aspenjim

RoflMao...

Type sfc /cachesize=0 first and then reboot and type sfc/purgecache in the run box

I got to thinking after I made the last post... put the script in the startup directory... Or in the registry in the run key... I also asked someone who works at a colorado university IT Dept and he told me profiles were for exchange clients. Are they used for anything else and where are they stored?

@nmX .Memnoch Thanks for all of the information. After our meeting Tuesday, I decided I wasn't at all qualified to do this. My server would have been better than my predecessors, but it would have been lame compared to a true server admin's. The company putting in the barcoding gear is going to do the server. They are going to put in a proliant w/ 2 - 3.6 xeons, and 2 - 15k scsi drives (as opposed to 2 sata drives in raid 1 originally proposed). He is going to show me what I've been asking about in this post and I'll have a much better clue for the future. Moving on to today... I sold another PC to the nursing home that I originally asked about in the post (logon to the domain vs. logon to the PC. This time I had much quicker success than the previous 5 PC's I added. Still ran into a couple issues I like an quick answer to. Here's what I did differently. I installed all of the software first, before I joined the PC to the domain logged on as the administrator. This PC was for the accounting lady and there was all several new programs I had to install (business works, timeclock s/w and intuits online payroll services). It went nearly perfect. Here are the 2 issues... When I sat up the user account, when I gave her only user rights, Almost every program errored (I could tell it was a permissions problem). Whenever I gave her admin rights, it was okay, user rights, it errored. Logged on as a admin, I right clicked on the directories (which were all on the root of the HDD) and gave her full read/write privileges with no luck. I ended up just making her a member of administrators to get out of there tonight. The other issue was I could not get her login script to work. (it just mapped drives to shared, her user dir and apps). I studied 2 other PC's that someone else had setup and did the login script and I couldn't figure out what I was doing wrong. (Later, I thought I may look tommorrow to see if the batch file was in the startup dir). I ended up just mapping the drives and she was able to use the programs without any errors. Back again to the original question... logon to the domain vs. logon to the PC... The first 2 PC's I added to this domain, I joined them to the domain very first and (not knowing any better) had them logon to the domain. However, after doing that, I studied all of the login bat files and was able to make their PC's do the login script thing you mentioned earlier in this post. I just decided to make all of the users logon to the PC from there on. When I get all of this figured out, I will make all of the logons the same. Then, I have to figure out a password scheme, as I mentioned before. After this Sheriff's install, I'll prolly be much more clear on this. BTW, this server at the nursing home is a w2k server. I ran into one other thing new to me and need to do a quick crash course on... That is profiles. I'm embarassed, but I truly don't know exactly what a profile is, except something to do with user pref's (I'm guessing). I have never paid attention to the profiles tab on setting up a user on a pc. I did look at the working ones that someone else had set up and they had the logon script file name in there, but when I added "tbone.bat" (tbone is the user), it still didn't map the drives. There is a directory of logon scripts for the users on the server with tbone.bat in it.

Just out of curiosity, if I were to search for a file extension for a SQL db, what would it be? *.mdb is for access and I know there are 2 *.mdb files there (on the server). They talk about cutting "ADE's" on everyones workstation every several months and one of the files may just be a backup file. Also on everyone's workstation, there is a directory with 2 *.mdb files in it around the 75 megs I mentioned earlier. That is where the shortcut from the desktop goes to. From there, I don't how it gets to the server. I want to do clean installs on all of the XP homes as opposed to upgrades, although just to regain some of the performance as most were upgraded from Win 98. I didn't know about the PDC and BDC change. Did that happen when 2000 came out? AD doesn't seem too difficult yet for me on local networks. (Problem I have is mentioned in the first post giving the user domain admin rights. I will have to correct that soon as they want pw's to be changed every several months to comply with HIPPA laws. Some people have passwords to the pc and others have pws for server access and I don't quite know an easy way to get everyone on the same page. I figured I would have a much better idea after this Sherriff dept install.) I just add a user and computer here and there and give them appropriate rights and pw's. I know that's a completely different scenario with bigger networks and multiple domains (as I've seen at some colleges).

