Caution: This paragraph will say Windows Restore has... issues, but the Windows Backup may be OK.
The major problem with system restore is that you can't mark any specific restore point "good". What you are left with are a few system restore points that usually are somewhat corrupted by the time you need them: EG: The Good restore points are wiped by infected ones by the time Mom tells you her computer was acting funny. You know she really didn't want to bother you about it... If you could mark a restore point as known, it would be better. (I would probably keep the initial set + just after each SP.).
To be fair, Windows Backup appears not to suck any more.. at least on 2008. (I have not seen Windows Vista or 7 yet.)Server wants a separate disk for the backup, You can pop in an install disk & use that backup image to restore the entire system. (although I don't think it's recommended for exchange...) If the backup on 7 is as good; can let me target a network share... that's a huge improvement for ms.
(has anyone tried this on 7? I'm too lazy to look this feature up. Kinda like OP who didn't notice that he made so little space for system restore that it overwrote each and every one.)
It's kinda like a backup that's available on Disk, you know - It takes a backup of system files on every successful boot. - it provides fast access to those backups if you need them. - best of all (for most users), it's infected with a virus by the time you want to use it;)
I just spent the weekend at work, due to an apparently "new" rootkit that hit our network.
Friday
9AM, Ticket was submitted to MaBigVendee (with sample of affected file).
At 3 PM: Admins had ticket escalated due to lack of response. (120 workstations affected)
At 6 PM: MaBigVendee responded that we did have "unwanted software" and asked for us to make some logs (Using Process Monitor; the vendor application internal scan log).
-> but hey: Try our latest beta virus definitions file.... it should work....
By 9 PM: The initial link for ftp sent by the vendor did not work; while the ftp client on our side said the second link worked & accepted & completed the upload... Ticket is escalated to the highest level it can go to. I am attached to the incident.
At 11 PM: We have shut down much or our core network. The contact and phone number I got with ticket escalation to the highest level of handling do not work. I have to sit in the main queue to contact anyone: who states something like: we have no files and can do no work on the issue.
I do not have a copy of the files; all engineers who were on the case
Saturday:
~9:30 AM: 20MB log file is broken into a multi-part zip file & emailed. (Don't forget to change the extension to.txt so it can stay as an attachment.)
11:30 AM: MaBigVendee states they are missing part of the zip. I get the copy I was cc'd on & forward it again.:: Someone is getting security to let them into offices to collect instances of affected laptops.
12:30 PM: MaBigVendee states that the logs sent were useless: Asks if we got the alert during them: (I did not know then, but it turns out MaBigVendee asked for us to create the alert popup condition while the log was being created. This was done per request.)
At this point, we are asked to test another virus definitions file: Why: no reason I can tell. I hear various refrains on response was delayed due to a lack of information being sent to the vendor. Apparently all information was sent to the vendor & vendor is so large that one hand is unaware of the other. (EG: Concierge service is unaware of anything touched by phone monkeys; researchers can neither access corporate nor concierge resources.)
12:45 MaBigVendee remote assistance site is blocked at the web-proxy due to environmental restrictions applied by admins:
2:15 PM Another admin & I go way out of policy to get MaBigVendee access to an instance of an infected workstation. MaBigVendee researchers play on this workstation for the next several hours.
At some point vendor asks for a VM of the infected machine: but seem to want the machine there researchers are attached to.. don't start until the researchers finish.
9PM: We get bored and VM another instance of an infected machine:
11:45 PM: Call MaBigVendee concierge service number: No response... leave message no response: go home.
Sunday:
9AM: Vendor will get to office to pick up VM image at noonish.
12:00: Cannot replicate symptoms in Virtual environment. Point this out to vendor. ("Hey, you know how some programs check to see if they are in a virtual environment & shutdown? I think this may be one.")
MaBigVendee response: "We don't think you virtualized an infected machine." Can you understand how insulting this is? I mean seriously? Go show vendor's guy Virtualization Log Summary on infected machine & that that machine has the issue we are trying to get resolved.
More Boring frustrating stuff here.
By 7:PM Vendor finally tests a method of removing "unwanted software", but neither of the vendor tools (2) that we own & could be used to push out a Virus definations file & force a full scan will work. We will have to wait for an approved definition, or sneaker-net the beta to 250+ workstations.
So: Having a company that will actually respond and put researchers on the problem is a good part of having a competent company, but big is no panacea & may work against you.
