CoffeeFiend

That's mostly my point. We don't keep ancient hardware around. Normal ACPI HAL always worked for every PC I've ghosted (hundreds of them - haven't seen non-ACPI compliant hardware in many years; and most businesses don't have or need fancy super-fast CPUs on desktops). As for Mass Storage Adapters, no, sysprep won't always reliably make a normal "plain IDE" install work on a SATA RAID/SCSI setup or such (you'll get Stop 0x7b Inaccessible Boot Device; in theory it should work, but in practice it often won't)

I don't even see what you're worried about. 55 PCs only? Over 4 weeks (20 work days), that's not even 3 PCs/day. One person could manage that installing everything manually (non-unattended). Make basic unattended setups (optional) to install from. Then sysprep, ghost, and use that image for the other PCs (much like IcemanND said). I don't ever recall having a single HAL issue over the years (the most likely part would be mass storage adapters and such). Pack the necessary drivers somewhere (right in the ghost image perhaps so it ends up on the HD), let it detect new hardware and you're done. It only takes a few minutes to restore the ghost image (depending on size of image mainly) and detect drivers. Assuming there isn't anything to keep on the PC's HDs, one guy could come in on the odd saturday or sunday (when no one's there) and do them all the same day by himself... (roughly 10 minutes/PC for a full day - not bad; worst case scenario a slightly longer day)

True to some extent. In the case of classic ASP, pretty much everything else is better. Not everything new is better, but .Net 2.0/C# is (or at least is amongst) the absolute best dev platform right now (IMHO), and ASP.Net is one of its strong points. It's truly great. Comparing it to what's more or less the worst... You say it does what you want, but seeing your post, it doesn't seem so (having problems with newbie stuff on an extremely simple script and not making sense of it). Honestly, it looks like you more or less have to learn the whole thing anyways, why not learn the new and dramatically better way - or perhaps use something like PHP while you're not locked in, and that you seemingly only want a simple scripting language without steep learning curve and all (and it's free too and a whole lot cheaper to host on the web if you ever need to). There's tons of free learning material for all of them (I can post some if you're interested). Your choice anyways. Best of luck to you.

We've got way more computers than that, and of many makes too (we buy in large qtys, but over the years you accumulate lots of different ones). Good old sysprep + ghost works best for our setup. We keep a generic image updated, plus a few more for some "more problematic" PCs (which we're having to do more and more lately, due to SATA mass storage adapters and such). But those "other" images aren't used quite as often, are updated as needed (deploy old to one PC, update, sysprep, create new image from that), so it's not that time consuming either. RIS, PXE and all wasn't of much use for us (we do ocasionnaly use ghost multicast server though). Anything that isn't straight ghosting is installed from a network share (boot via good old custom/heavily-tweaked Bart floppy with tons of NDIS2 drivers and all, burnt onto a CD). One day we might switch to WinPE, but so far the floppy is still a better method for the way we do things. Plain unattended installs? That would be the ultimate last resort. Not. Gonna. Happen. Takes too much time for mass deployments, it's just not realistic - even for "daily" reinstalls by the guys at the bench when something's gone wrong. ~3 minutes for reimaging right at the user's desk while he grabs some java VS take his PC away from his desk for over an hour? (more like a full day if you count everything else that has to be installed after windows: AV, line of business apps, ms office, in-house apps, more patches, etc). Unattended installs from media (CD/DVD) is even more out of the question (if that's even possible). Walk around many, many buildings, with media in hand, insert in drives, wait on it for ages, find out the bios boot sequence wasn't setup right, make sure no one makes a copy of the CD for home use (can't let anyone just use our keys and all), etc. Sounds like a nightmare. If they ever decided to do something like that, I'd let them know that I'll be gone in 2 weeks. (Ok, never gonna happen, as my job doesn't involve doing this stuff ) [edit] looks like HwagPo beat me to it really

Uh? The thing is in digital format already. There's no point converting it to analog, then thru an amp (and those things DO clip BTW - I've designed audio amps for years...), then back to digital again. That will distort it and add some clipping (which is what you seem to be against). Amps don't make peaks disappear (more like they tend to introduce extra clipping of their own). If you got to boost so much that you get clipping using software, using an amp will hardly be better. Your >0dB peaks will still be too strong, will still clip, and will produce a voltage that will end up as "clipping" once captured again (and you're adding more distorsion regardless - even if you have some fancy class A tube amp with really good filtering and all). That would likely require him to spend some $$ too (for something half decent), and sounds quite overkill for most purposes (home videos namely). Unless you're talking about amplifying whatever your signal is before recording it i.e. recording it at an acceptable level in the first place... (obviously a well placed decent mic can't hurt either) Maximizing/normalizing volume if it's at a low level (using software) should not produce any clipping whatsoever (unless you overdo it). The app will detect the peaks (max values of -32768 to 32767 for 16bit) and will only boost so much. And there even more ways using software to boost even more without clipping (or minimizing it), using things like dynamic range compression (boost quiter parts more and louder parts less) and what not. There is lots of good apps that can do this, from high end commercial mastering apps and plugins to free/OSS ones.