We are having a big pow-wow with all of the powers to be on Tuesday. I'm just not used to thinking on such a large scale, but I guess this is the government w/ a $15M budget and they do have $24k for this particular project. They spend as a whole about $15k a month for the software support for the inmate database and the ACS software to run the county. One more side question. the database is an access db about 75meg in size. What does SQL have to do with this. All I know about SQL is it is searchable (as opposed to MSDE). I am a MS registered partner and I subscribe to their actionpak and have all of the software that comes with that. I've installed SQL2005 on my server and don't even have a clue about how to configure it. In other instances, I've installed a trial version of Webroot Enterprise trial and couldn't figure out how it works. So in a nutshell, What exactly does SQL do when the db is in Access? (I don't have to worry about configuring this on the real server, because the company that does the support will dial in and do the configuration. However I should have a clue going into this. Incidently, 2 of the engineers of this software will be at the meeting on Tuesday. Again, I am an indepent contractor for this account and get paid $1400/ month to support about 80 county owned or leased PC's. We will probably renew the contract soon and I should get $1600/mo. I want to keep my time fairly minimal and when I talk about downtime, I mean just getting it installed and getting the kinks worked out and all of the XP Home's upgraded to XP Pro. One other thing I haven't thought about until your last 2 posts is the idea of keeping the old server as a BDC. How does that work for the database. I know how PDC's and BDC's work, but is the db kept mirrored on both servers in case the BDC has to be promoted. These are questions that they will have. An offsite backup is something that they really want. I know this would be just a walk in the park for most all of you, but it is keeping me up at night trying to think of all that needs to be considered. This is the first time in 7 years I've felt the need for some formal training. I've worked with the John Deere dealerships and have had to do a BDC promo (with JDIS's help), but they had a UNIX server that ran the business system and their NT servers just acted as a parts catalog server and connected to the unix server for pricing and inventory, so the temporary downtime (HDD failure) was minimal. This was in 2001. This my chance to raise my competency to the next level. I've complained about the Sheriff's server since day one and they just keep buying XP home from their previous HW vendor, just p***ing $$ away. I can appreciate that kind of loyalty, but the guy doesn't have a clue about the real world networks (even in a town of 2000 people).

You have to remember that I have never designed a domain before and need to keep it simple to minimize any downtime and keep their confidence in me. I have had this account for 18 months now and they have alot of faith in me (and I want to keep it that way). My head is swimming now with all of this info and I've also been looking at other resources. I also want to use their existing 2003 server OS (or maybe the R2 upgrade they should be entitled to) to save them money and ensure that I get the hardware business. The barcoding vendor has proposed about $20k for a new server also. By me keeping their existing OS, it should be a shoe-in for the hardware. BTW, I am probably going to load XP on the existing server P3-933 and use it for a backup PC as opposed to getting a new tape drive, since it already had scsi drives in RAID 1.

To whomever asked is is 2003 standard. They purchased a volume license of 2003 in may 2004 and got 20 CAL's and also purchased a volume license for SQL 2000 and 20 CAL's for that. (they also got software assurance which should entitle them to get 2003 - R2 and SQL 2005). However, I don't think the CAL's were installed and this week their Sonic Wall just exceeded it's 25 user license. That tells me that 5 more PC's than CAL's are all connected to the server or on the network. Also as I mentioned before, the licensing mgr is disabled on this server and another in the courthouse. That particular server was just installed a month ago as the former lease was up on it and about 10 clients. (I haven't mentioned server #3 before). It is setup as a workgroup peer and used as a file server and also has SQL on it. So having said that, is it possible the CAL's aren't installed? This licensing business is SO **** complicated. When you all refer to devices, does that mean the AP's and networked printers also? It's pretty intense trying to figure out what all they really need and what all they already have. (right now they have a mess) All of the doc's are local and not on the server and they have at least 10 XP home's. I've got 2 or 3 weeks to get this down in my mind and to make a solid proposal. I never mentioned also that they are installing a T1 to connect 5 entities together, with an option to add a gigabit fiber optics line for future growth. One at the Sheriff Dept, one at the courthouse, one at the nursing home and a 4th a human services and a 5th at a "events center" (a couple of meeting rooms and the fairgrounds). They are also purchasing 15 Cisco 1200 AP's. The main point of all of this is to eventually get them all on the same financial system. It is ACS Leasing, apparently a nation wide company providing govt accounting software for treasurer, assessor and finance. The entire network has about 75 PC's networked and a couple smaller networks (ambulance and recording) with about 10 more PC's and another 2000 domain server (I don't know why it is a tiny domain as it is just used to scan and record legal documents). I guess before getting all of this advice, I should have drawn the whole picture clearly. Right now the Sheriff Dept is what I'm working on. I want to do this right and professionally, as this county is barely crossing the "digital divide". They are having a website designed right now.... http://co.washington.co.us/ As for me, I have been a PC tech since 1999 and have been in business for myself for 3 full years now. I went through a semester of a CNA course and was hired by a company in Aspen, Colorado to help get them through Y2K and picked the rest up on my own. That particular company had about 30 - 40 networks, but almost all Novell Servers. I also worked for a year at a John Deere dealer, and got a good feel for JDIS domains, as this dealer had 5 branches. I've gotta admit this website is an absolute goldmine of information and thank all who offer advice for this project.