Personally, I think Microsoft has much better rootkit d
I saw it on a website yesterday. Therefore the correct answer is a difficult to use, obtuse unix variant: Therefor, try a BSD variant. Tell em it runs VMWare (an OpenBSD if I'm interpreting some of the config files correctly.) How about PC BSD.
Seriously: go to Distrowatch.com Scroll down the right-hand side & look at the current page-hit ranking. Try the top ten & see how they work for you & offer to install one for your users. You will be the ONLY tech support for converted users & will have to train people who no nothing about computers how to google for themselves. Be aware of what you are getting into.
The Government does not pay all that well (and previously less well). You are talking about large networks, that are very complicated. As a result, you do not have a whole lot of government staff with experience to run a network that is that complicated.
I work in a very small (5K users) government (federal) office. I have to deal with 12 windows domains, 11 Political groups, and offer support to all Regional Admins, and departmental admins - as well as dealing with a help desk which has been told "we don't investigate error logs."
Unfortunately, some of the government staff can't find their ***es with both hands. This is because 12 years ago, the government paid much less than the contractors. Good technical people could earn twice a much contracting a working for the government. Those people are still contracting (mostly), and are the ones that you would want in the government running the show. The people who have "more senior" positions in gvt now? They are largely the ones who couldn't get the better paid contracting jobs, and state: Helpdesk personnel should not be investigating application event logs.
Furthermore, this is also the case for many large businesses: They outsourced the tech support years ago (cheaper); most users get someone in india to change passwords, while sr. staff get concierge service. Those large businesses have similar issues as well: but they have an explicit 2-tier service system.
It's been going on for years, but I don't see any way to rectify it: especially as the job listings still seem to be opaque, and difficult to decode.
Yes, Two stroke engines have been around for a long time. However, this engine purports to be a clean two stroke - something that has not been around a long time. Anyone with an mid-70's two stroke motorcycle could probably go around the block before biking in their own smoke - so yes, this is new.
The advantage of this "system" is obviously 1) it's light, 2) it's clean; 3) it can use multiple fuel types. 1) A light engine can be combined with a generator; a battery. Think Electric-Car.
If the battery in an electric car is large enough to run ~30 miles; the car has a sufficiently strong auxillary motor (not enough to drive the car fast uphill, but enough to repower the battery between the downhill & uphill) - this makes an electric type car better. A "more complex" two stroke should be lighter than a four stroke; make the Electric car significantly better. (Personally, I drive under six miles most days. Occasionally I want to visit friends who live outside the range for a purely electric vehicle - requiring me to have a conventional vehicle, or an expensive one with multiple power systems.)
2) If the engine is as clean as a four stroke, then the engine is as clean as a four stroke. EG: you will be able to use it in a production vehicle without as much pollution as a conventional two stroke.
3) It can use multiple fuel types: EG: You can fill it with Gas, Diesel, Algie-Diesel - or if you're in a 3rd world country: you can use Strained Fryer Grease (Diesel Fuel) from Bob's Yak stand. (May only work in warm climates, not recommended for stoned hippies, etc...)
So yes, if this works as implied this is a good solution that represents a significant improvement over a four stroke engine. (Not to say that the moving-puch cylinder head would not work in a four-banger.) For a company that makes very light vehicles, and is working on an "electric-type" vehicle - this solution makes emminent sense. Please insert this in your tin-foil hat so the Govenment does not leak it to the Big Oil companies.
Guys, Us stewpid windows guys don't haff to know this stuuf. Micro$oft locks out any account that has 3 failed logons withing 15 minutes by default. (Not that it would not be trivial to get around, it just means that you have to try each password for a specific account once per five mintues. And increases the amount of time it takes to break a password. Hopefully to longer than the period to change the password....)
Dude, you misunderstand: You wanted the "Department of Inadvisably Applied Shiny Shit." This is the "Department of Keeping Crap Moving." If you wanted Shiny Shit, you are in the wrong department - cause this is a crappy job.
I suspect that managability means that someone (who is familiar with group policy objects) can manage a myriad of IE settings and restrictions across many machines in one go.
To do this with Firefox (and someone please correct/confirm me), you need to push a firefox configuration file to many machines at once & lock it down.
In the case of my office, we lock every IE user to machine configuration registry keys, and set those with a GPO. End users cannot reconfigure their IE settings to allow things like active-x,
With Firefox, while I suspect it's just as configurable - I also suspect that it's slightly more difficule, and less documented.
