Techsupportwhiz
asked on
DIs there a better explaination for a slow system other than it's 5 years old?
I am dealing with a company that has 5 Windows 2000 pro desktops (1GHz processors, 128 MB RAM) that are 5 years old. they were on a windows 2000 domain. that server was replaced with a win 2k3 SBS server and each machine was joined to the domain. 1 machine - the office manager's - has gotten very slow for opening apps, logging on, logging off, etc. he said it started soon after the switch of the servers.
is the fact that the desktops are 5 years old (and the related bloat in the registry, etc) enough of an answer? that machine is doing the backups to tape using backup exec dektop (again, this is a desktop running windows 2K pro). the other desktops are not as slow, but don't see as much use. that machine gets on the web via dial up, there's no other internet access / outside contact with the real world. All the desktops are wireless to the router, which is wired to the server. the server is a single nic machine
I've checked things like dns server is the server for each desktop, etc. there's no spyware / viruses on his machine... it's just really slow. while there are likely loads of things you could offer to have me check, when you factor in the cost to have a tech (me) do that, again, is it wrong to say 'it's old, the OS gets bloated, etc... and push for a new machine? What is a 5 year old 1GHz machine worth? I keep liking this type of decision to a used car - sure the mechanic could likely keep it running. but at some point, the owner (or mechanic if he's altruistic) has to say it's just not cost effective to play with this?
your thoughts?! Points will be given for everyone that shares their thoughts.
is the fact that the desktops are 5 years old (and the related bloat in the registry, etc) enough of an answer? that machine is doing the backups to tape using backup exec dektop (again, this is a desktop running windows 2K pro). the other desktops are not as slow, but don't see as much use. that machine gets on the web via dial up, there's no other internet access / outside contact with the real world. All the desktops are wireless to the router, which is wired to the server. the server is a single nic machine
I've checked things like dns server is the server for each desktop, etc. there's no spyware / viruses on his machine... it's just really slow. while there are likely loads of things you could offer to have me check, when you factor in the cost to have a tech (me) do that, again, is it wrong to say 'it's old, the OS gets bloated, etc... and push for a new machine? What is a 5 year old 1GHz machine worth? I keep liking this type of decision to a used car - sure the mechanic could likely keep it running. but at some point, the owner (or mechanic if he's altruistic) has to say it's just not cost effective to play with this?
your thoughts?! Points will be given for everyone that shares their thoughts.
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Lee W, MVP
I disagree with the age comment. It really depends on how reliable that model has been for you and what you do with it. I'm replacing machines that are 7 years old at my client right now... because 7 year old systems are Celeron 400's. They've been reasonably good - and the people that use them are NOT power users. I still got a 500 MHz box to replace next year and late next year we're planning on getting rid of the 700 MHz systems. If you're doing CAD and major number crunching or video editing, etc, then yes, those are old... but for basic Web browsing, email, and basic office apps (letters, small excel files, etc), I'd give those 1GHz systems another 2+ years at life... WITH a quadrupling of the RAM.
Each to their own, In my experience, with IBM machines mostly, they have a lifespan of around 4 years. After that they bog down to a point where they are ridiculously slow to work with and become a cost to business unless they are rebuilt. All depends on what they are using them for I guess, most of the machines I am referencing were used for POS applications and the like
7 years on a Celeron is no small feat!
Works fine - they even have photoshop CS installed on one and are using it like a terminal server - one user at a time of course (XP system) to use photoshop - usually just basic lightening and/or darkening of images, nothing too fancy, but it's been working just fine for them. Now go back 10 years and yes, 5 years was ANCIENT for a machine... think about it... in 1993 I had a 486SX 25 Mhz laptop... in 1998 you had, if I remember correctly, 300 MHz systems and if you dared to try to use a 486 - even a "fast" one with a PCI bus, it was still dog slow - for virtually ANYTHING modern.
On the other hand, I'm using a 4+ year old P4 Laptop to type this on and it's fine... 768 MB RAM, 1.9GHz CPU... no real complaints... I don't see myself getting a new one (unless this fails) for at least another couple of year... possible 3 or 4 more...
On the other hand, I'm using a 4+ year old P4 Laptop to type this on and it's fine... 768 MB RAM, 1.9GHz CPU... no real complaints... I don't see myself getting a new one (unless this fails) for at least another couple of year... possible 3 or 4 more...