So that means "logon to the domain", right? From what I see, (and maybe it's a one time thing), it takes a long time for the first logon. I initially sat clients up that way, but shied away when I found out I couldn't do windows updates without first logging off the domain. Also, in another government account I picked up, they have an access database and I couldn't install the client software being logged on to the domain, because of the MDAC components it had to install. It seems like the drive mappings were way harder to get right. They liked U for users, s for shared, and H for the database. I would install the software as the Administrator, and then set up a user account and logon as the new user and the U, S and H drive letters weren't available. Then I would get all ugly errors, because they couldn't get into the database, etc. So from a previous post I made here at MSFN, someone suggested giving them domain admin rights. I did this and the installs started getting a little easier. The first couple took me about 3 hours each to do a seemingly simple task. Now doing the "domain admin" way, and then installing the client software was much easier. I realize this is a security hazard, but in this case, it is a nursing home and the users are quite naive. (It is pretty embarassing to me to take 3 hours to do a 15 minute job, but I'm a rookie at this domain business). I also fully realize that XP home can't join a domain. I didn't sell the PC's, just took over the accounts. So when one is logged on the domain, then are all of the doc's, favorites, etc are all stored on the server by default? I have set everyones docs up on the server for backup purposes in a data partition. Back to the Sheriff Dept where the new server is going to be..... Right now they have a P3-933 w/ 512 meg of Ram supporting about 25 users. I plan to install a Dell 1800 with 2-3.0 Xeon Procs and 2 - 73 gig 15k SCSI drives in RAID 1 and 2 gigs of RAM. If I do manage to pull this off this way (logging on to the domain), will they notice a speed decrease? (in exchange for security). If so, I need to prewarn them that security is the trade-off for speed. They have $24000 they need to spend pretty quick, so I was going to suggest getting everyone upgraded to XP Pro from Home and 2000. They are also getting a T1 and a new SONIC Wall and Cisco Router and 3 Cisco 1200 APs along with a barcoding setup for inmate wristbands with this 24k. I'm only providing the Server and XP software. This is going to be a major stretch for me to pull this off and look like I fully know what I'm doing, working with these other vendors (the barcoding people and the ISP). Right now the entire setup is a joke. But in their defense, they built a new jail with 125 beds and originally had one with 5 and over the last 5 years went from about 5 PC's (including the P3 server) to around 35, some provided by the CBI, Canteen people etc. I will read the links you provided and see if I can clarify some things on my own. Also as far as the login scripts, the nursing home account I got has a 2000 server and it is set up with scripts. It is pretty impressive of how clean it works. I've tried to use it as my example and have studied it and as I have added clients, I copied the Scripts and made slight modifications to the drive mappings for individual users. that is however, where I was saying that it originally took me 3 hours to do a 15 minute job. thanks again for the advice. I will have more questions, I'm sure. aj BTW.. those are the same link (and the terminology is hard to grasp eg. "schema"). Also. would you kindly explain the role of CAL's? Is it just a legal requirement from microsoft? Both of the 2003 server networks I support have the "license manager" service disabled.

Hey everyone. I am getting ready to install a new server 2003 in a local sheriff's dept. Right now, the old server is 2003 with around 25 clients. It is setup as a workgroup server (no domain), which all of the clients (2000, xp home and pro) login to a database which is apparently ran by SQL server 2000 all using Administrator and the same password. I don't have to worry about reconfiguring SQL, because the vender that supplies the software will do that. My question is.... which is the prefered method (or advantages) of logging on to the computer or the domain. I've added other PC's to a similiar 2000 domain and some are logging on to the local PC and others are logging on to the domain. I've learned through another post I made in here while back to setup the users with domain admin rights and then I get away from having to log-off of the domain to do Windows updates, etc. Is one advantage that users can logon to domain from any PC on the network and have access to their doc's, etc. Right now also, this particular account has a volume license and 20 CALS for both SQL and 2003. However, the CAL's were never installed and the "License MGR" is disabled. That is probably because they have about 10 XP Home PC's. I've only setup one other domain (with success) and they have 4 XP Pro and 5 XP home Clients and I've never ran into the issue of having to install additional CAL's. What would be nice is if a few of the readers of this might post a link or 2 to guides to setting up a domain, with the pro's and con's of different scenarios. Unfortunately, I live in a small rural town, and don't have access to servers setup professionally to study their methods and obviously I'm not certified. TIA for any input aj