"certs for linux but anyone who knows anything in HR will sneer at them as the meaningless drivel they are."
Anyone who knows anyone in HR is nobody. I work for a fairly savy IT company, and have worked for several other IT companies. The percentage of people in HR who know about technology close to 0%.
If those people knew about IT, they woule be called managers, not HR. In their defense have to know a verly little bit about a wide variety of work sectorys, and a lot about HR.
Take a look at some of the postings out there: "We want a MCSE/MCITS certified administrator, who is also AIX certified and certified as a CCIE. Pay
If you really want to move from help desk to a *glorified position track*, figure out which one you want. You want to be a MS admin, figure out which technologies sound interesting to you. Then figure out how to get your MSITS (or whatever they are calling MCSE these days.) Get your company to *lend* you a good computer or two to set up a virtual environment to test this. If you want to do Linux admin, try the same cert BS with Linux stuff.
As a techie,I may think that certifications are bogus, and only serve to tell you what advertized features actually work - but they tell a HR drone something else: 1)You know enough about the tech to have passed a vendor test. 2)The vendor *may* help you more than some uncertified schmuck. 3) You *may* know how to learn, and be able to apply it in a beneficial way to the company.
Personally, as a SR Sys Support Specialist (Dealing with MS, VMware, and Citrix mostly), I find that I have help desk zombies interrupting me every five minutes with issues that they should have done themselves...
It sounds like you want a backup to store the drive allready in the computer, although it could be you don't have enough storage and are just storing files on external drives. (Say movies ripped from DVD or so...)
In either case, it's probably easiest to make a network attached storage device (aka Linux server) to copy everything to.
Computer1: Primary use computer
OldComputer2: NAS in closet... You can get an old P3 (low heat producing) with a bunch of drive bays, and a PCI SATA card ($50). Use a junky IDE drive for the OS, and make a raid5 of several large capacity drives.
The advantage of this is that you can synch your existing "workstation" to the NAS, and get the files you were considering on external HD's on the fly. Moving HDs around is not really recommended, as there is a good chance you will damage them. There are many guides to configuring this, which you can find via google.
Note: a PCI SATA adaptor will limit you to about 1Gb/s throughput. Convient as a Gb Ethernet is the current networking standard.
In a environment as described, everyone knows to buy one of what allready exists. In my office it's a Latitude e6400 (latest and heaviest;). With a PKI card reader. This configuration unit is a US government special. The "Secure Offices" poster is going to have an equilivent set of Super Sucky Specials.
However, the key is not the purchasing what everyone else has - it's that the question is relatively bogus: In most secure environments, you have to use company equipment, and have papers stating you are allowed to take it out of the building. In the few courthouses I've been in (DC, Ohio; Federal) I was not allowed to bring in electronics. It may be that poster has a specific ? regarding courts - but the question is overgeneralized to the point of misleading: Secure US government facilities don't let you use your own equipment. (My insecure one will let you bring it in, but no touching the network: The night security guys have EEEs with wireless modems for between log checks.)
You need DOS and Win96 compatibility: You can virtualize the existing system into a new system, and make it portible; back-up-able (as a Virtual Machine) by virtualizing it.
As an aside, I always thought Win95 was a dog. You may wish to check to see if XP compatibility mode will work, or check (ha ha) to see if WINE will work. (Actually, trying the application set with WINE is not a bad idea - it should be compatible with Windows 95 by now.)
Remember it could be worse: I have a friend who deals with Vet who has an old Xenix system - they buy parts of ebay in bulk;)
The VMWare Player is free, as is VMWare Converter.
1)Create new image on box (smallish disk). Update same 2)Create an image of that box's C: Drive, place on another drive 3)Make copy of that image file/folder 3)Run that image, throw away when done.
VMWare server is also free, and not hard to get running on Linux.. on WinServer 2k3 it's a doddle.
Went through initial check at counter, went through security, got to gate.. adroid: Do you have any weapons or explosives? me: Just some nitroglyceryn (sp?) that I drove from DC (about 40 miles), through every pothole I could find. adroid: proceeds to call FBI paraphrased FBI: Why are you an ass me: dude, you are the jerk being an ass. The security here is asstastic: An idiot could drive through that chain link fence and blow up a few planes on the runway. Or just ship bombs with altitude detonaitors: plane goes up, bomb goes off. People and their stuff are only 1/3rd of the "cargo" on a typical us airliner... paraphrased FBI: Dude, you are so going to jail....