Thats awesome! Guess I have been around machines that get flogged to hard by questionable individuals.... in 1993 it was year 3 at school for me and I didnt even know what a computer was :-) except for going to mates house after school and playing the original flight simulator with a one button joystick... in 98' i would have just hit year 8 at school **grin**...think i got lucky and landed in IT at the point where its all modern machines that are so cheap nowdays its easy and feasible to replace which probably gives a bit of a "trigger happy" approach to it all
ASKER
leew - thanks for the thoughts, but are you salaried where you are doing this or paid hourly / consulting? Again, yeah, older machines will keep running, but is it cost effective?! When paying hourly and looking to quadruple the ram....... older ram I think is more expensive / MB than newer memory? add the labor and ram in there and it doesn't pay to keep milking along a system? Yeah, if its running, leave it be. but if it gets gummed up, does it pay to reinstall the OS and all the patches and apps?!
Hourly consultant (It's a bit of an odd arrangement, at least in my opinion - I'm present to field questions a set number of hours per week and because it's "guaranteed time" I charge them notably less than other clients.
PC133 RAM is not very expensive and it's quick to install. And it's a REASONABLY safe bet those machines are using PC133 - if they are using RDRAM, forget it - that's insanely expensive. At that point I'd say go buy new systems. But anyone who's worked on 2000 and XP systems regularly will tell you 128 MB is too little. AT LEAST double it... and preferably quadruple it. And unless these are cramped machines, installing RAM really shouldn't take more than 5-10 minutes per system. So labor really shouldn't be a consideration.
If they do use PC133 SDRAM, make sure you use a tool like Crucial or Kingston's web site to ensure you get the correct RAM, but I would expect an upgrade to 512 MB for each system to cost $75. 5 systems, $375 + 1 hour labor. Again, depending on what they do, that can be a REAL good price to extend the life - and SERIOUSLY enhance the performance - of 5 computers by another 18 (minimum) - 36 months. Again, if you have experience working on 2000 systems, you KNOW how RAM can play a HUGE role in system performance. At worst, experiment - charge $100, RAM included, and upgrade ONE system. See what happens. That's a $25 install fee plus 384 additional MB of RAM.
As for what you should do about "gummed up" systems, that depends on the system and what it's used for. As I said, if these are used for basic purposes, then I would expect to keep them running another couple of years. Will OS, patch, and app loads take longer... sure... and you as the consultant (or IT support person) needs to determine the costs involved to the client - I know how much I charge and what I would and I have confidence in the my skills so that I will go so far as to provide a reasonable guarantee to most clients with regards to performing a fresh install.
PC133 RAM is not very expensive and it's quick to install. And it's a REASONABLY safe bet those machines are using PC133 - if they are using RDRAM, forget it - that's insanely expensive. At that point I'd say go buy new systems. But anyone who's worked on 2000 and XP systems regularly will tell you 128 MB is too little. AT LEAST double it... and preferably quadruple it. And unless these are cramped machines, installing RAM really shouldn't take more than 5-10 minutes per system. So labor really shouldn't be a consideration.
If they do use PC133 SDRAM, make sure you use a tool like Crucial or Kingston's web site to ensure you get the correct RAM, but I would expect an upgrade to 512 MB for each system to cost $75. 5 systems, $375 + 1 hour labor. Again, depending on what they do, that can be a REAL good price to extend the life - and SERIOUSLY enhance the performance - of 5 computers by another 18 (minimum) - 36 months. Again, if you have experience working on 2000 systems, you KNOW how RAM can play a HUGE role in system performance. At worst, experiment - charge $100, RAM included, and upgrade ONE system. See what happens. That's a $25 install fee plus 384 additional MB of RAM.
As for what you should do about "gummed up" systems, that depends on the system and what it's used for. As I said, if these are used for basic purposes, then I would expect to keep them running another couple of years. Will OS, patch, and app loads take longer... sure... and you as the consultant (or IT support person) needs to determine the costs involved to the client - I know how much I charge and what I would and I have confidence in the my skills so that I will go so far as to provide a reasonable guarantee to most clients with regards to performing a fresh install.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
Sorry, I diidn't intend to ignore it... it's that I am not on site, and it's thanksgiving.
and certainly not to argue Jeff, I difer to your experience for sure... but you are questioning my the idea of no right or wrong answer is debatable?
