. Of course, this was on a mid-range AthlonXP with 256MB RAM. I have (in boxes at the moment) a new dual-core Athlon. I expect it to take about 48 hours from fdisk to KDE.
There are precompiled packages avalible to speed things up, but where's the fun in that.
Yeah, Debian to Gentoo is quite easy. Just use "emerge" vice "apt-get".
AdBlock and FlashBlock are designed to block ads and flash. Modifying your hosts file to block spyware is a false sense of security. Dangerously false.
If you use Windows, AutoUpdate at least weekly. Nightly may be overkill, but isn't really hurting anything.
Turn on the Firewall and do not allow exceptions unless you know what they are for.
Install and use Opera or Firefox.
Install and update AVG and/or Avast. Norton is overkill for most home users. Why pay $50 for something only marginally better than the free editions?
Install and run SpyBot, MS Ad Scanner, and AdAware.
If you get any Spy/Adware, your only real recourse is to treat it like a virus; nothing can be trusted. You should back up your data and reformat/reinstall. Then scan your backups for malicious programs before you restore them.
An Ad-Blocking Hosts file is a dumb suggestion. If you can modify the Hosts file, what makes you think that a program you launch can't modify the same file?
And before you suggest running as a non-admin user, don't forget that a lot of programs will not run properly unless you have admin rights.
Now, I guess you could put the hosts file on a floppy and write-protect that. Then you can create a symlink to the file on the floppy.
An additional thing about RAID. Always reboot to rebuild your array. If at all possible, come in at like 3am on a Sunday to replace a failed drive. Shutdown and remove the failed drive. Insert a new drive and turn the power back on. Never trust the "hot swap" scam.
Seriously, downtime isn't that hard to get. Especially if you tell them that not granting a reboot window could potentially cost them 5+ hours of unscheduled downtime.
If you work in a place where they refuse to grant downtime, have your boss draft a memo stating that you advised him of the potential problems and he will accept all responsibility in case of an outage.
If he won't do that, then you draft the memo and have a coworker sign statiing that the boss ordered you to hot swap agianst your wishes.
The chances of losing two disks at once are slim. RAID 0+1 will provide great performance and good fault tolerance if you react to problems as they happen.
But I guess it depends on what your users need. If they need raw throughput, RAID 0+1 is better. If they need low latency, then RAID 10 may be the answer. Or maybe both systems would fall within the margin of error of each other.
In any event, once you get into what-if situations, no RAID will be good enough. What if you lose a disk? What about two? Five? Well, what if lightning hits the chasis or the janitor unplugs it to buff the floor?
The best you can do is roll the dice and play the odds. You'll see that I told him to use RAID 0+1. I also told him to use good monitoring setups to mitigate problems. I also suggested a tape backup. Actually, maybe I didn't, but I did tell him to verify his backups work and that he is able to restore from them, so that's kind of the same thing.
When it gets down to it, oppinions are like assholes; everyone has one. And most people only care about their own and don't really want to look at their coworkers'. I guess I'm the same in that respect.
Some more tips for you:
First, make sure you get a SATA2 controller. NCQ is a must for multiuser environments.
Second, whatever controller you buy, grab 3 of them. RAID is great for disk failure, but people rarely think about what they'll do when the controller fails.
Look at some of the stranger RAID options. If you just use RAID5, you'll be selling yourself short. RAID3 is worth a look. I'd actually suggest you put two controllers in a machine. Run RAID0 on 4 drives on a single controller. Run RAID0 on 4 drives on the other controller. Then use Windows or Linux software RAID to run RAID1 between the two RAID0 drives. Very fast performance and fully fault tollerant.
Keep the OS on a small, slow hard drive seperate from the array. You can do funny things there, but I'd suggest you set it up properly and then use a disk clone utility to create an offline backup to store somewhere.
Arrange for testing in the first few months. Unplug drives from the array and see what happens. Verify that you can restore from tape backup to the array. Veryfy that your cloned OS hard drive can actually get the array online. Extensive testing before you go live. Two tests in the first month after going live. One test every 6 months after that.
If you are using commodity servers, get a spare for everything in there.
If you want high avalibility, look into DRBD. It's like RAID1 over a network.
Monitor the damn thing! My last job someone let the server die. It had RAID5 over 5 drives. One drive had failed and no one noticed. When the second failed, that was the end of it. Learn to use SNMP or get some good monitoring utilities that will notify you of problems. You need to know if a drive fails, if it reports SMART errors, drive temp, proc temp and usage, NIC utilization, drive utilization, system temp, and memory utilization. MRTG+RRD Tool and SNMP will give you pretty charts for all that crap. MotherboardMonitor will also give some nice readouts for Windows. If it's commodity server, look at installing a Crystalfontz LCD so that you can walk by and get a quick status without the need to login.
Another possibility is to get a single Windows box with a shit-ton of RAM. 1GB for general testing and maybe 4GB for production. Also get at least 6 physical drives divided between 3 seperate RAID controllers configured for RAID1. Redundant power supplies would be nice, but only if your environment also provides seperate power feeds in a rack.
Install whatever OS you are comfortable with. Regardless of what Lin or Win fanboys think, you'll get more uptime and beter security with an OS you are used to using.
Next, install VMWare. Create a VMware image for WinXP or 2k3 on the second RAID controller. Create a VMware image for Linux 2.4 Kernel on the third RAID controller.
Install Windows and all its server stuff on one image and install Debian stable on the other image.
Now, at first, this may seem like overkill. But you save a lot of overhead by using a single box. The actual price is comparable to a two servers. The CPU and RAM costs are cut considerably because the two servers will keep a single CPU in constant use. Compare this to dual servers where the CPU may sit idle for the vast majority of the time. Power usage for this setup will be less than dual servers and heat generation will be lower.
That has to be some of the best advice I've ever read online.
A few things I'd like at add:
First, if you stay in management, read about it. You can't be an effective programmer unless you understand your tools. You can't be an effective manager unless you do the same. Commit to reading at least one hour a day from a management book. Start with "7 Habits" and then progress to "Execution: The Discipline of Getting Things Done". After that, just browse similar areas and try to soak in as much as possible.
Second, do some traditional baselining. How many outages do you work a week? How many lines of code or milestones have you reached? Are you on or ahead of schedule? Are you under budget? Work on taking geeky information and presenting it to your boss in a money-based manner. Show him how you need overtime to patch the WMF exploit or the company could potentially lose hundreds of hours of computer time.
Third, look at streamlining the processes. Don't think of it as a "do more with less" scenario. Just look at your teams and make sure they do their thing really well. In fact, you should make your teams focus on obtaining excellence. You will have to realign the values of your people so that they actually have the desire to do the best job possible.
As do WoWs. That still does not prevent instances where 50 or more people take part in a raid. How will D&D adapt in a landscape where 50+ people expect a 5+ hour raid culminating with a single bad guy?