There are a few different solutions to your question, and many of them depend in the Webserver used.
For Windows, you do not need to use clustering for the webserver: you can use Network Load Balancing. If you have a SQL back-end, then you will need clustering...
To explain:
Clustering is used for programs that can only be run in single instance (Exchange, SQL, etc...) IIS (Microsoft's Web Server) can be run in multiple instances. Therefore you can use Network Load Balancing (NLB). With NLB, the request will be directed to any available server, if one goes down - the other is available.
NLB can be implemented (badly) with round-robin DNS - which will send the request to each server in sequence.
These same techniques can be used in Linux (see thread), and other Unices. For the Web-Front-end, use some form of NLB. For the back-end, use clustering.
At this point, I should point out that eventually you will have a single point of failure-unless you are very careful. (Do you have a redundant SAN? Are all etherent paths seperate and redundant? How about power to your server space?) I should also point out that everything above is overly simplified.
I certainly know that they have not helped with murders around the block from my house. Essentially, the cameras are never pointed in the right direction at the right time, and have never had the tape used as evidence for court purposes.
The last time I looked at Haiku/BeOS it required a PPC computer, long after they were generally available. (I'm not sure Mac's even used them any more.) Looking at the website (http://www.haikuware.com) it looks like they are shooting for i586+ hardware, and the supported stuff is a bit.. older.
Perhaps I could try it, but I'll need a better hardware database before I take the time. There are 0 systems that are well rated (of 5), and few motherboards. Let me know when more hardware works: If it does not mostly work with stuff in my junk drawer - I'm not buying a new system to test it. Even if it was the bees knees fifteen years ago.
Slashdot Mirror
Duh.
Caution: This paragraph will say Windows Restore has ... issues, but the Windows Backup may be OK.
The major problem with system restore is that you can't mark any specific restore point "good". What you are left with are a few system restore points that usually are somewhat corrupted by the time you need them: EG: The Good restore points are wiped by infected ones by the time Mom tells you her computer was acting funny. You know she really didn't want to bother you about it... If you could mark a restore point as known, it would be better. (I would probably keep the initial set + just after each SP.).
To be fair, Windows Backup appears not to suck any more .. at least on 2008. (I have not seen Windows Vista or 7 yet.)Server wants a separate disk for the backup, You can pop in an install disk & use that backup image to restore the entire system. (although I don't think it's recommended for exchange...) If the backup on 7 is as good; can let me target a network share... that's a huge improvement for ms.
(has anyone tried this on 7? I'm too lazy to look this feature up. Kinda like OP who didn't notice that he made so little space for system restore that it overwrote each and every one.)
It's kinda like a backup that's available on Disk, you know
- It takes a backup of system files on every successful boot.
- it provides fast access to those backups if you need them.
- best of all (for most users), it's infected with a virus by the time you want to use it;)
with the requirement that you have ten times the processing power.
Let Lesson
always preview use
or smash text
I just spent the weekend at work, due to an apparently "new" rootkit that hit our network. Friday 9AM, Ticket was submitted to MaBigVendee (with sample of affected file). At 3 PM: Admins had ticket escalated due to lack of response. (120 workstations affected) At 6 PM: MaBigVendee responded that we did have "unwanted software" and asked for us to make some logs (Using Process Monitor; the vendor application internal scan log). -> but hey: Try our latest beta virus definitions file.... it should work.... By 9 PM: The initial link for ftp sent by the vendor did not work; while the ftp client on our side said the second link worked & accepted & completed the upload ... Ticket is escalated to the highest level it can go to. I am attached to the incident.
At 11 PM: We have shut down much or our core network. The contact and phone number I got with ticket escalation to the highest level of handling do not work. I have to sit in the main queue to contact anyone: who states something like: we have no files and can do no work on the issue.
I do not have a copy of the files; all engineers who were on the case
Saturday:
~9:30 AM: 20MB log file is broken into a multi-part zip file & emailed. (Don't forget to change the extension to .txt so it can stay as an attachment.)
11:30 AM: MaBigVendee states they are missing part of the zip. I get the copy I was cc'd on & forward it again.:: Someone is getting security to let them into offices to collect instances of affected laptops.