You say that running XP on a new machine is a decision nobody should disagree with and point out Leew asking for the ipconfig. But he's also pointing out that PC133 ram isn't all that expensive, etc... (keep the older machine working). I read that and looked - for the dell diemsnion L9333r that they have, it has 1 of it's 2 slots with 128 MB ram in it currently. the box has a max of 512 MB. so do you pop for the $80 in ram, for 2 of the 256 MB sticks and fart around with getting that machine to work or put it towards a new machine. I agree with you jeff - it's a no brainer to be on xp on a newer box. so why spend the time on the current network to make it run faster when a new box could be in place within days? You want to see if the dns server is the SBS box< i would think at first glance. I know it is and on monday I'll post the settings... being that only 1 machine is having the seriousl slowdown, I'd want to post its ipconfig /all and from other faster machines.....
i think my point is that the machine is old and while, if it was running OK, it would be fine for them to use... being that it's not OK, simply replacing it is / isn't a logical way to resolve the problem?
and certainly not to argue Jeff, I difer to your experience for sure... but you are questioning my the idea of no right or wrong answer is debatable?
You say that running XP on a new machine is a decision nobody should disagree with and point out Leew asking for the ipconfig. But he's also pointing out that PC133 ram isn't all that expensive, etc... (keep the older machine working). I read that and looked - for the dell diemsnion L9333r that they have, it has 1 of it's 2 slots with 128 MB ram in it currently. the box has a max of 512 MB. so do you pop for the $80 in ram, for 2 of the 256 MB sticks and fart around with getting that machine to work or put it towards a new machine. I agree with you jeff - it's a no brainer to be on xp on a newer box. so why spend the time on the current network to make it run faster when a new box could be in place within days? You want to see if the dns server is the SBS box< i would think at first glance. I know it is and on monday I'll post the settings... being that only 1 machine is having the seriousl slowdown, I'd want to post its ipconfig /all and from other faster machines.....
i think my point is that the machine is old and while, if it was running OK, it would be fine for them to use... being that it's not OK, simply replacing it is / isn't a logical way to resolve the problem?
I'll say again, it's Dependant upon WHAT the users do. For e-mail, web surfing, basic Word and Excel, 1 GHz and Windows 2000 is fine. The productivity benefits you might see because you use XP are negligible unless you intend to manage it like an IT enterprise. And if you've ever installed RAM in a computer, you know there's no real "farting" around to get it to work - RAM is a very easy, very cheap upgrade, and in my opinion, for any machine running 2000 or XP, it's a very NECESSARY upgrade if you have less than the previously mentioned minimum.
Don't forget what a new machine running XP will cost you - $650 - 8x the RAM upgrade - before new machine install costs. I just bought 3 to replace the celeron 400's I mentioned... They were MINIMAL machines with XP Pro, only upgrade was 1 GB of RAM total - 2.66 Celerons with 80 GB hard drives, DVD READER/CD Writer, NO monitor. Remember the XP Pro license alone tacks on $100.
So you can tell your client he needs to spend $3000-$3500 PRE labor on 5 new systems, or you can tell him he needs to spend $400 PRE labor to upgrade the RAM. Or, as I said, you can be nice if you don't believe me and upgrade ONE system and see if it helps (upgrade the most used system).
Don't forget what a new machine running XP will cost you - $650 - 8x the RAM upgrade - before new machine install costs. I just bought 3 to replace the celeron 400's I mentioned... They were MINIMAL machines with XP Pro, only upgrade was 1 GB of RAM total - 2.66 Celerons with 80 GB hard drives, DVD READER/CD Writer, NO monitor. Remember the XP Pro license alone tacks on $100.
So you can tell your client he needs to spend $3000-$3500 PRE labor on 5 new systems, or you can tell him he needs to spend $400 PRE labor to upgrade the RAM. Or, as I said, you can be nice if you don't believe me and upgrade ONE system and see if it helps (upgrade the most used system).
To me, THAT'S the no brainer.
If your client is, however, more than happy to spend the money..... then go for it I say :)
ASKER
THanks guys. I WILL post the ipconfig settings... I just have a mental block of spending money on a maching that I consider past their prime. - sure it's 'only' $100 per machine and you may get a well running machine, but the os does get bloated over those 5 years time (so ram may not be all that it needs) and how long will the machine keep running.
not all the sytstems need the ram.. it's just one that is much slower than the rest.
thanks guys!
not all the sytstems need the ram.. it's just one that is much slower than the rest.
thanks guys!
leew... just to put your comment into perspective...