i did a quick search of MS.com and found this with a valid phone#... http://support.microsoft.com/kb/284230/ USA... 1-800-360-7561 Canada... 1-800-933-4750

MS will happily replace damaged (legit) media and ME doesn't require activation

WD all the way, cept Seagate now offers 5 year warranty

Very nice and clean.... Thx man!

If you have SBS2003 and haven't promo'd it to a DC, it'll shutdown. You'll see the error in event viewer.

Thats what hyper-threading does is makes a virtual 2nd processor so since you have 2 physical CPU's, you also have 2 virtual CPU's.

I'm embarrassed a little. I've been working on PC's professionally since 1999 and can get them connected to a domain, usually with no problem. Installing software is sometimes a problem. I've never set up a true domain w/ 10 users. Lots of peer networks though. Just thought this would be a good place to learn the best methods. @Ghostrider... that's what I need to do. Thanks.

What's the fundamental difference in the logons - logon to this computer and logon to the domain? Sorry for such a basic question. I want to get everybody the same so we can start a password reseting schedule. What would the preferred logon be and why? I also want to setup logon scripts so everyone gets to the right documents.

I'd leave it as a workgroup server and not a domain controller, but it shut down every few hours and complains about violating the EULA. My server always complains about the router and then I have trouble getting by the "Configure" DHCP scopes when the wizard comes up. I also want to learn the RIS stuff and in that case it has to be a domain controller and have AD installed. I'd let it be the DHCP server, but I have a customer in the offices next to me tapped into my internet and I need to keep the internet going 24/7 for them and occasionally my server is down.

I'm going to setup my first domain controller (SBS2003) for a customer next week and would like to know the pro's and con's and difference's between logging on to the computer and logging on to the domain. I know one diff is I can't do windows updates or otherwise alter any files in the windows directory while logged on to the domain. This is going into a small headstart school with about 10 clients (2 being mobile). The reason I ask is I also do the IT work for a nursing home and they have a domain. I've noticed (through doing upgrades and software updates) that I almost always have to be logged off the domain to make changes. More specifically why I ask is about 5 out of 10 clients logon to the computer and the other 5 logon to the domain. I'd like to get some consistency in new install, but need some advice in this next week. I also have trouble with configuring DNS and DHCP when the server has 2 nics and a linksys router connected to a wireless bridge supplying the internet. My long term goal with my personal server is ISA 2004, but without the firewall function (just the acceleration part). Could someone also tell me the preferred settings for the 2 nics when one will be connected to a router and the other to the local network. I use the 10.10.10.0 subnet for the LAN and leave the nic connected to the router dynamic. Thx, aj

Title says it all... Kind of a WinPE n00b question, I know, but I'm a WinPE n00b!!! thanks for any infos AJ edit... Okay I found out it means Windows Image Management and new feature of Vista that does a file based image as opposed to a sector based image that will install on different hardware. Any further explanation would be great. Thx.

Thanks for the info. I'm dying to learn what WinPE can do and also getting a RIS server working. I just don't have much spare time learning and trying this stuff out to work out the bugs and get it into production. I have, however, got a nice unattended install CD going for the various flavors of XP with several APP's working. I think my first step with winpe is using the winpe files with barts pebuilder and make one and use the sherpa xpe plugin. Jim

@odopey... What is BDD\OSD ? aj

I have a PC (w/legit Windows I might add) that won't install the updates properly. I did use the update pack 1.31 integrated, but several other PC's with the same CD installation works okay. It is rather annoying to have the yellow shield keep popping up telling me I have updates to install, but they fail to install. I'd like to go into the "software distribution" directory and delete everything and let it see if it can redownload the updates and install them properly. I also have a customer who complains of this problem. How can I properly delete the update and have it install properly? Anyone else with this problem? Thx Jim

I'd be willing to bet that it is overheating. Those CPU's run hot. D/L motherboard monitor or everest 200 and checkout your temps.