12:30 PM: MaBigVendee states that the logs sent were useless: Asks if we got the alert during them: (I did not know then, but it turns out MaBigVendee asked for us to create the alert popup condition while the log was being created. This was done per request.)
At this point, we are asked to test another virus definitions file: Why: no reason I can tell. I hear various refrains on response was delayed due to a lack of information being sent to the vendor. Apparently all information was sent to the vendor & vendor is so large that one hand is unaware of the other. (EG: Concierge service is unaware of anything touched by phone monkeys; researchers can neither access corporate nor concierge resources.)
12:45 MaBigVendee remote assistance site is blocked at the web-proxy due to environmental restrictions applied by admins:
2:15 PM Another admin & I go way out of policy to get MaBigVendee access to an instance of an infected workstation. MaBigVendee researchers play on this workstation for the next several hours.
At some point vendor asks for a VM of the infected machine: but seem to want the machine there researchers are attached to .. don't start until the researchers finish.
9PM: We get bored and VM another instance of an infected machine:
11:45 PM: Call MaBigVendee concierge service number: No response... leave message no response: go home.
Sunday:
9AM: Vendor will get to office to pick up VM image at noonish.
12:00: Cannot replicate symptoms in Virtual environment. Point this out to vendor. ("Hey, you know how some programs check to see if they are in a virtual environment & shutdown? I think this may be one.")
MaBigVendee response: "We don't think you virtualized an infected machine." Can you understand how insulting this is? I mean seriously? Go show vendor's guy Virtualization Log Summary on infected machine & that that machine has the issue we are trying to get resolved.
More Boring frustrating stuff here.
By 7:PM Vendor finally tests a method of removing "unwanted software", but neither of the vendor tools (2) that we own & could be used to push out a Virus definations file & force a full scan will work. We will have to wait for an approved definition, or sneaker-net the beta to 250+ workstations.
So: Having a company that will actually respond and put researchers on the problem is a good part of having a competent company, but big is no panacea & may work against you.
Personally, I think Microsoft has much better rootkit d
I saw it on a website yesterday. Therefore the correct answer is a difficult to use, obtuse unix variant: Therefor, try a BSD variant. Tell em it runs VMWare (an OpenBSD if I'm interpreting some of the config files correctly.) How about PC BSD.
Seriously: go to Distrowatch.com Scroll down the right-hand side & look at the current page-hit ranking. Try the top ten & see how they work for you & offer to install one for your users. You will be the ONLY tech support for converted users & will have to train people who no nothing about computers how to google for themselves. Be aware of what you are getting into.
The Government does not pay all that well (and previously less well). You are talking about large networks, that are very complicated. As a result, you do not have a whole lot of government staff with experience to run a network that is that complicated.
I work in a very small (5K users) government (federal) office. I have to deal with 12 windows domains, 11 Political groups, and offer support to all Regional Admins, and departmental admins - as well as dealing with a help desk which has been told "we don't investigate error logs."
Unfortunately, some of the government staff can't find their ***es with both hands. This is because 12 years ago, the government paid much less than the contractors. Good technical people could earn twice a much contracting a working for the government. Those people are still contracting (mostly), and are the ones that you would want in the government running the show. The people who have "more senior" positions in gvt now? They are largely the ones who couldn't get the better paid contracting jobs, and state: Helpdesk personnel should not be investigating application event logs.
Furthermore, this is also the case for many large businesses: They outsourced the tech support years ago (cheaper); most users get someone in india to change passwords, while sr. staff get concierge service. Those large businesses have similar issues as well: but they have an explicit 2-tier service system.
It's been going on for years, but I don't see any way to rectify it: especially as the job listings still seem to be opaque, and difficult to decode.
Yes, Two stroke engines have been around for a long time. However, this engine purports to be a clean two stroke - something that has not been around a long time. Anyone with an mid-70's two stroke motorcycle could probably go around the block before biking in their own smoke - so yes, this is new.
The advantage of this "system" is obviously 1) it's light, 2) it's clean; 3) it can use multiple fuel types.
1) A light engine can be combined with a generator; a battery. Think Electric-Car.
If the battery in an electric car is large enough to run ~30 miles; the car has a sufficiently strong auxillary motor (not enough to drive the car fast uphill, but enough to repower the battery between the downhill & uphill) - this makes an electric type car better. A "more complex" two stroke should be lighter than a four stroke; make the Electric car significantly better. (Personally, I drive under six miles most days. Occasionally I want to visit friends who live outside the range for a purely electric vehicle - requiring me to have a conventional vehicle, or an expensive one with multiple power systems.)