Let's say that it's $750.00 per machine. This amount would be amortized over five years (if not expensed) which makes the cost around $150.00 per year. Buying the RAM, therefore, is only about $50.00 less as this is only a short-term fix.
Based on a very rough ROI formula, a user making $20,000 per year would only need to save about 5 minutes a week in increased productivity to pay for the upgrade. (I can provide the forumla details if you like).
So, if the user can't increase their productivity by one minute a day using XP, then I'd be rather surprised.
Everything in business comes down to the bottom line... but too often, there is not enough consideration about what may be negatively affecting that number. Employee productivity is probably the single most contributing factor to any company's bottom line, and in my opinion, not giving them the proper tools to do the job is penny wise and pound foolish.
TechSupportWhiz...
"why spend the time on the current network to make it run faster when a new box could be in place within days?"
Because I have a feeling that things may really not be configured properly here... and if you aren't able to recognize what the issue is, then you may end up adding new machines with the same configuration problem. It doesn't hurt to check that out. If I'm right, it'll take you about 5 or 10 minutes to fix the problem as things stand currently... and will save you a ton of time if you proceed with the upgrade.
Jeff
TechSoEasy
Let's say that it's $750.00 per machine. This amount would be amortized over five years (if not expensed) which makes the cost around $150.00 per year. Buying the RAM, therefore, is only about $50.00 less as this is only a short-term fix.
Based on a very rough ROI formula, a user making $20,000 per year would only need to save about 5 minutes a week in increased productivity to pay for the upgrade. (I can provide the forumla details if you like).
So, if the user can't increase their productivity by one minute a day using XP, then I'd be rather surprised.
Everything in business comes down to the bottom line... but too often, there is not enough consideration about what may be negatively affecting that number. Employee productivity is probably the single most contributing factor to any company's bottom line, and in my opinion, not giving them the proper tools to do the job is penny wise and pound foolish.
TechSupportWhiz...
"why spend the time on the current network to make it run faster when a new box could be in place within days?"
Because I have a feeling that things may really not be configured properly here... and if you aren't able to recognize what the issue is, then you may end up adding new machines with the same configuration problem. It doesn't hurt to check that out. If I'm right, it'll take you about 5 or 10 minutes to fix the problem as things stand currently... and will save you a ton of time if you proceed with the upgrade.
Jeff
TechSoEasy
Little more perspective...
My client purchased retail copies of Office - this allows them to transfer the licenses from machine to machine, unlike OEM licenses. BUT, if they bought OEM licenses, as many companies do, then your new system costs rises to about $1000 per machine.
ROI is a great thing... but it's also WAY over rated, especially in some small businesses. You can quote ROI all over the place, but most small businesses have little way to quantify it - especially with a minor upgrade in OS. XP is NT 5.1. 2000 is NT 5.0. It's NOT a major upgrade. Yes, there are some things that are useful and can only be done with XP (Remote Desktop, for one), but on the whole, XP or 2000, users are not going to notice a difference going from one to the other, in my experience and opinion. There may be good reasons from a safety, support, and reliability standpoint, but again, if you do the math, I don't think the "loss" would be significant to the "average" small business.
Now, I'm not saying ROI should be ignored. But the secretary in a small office who does little more than help keep someone's schedule and does some filing and e-mail is can do their job equally well on a Pentium III 500 or a Pentium Dual Core running at 3 GHz. This is what I've been saying all along - it depends on how they use the computers. One of the biggest causes of lost productivity with computers is inadequate RAM. With enough RAM programs can stay in RAM and load/be used without swapping out to disk. With too little RAM you start swapping RAM to disk and end up waiting. For e-mail and BASIC office stuff (as I've been saying all along) a "low end" 500-1000 MHz system - WITH enough RAM - will be more than sufficient without any significant loss of productivity.
@TechSupportWhiz:
> I just have a mental block of spending money on a maching that I consider past their
> prime. - sure it's 'only' $100 per machine and you may get a well running machine,
> but the os does get bloated over those 5 years time (so ram may not be all that it
> needs) and how long will the machine keep running.