2) If the engine is as clean as a four stroke, then the engine is as clean as a four stroke. EG: you will be able to use it in a production vehicle without as much pollution as a conventional two stroke.
3) It can use multiple fuel types: EG: You can fill it with Gas, Diesel, Algie-Diesel - or if you're in a 3rd world country: you can use Strained Fryer Grease (Diesel Fuel) from Bob's Yak stand. (May only work in warm climates, not recommended for stoned hippies, etc...)
So yes, if this works as implied this is a good solution that represents a significant improvement over a four stroke engine. (Not to say that the moving-puch cylinder head would not work in a four-banger.) For a company that makes very light vehicles, and is working on an "electric-type" vehicle - this solution makes emminent sense. Please insert this in your tin-foil hat so the Govenment does not leak it to the Big Oil companies.
Re: Windows Admins...
Guys, Us stewpid windows guys don't haff to know this stuuf.
Micro$oft locks out any account that has 3 failed logons withing 15 minutes by default. (Not that it would not be trivial to get around, it just means that you have to try each password for a specific account once per five mintues. And increases the amount of time it takes to break a password. Hopefully to longer than the period to change the password....)
duhhhhhh!
Dude, you misunderstand:
You wanted the "Department of Inadvisably Applied Shiny Shit."
This is the "Department of Keeping Crap Moving."
If you wanted Shiny Shit, you are in the wrong department - cause this is a crappy job.
As I recall Apple (DRM) was stating that jailbreaking cellphones was something to be done by terrorists who want to destroy cellphone infrastructure.
Interesting that a SMS message can destroy apples;)
I suspect that managability means that someone (who is familiar with group policy objects) can manage a myriad of IE settings and restrictions across many machines in one go.
To do this with Firefox (and someone please correct/confirm me), you need to push a firefox configuration file to many machines at once & lock it down.
In the case of my office, we lock every IE user to machine configuration registry keys, and set those with a GPO. End users cannot reconfigure their IE settings to allow things like active-x,
With Firefox, while I suspect it's just as configurable - I also suspect that it's slightly more difficule, and less documented.
$.02
"certs for linux but anyone who knows anything in HR will sneer at them as the meaningless drivel they are."
Anyone who knows anyone in HR is nobody. I work for a fairly savy IT company, and have worked for several other IT companies. The percentage of people in HR who know about technology close to 0%.
If those people knew about IT, they woule be called managers, not HR. In their defense have to know a verly little bit about a wide variety of work sectorys, and a lot about HR.
Take a look at some of the postings out there: "We want a MCSE/MCITS certified administrator, who is also AIX certified and certified as a CCIE. Pay
If you really want to move from help desk to a *glorified position track*, figure out which one you want. You want to be a MS admin, figure out which technologies sound interesting to you. Then figure out how to get your MSITS (or whatever they are calling MCSE these days.) Get your company to *lend* you a good computer or two to set up a virtual environment to test this. If you want to do Linux admin, try the same cert BS with Linux stuff.
As a techie,I may think that certifications are bogus, and only serve to tell you what advertized features actually work - but they tell a HR drone something else: 1)You know enough about the tech to have passed a vendor test. 2)The vendor *may* help you more than some uncertified schmuck. 3) You *may* know how to learn, and be able to apply it in a beneficial way to the company.
Personally, as a SR Sys Support Specialist (Dealing with MS, VMware, and Citrix mostly), I find that I have help desk zombies interrupting me every five minutes with issues that they should have done themselves...
Actually, from work on IE7 Slashdot looks almost as bad. Half the time I can't see response titles.
It sounds like you want a backup to store the drive allready in the computer, although it could be you don't have enough storage and are just storing files on external drives. (Say movies ripped from DVD or so...)
In either case, it's probably easiest to make a network attached storage device (aka Linux server) to copy everything to.
Computer1: Primary use computer
OldComputer2: NAS in closet... You can get an old P3 (low heat producing) with a bunch of drive bays, and a PCI SATA card ($50). Use a junky IDE drive for the OS, and make a raid5 of several large capacity drives.
The advantage of this is that you can synch your existing "workstation" to the NAS, and get the files you were considering on external HD's on the fly. Moving HDs around is not really recommended, as there is a good chance you will damage them. There are many guides to configuring this, which you can find via google.
Note: a PCI SATA adaptor will limit you to about 1Gb/s throughput. Convient as a Gb Ethernet is the current networking standard.
I call bogus question.
In a environment as described, everyone knows to buy one of what allready exists. In my office it's a Latitude e6400 (latest and heaviest;). With a PKI card reader. This configuration unit is a US government special. The "Secure Offices" poster is going to have an equilivent set of Super Sucky Specials.
However, the key is not the purchasing what everyone else has - it's that the question is relatively bogus: In most secure environments, you have to use company equipment, and have papers stating you are allowed to take it out of the building. In the few courthouses I've been in (DC, Ohio; Federal) I was not allowed to bring in electronics. It may be that poster has a specific ? regarding courts - but the question is overgeneralized to the point of misleading: Secure US government facilities don't let you use your own equipment. (My insecure one will let you bring it in, but no touching the network: The night security guys have EEEs with wireless modems for between log checks.)
Mod parent up.
You need DOS and Win96 compatibility: You can virtualize the existing system into a new system, and make it portible; back-up-able (as a Virtual Machine) by virtualizing it.
As an aside, I always thought Win95 was a dog. You may wish to check to see if XP compatibility mode will work, or check (ha ha) to see if WINE will work. (Actually, trying the application set with WINE is not a bad idea - it should be compatible with Windows 95 by now.)
Remember it could be worse: I have a friend who deals with Vet who has an old Xenix system - they buy parts of ebay in bulk;)
The VMWare Player is free, as is VMWare Converter.
1)Create new image on box (smallish disk). Update same
2)Create an image of that box's C: Drive, place on another drive
3)Make copy of that image file/folder
3)Run that image, throw away when done.
VMWare server is also free, and not hard to get running on Linux.. on WinServer 2k3 it's a doddle.
This was about 1998:
Went through initial check at counter, went through security, got to gate..
adroid: Do you have any weapons or explosives?
me: Just some nitroglyceryn (sp?) that I drove from DC (about 40 miles), through every pothole I could find.
adroid: proceeds to call FBI
paraphrased FBI: Why are you an ass
me: dude, you are the jerk being an ass. The security here is asstastic: An idiot could drive through that chain link fence and blow up a few planes on the runway. Or just ship bombs with altitude detonaitors: plane goes up, bomb goes off. People and their stuff are only 1/3rd of the "cargo" on a typical us airliner...
paraphrased FBI: Dude, you are so going to jail....
There are a few different solutions to your question, and many of them depend in the Webserver used.
For Windows, you do not need to use clustering for the webserver: you can use Network Load Balancing. If you have a SQL back-end, then you will need clustering...
To explain:
Clustering is used for programs that can only be run in single instance (Exchange, SQL, etc...)
IIS (Microsoft's Web Server) can be run in multiple instances. Therefore you can use Network Load Balancing (NLB). With NLB, the request will be directed to any available server, if one goes down - the other is available.
NLB can be implemented (badly) with round-robin DNS - which will send the request to each server in sequence.
These same techniques can be used in Linux (see thread), and other Unices. For the Web-Front-end, use some form of NLB. For the back-end, use clustering.
At this point, I should point out that eventually you will have a single point of failure-unless you are very careful. (Do you have a redundant SAN? Are all etherent paths seperate and redundant? How about power to your server space?)
I should also point out that everything above is overly simplified.
Actually, I have been arrested for being that obnoxious at the airport...
The washington city paper has a story on the waste of resources, known as the survelience cameras in DC.
http://www.washingtoncitypaper.com/display.php?id=36798
I certainly know that they have not helped with murders around the block from my house. Essentially, the cameras are never pointed in the right direction at the right time, and have never had the tape used as evidence for court purposes.
The last time I looked at Haiku/BeOS it required a PPC computer, long after they were generally available. (I'm not sure Mac's even used them any more.) Looking at the website (http://www.haikuware.com) it looks like they are shooting for i586+ hardware, and the supported stuff is a bit .. older.
Perhaps I could try it, but I'll need a better hardware database before I take the time. There are 0 systems that are well rated (of 5), and few motherboards. Let me know when more hardware works: If it does not mostly work with stuff in my junk drawer - I'm not buying a new system to test it. Even if it was the bees knees fifteen years ago.
However,
Good work guys!
Don't know, all my posts don't get modded up...