YOU have to decide - with input from the client - if these machines are (motherboard and CPU) past their prime for their task. If the user ONLY uses a web site for data entry, then a low powered PDA may be sufficient. But if they do Computer Aided Design, then yes, the machines ARE past their prime. You must ANALYZE the usage and not rely on what YOU and YOUR usage would consider "past their prime"
Let me try to rephrase - your client's usage dictates if the machines are past their prime. A stock broker who can make $10,000,000 in a minute needs to have a reliable, fast system and network. Possibly several because being down even one minute can cost FAR more than buying 5 computers and having them all setup to use at a moment's notice. But a small retail store that sees 100 customers a day during the busy season won't likely find upgrading every year or two or having a seemingly excessive number of systems to be worth it.
If the client does not want or would prefer not to spend $3000-$5000 on new computers, plus, I figure, AT LEAST 1 hour installation and data transfer labor PER MACHINE, which, depending on how much you charge (and assuming it IS only an hour per machine), can tack on another $500-750 (assuming you charge relatively common industry rates), then the costs are $3500-$5750 or more. Based on the clients I've had, this is not a non-trivial amount and you can expect at best a pause before they agree and worst a flat out NO. Unless they are already resigned to doing it. One of the best and cheapest options - and you should always provide options - you can recommend one thing over another, but give them options - is to increase RAM in these systems. Assuming they agree, you're looking at $400+1 hour labor (Dell systems are generally easy to upgrade RAM in). So your client has a choice... upgrade the RAM to 512 in all systems at a cost of $475-$550 and see a performance increase of probably 33-50% for "average" use, or spend 7-11x as much on new systems and the LOST productivity of having to move users out of the way as you do the upgrades (unless you can do them after hours - but that still tends to have problems as "little programs" users like to use are often no reinstalled.
And yes, over time, things can develop problems. ABSOLUTELY. So, look over the event logs. Are the systems healthy or not? Do system state backups if they are. Then you can be reasonably sure the systems will remain reliable and stable.
My client purchased retail copies of Office - this allows them to transfer the licenses from machine to machine, unlike OEM licenses. BUT, if they bought OEM licenses, as many companies do, then your new system costs rises to about $1000 per machine.
ROI is a great thing... but it's also WAY over rated, especially in some small businesses. You can quote ROI all over the place, but most small businesses have little way to quantify it - especially with a minor upgrade in OS. XP is NT 5.1. 2000 is NT 5.0. It's NOT a major upgrade. Yes, there are some things that are useful and can only be done with XP (Remote Desktop, for one), but on the whole, XP or 2000, users are not going to notice a difference going from one to the other, in my experience and opinion. There may be good reasons from a safety, support, and reliability standpoint, but again, if you do the math, I don't think the "loss" would be significant to the "average" small business.
Now, I'm not saying ROI should be ignored. But the secretary in a small office who does little more than help keep someone's schedule and does some filing and e-mail is can do their job equally well on a Pentium III 500 or a Pentium Dual Core running at 3 GHz. This is what I've been saying all along - it depends on how they use the computers. One of the biggest causes of lost productivity with computers is inadequate RAM. With enough RAM programs can stay in RAM and load/be used without swapping out to disk. With too little RAM you start swapping RAM to disk and end up waiting. For e-mail and BASIC office stuff (as I've been saying all along) a "low end" 500-1000 MHz system - WITH enough RAM - will be more than sufficient without any significant loss of productivity.
@TechSupportWhiz:
> I just have a mental block of spending money on a maching that I consider past their
> prime. - sure it's 'only' $100 per machine and you may get a well running machine,
> but the os does get bloated over those 5 years time (so ram may not be all that it
> needs) and how long will the machine keep running.
YOU have to decide - with input from the client - if these machines are (motherboard and CPU) past their prime for their task. If the user ONLY uses a web site for data entry, then a low powered PDA may be sufficient. But if they do Computer Aided Design, then yes, the machines ARE past their prime. You must ANALYZE the usage and not rely on what YOU and YOUR usage would consider "past their prime"
Let me try to rephrase - your client's usage dictates if the machines are past their prime. A stock broker who can make $10,000,000 in a minute needs to have a reliable, fast system and network. Possibly several because being down even one minute can cost FAR more than buying 5 computers and having them all setup to use at a moment's notice. But a small retail store that sees 100 customers a day during the busy season won't likely find upgrading every year or two or having a seemingly excessive number of systems to be worth it.
If the client does not want or would prefer not to spend $3000-$5000 on new computers, plus, I figure, AT LEAST 1 hour installation and data transfer labor PER MACHINE, which, depending on how much you charge (and assuming it IS only an hour per machine), can tack on another $500-750 (assuming you charge relatively common industry rates), then the costs are $3500-$5750 or more. Based on the clients I've had, this is not a non-trivial amount and you can expect at best a pause before they agree and worst a flat out NO. Unless they are already resigned to doing it. One of the best and cheapest options - and you should always provide options - you can recommend one thing over another, but give them options - is to increase RAM in these systems. Assuming they agree, you're looking at $400+1 hour labor (Dell systems are generally easy to upgrade RAM in). So your client has a choice... upgrade the RAM to 512 in all systems at a cost of $475-$550 and see a performance increase of probably 33-50% for "average" use, or spend 7-11x as much on new systems and the LOST productivity of having to move users out of the way as you do the upgrades (unless you can do them after hours - but that still tends to have problems as "little programs" users like to use are often no reinstalled.
And yes, over time, things can develop problems. ABSOLUTELY. So, look over the event logs. Are the systems healthy or not? Do system state backups if they are. Then you can be reasonably sure the systems will remain reliable and stable.
Hey I read this whole paq and still no ipconfig /ALL posted!!! -gsgi
Which is why I've not responded yet to this... since I think that would be central to the entire discussion.
Jeff
TechSoEasy
Jeff
TechSoEasy
ASKER
they are buying new PCs.... end of year tax issue.
problem solved!?
problem solved!?
ASKER
I personally like JayJay's thinking - replace it/ sure, you can add ram, reinstall the OS, apps, drivers, etc... but I think you are putting in too much time to do that and you still have an old machine that's running OK. yes, an older machine's speeds are OK for simple work that they are doing.. but statistically, that 5 year old machine is more likely to fail after you spent all that time working on it vs. a new machine.
I think it's not an issue of the machines being past their prime for the apps... they certainly are not... they are past their prime for investing time / money to get them running.
an old clunker car.... gets you to work.. it does what you need it to do... so why replace it?!. but statistically, it's older / more likely to fail than a new machine. would you spend hundreds on tires, new exhaust, replace hoses, etc. etc... or put it into a new car? It's a much wider $$ spread between new car and old car. but the analogy is there?
thanks guys
I think it's not an issue of the machines being past their prime for the apps... they certainly are not... they are past their prime for investing time / money to get them running.
an old clunker car.... gets you to work.. it does what you need it to do... so why replace it?!. but statistically, it's older / more likely to fail than a new machine. would you spend hundreds on tires, new exhaust, replace hoses, etc. etc... or put it into a new car? It's a much wider $$ spread between new car and old car. but the analogy is there?
thanks guys
I still would like to see the ipconfig / ALL so I can learn from leew and the group how to tell from ipconfig / ALL if dns is setup properly or if not then how to fix it. I like to learn here, that's why I help out here. It's a simple concept I think. -gsgi
Glad they made life easy for you with new machines...I personally agree with your analogy as well.....different views and opinions though is what makes these boards what they are :-) Cheers mate
gsgi - you want DNS rules - For a simple domain structure:
Client points to local internal DNS server only.
Server points to itself (If DNS server) OR a local DNS server the same as a client.
DNS Server should be set with ISP forwarder if given, though you dont HAVE to have a DNS forwarder, you can resolve off of root hints just fine.
If you have a mult tiered forest OR you are looking at a multi site domain, then your primary and secondary DNS servers come into play but thats a separate question completely unrelated to this topic :-)
There isnt much more to be learnt from an IPCONFIG in this scenario, sounds to me like the machines were just old
J
gsgi - you want DNS rules - For a simple domain structure:
Client points to local internal DNS server only.
Server points to itself (If DNS server) OR a local DNS server the same as a client.
DNS Server should be set with ISP forwarder if given, though you dont HAVE to have a DNS forwarder, you can resolve off of root hints just fine.
If you have a mult tiered forest OR you are looking at a multi site domain, then your primary and secondary DNS servers come into play but thats a separate question completely unrelated to this topic :-)
There isnt much more to be learnt from an IPCONFIG in this scenario, sounds to me like the machines were just old
J
thanks, gsgi
gsgi,
If you want to see a few examples of IPCONFIG's and their problems, put this search string in Google:
site:experts-exchange.com intitle:sbs ipconfig /all
Jeff
TechSoEasy
If you want to see a few examples of IPCONFIG's and their problems, put this search string in Google:
site:experts-exchange.com intitle:sbs ipconfig /all
Jeff
TechSoEasy