There's about a million things it could be. From so-so quality network drivers (seen that too many times, kinda common), a [really] badly fragmented NTFS partition, un-optimized network stack, network utilisation somewhat high for some reason (something running in the background - spyware included), OS "too busy" with a process, you could be browsing the files using explorer (which loves to bog down the network to read metadata and things like resolutions of video files and the like - kinda painful on a video server), a firewall making the network slow overall, SMB speed issues (sometimes due to bad time sync as it's used by kerberos within SMB for authentication), etc. That list could be much longer... Change drivers, try a different NIC, check the event log, check your performance counters, check what's coming across the wire via various network capture/logging/debuging utilities for various protocols (MS has a network monitor app, kerberos tools and all)... You'll have to find the problem, and there's tons of places to look, tons of things that could've gone wrong, and it'll take time and knowledge to find and fix it. Good luck

.

That's a versionning system alright. But I doubt that's what you really want. CVS is primarily intended for text files (source code) - not binary files. It could be somewhat problematic in a few ways. And CVS is a old-ish system (why not use SVN instead at least?). The other issue is although TortoiseCVS is a lot more user friendly than having to type CVS commands, I'm not sure it's going to be really good or user-friendly enough for the end-user... They just might not understand or use it. And TortoiseCVS is a CVS client - you'll still need the backend. You'll need to learn CVS yourself pretty well if you pick that (or even to evaluate it) and its shortcomings (can't rename files, etc). I still fail to see the relevance of this. Is there even a server i.e. "back-end" part to this app? Sounds like a plain desktop app to me (that has NOTHING to do with your server scenario). But then again, I've never seen the app itself... Seemingly option #1 is too expensive (which was my point)... Understood. Scenario #2 may be the other extreme though. That machine should be able to do a whole lot more than just serve some files to a handful of people... (i.e. if their needs are "modest"). As for adding another box running on an open source platform, it's an option, but don't necessarily think it's always the cheapest/first option to pick. Look at the alternatives and *total* cost - and needs/resources... RHEL is about the same Win2003 Web Server ed (runs IIS6 and could run SQL Server 2005 Express Ed too - or some other DB of your choice and whatever), i.e. 350$ for RHEL ES w/ basic support vs 400$ for Win2003 Web Svr Ed (50$ is hardly a deciding factor for 2 systems with such fundamental differences really). Mind you there are more options here too (FreeBSD, Solaris, CentOS, SUSE, ...) What it seems to come down to is what do you want from the server? Run ASP.Net apps on IIS6 (C# or VB.Net, .NET Framework, dev using Visual Studio, and everything centric/related to that setup), or do you want to serve simple PHP pages and likely use some of the open source PHP-based apps (like perhaps a CMS?) on a fairly typical "LAMP" stack? What are you (and the people that'll use the server - i.e. write apps/pages for it and maintain them) more comfortable with? (if you have no *nix knowledge, administering a freebsd web server box or such without even X running could be a "whole lotta fun"). Lots of people just aren't willing to go thru all this really (or they're a "MS-only shop" and intend it to stay that way). That's totally up to you. No one can force you to learn nor will do it for you or whatever. I'd say both are good, viable options, but they address vastly different needs. Some even opt for a "mix" of the previous 2 solutions i.e. WAMP or apache w/ mod_mono on linux. Both are so different that price here is almost an afterthought really. Some places will pick IIS6 on Win2003, another will pick LAMP (be it PHP, Python or Perl) on FreeBSD, while another may pick JBoss on RHEL or Websphere on SUSE, ... Mail wise, if they don't need outlook calendaring and the like, there are tons of other options (hMailServer being one - and free).

.

Standard ed doesn't have SQL Server (nor ISA for that matter)! That's the main reason why I truly would want the premium instead, but it's more expensive (900$ more right off the bat - and seemingly "1000$ is too much"). But 10-15 people doesn't really mean much in terms of how much they'll need (you can't even tell for sure if sharepoint is really useful for their particular version control scenario really). SQL express might work OK for small sharepoint installs (again, if they go for that in the first place), but if they need it for hosting their website (not small intranet - on the internet) etc, then it just might not cut it... Very nice DB indeed, but it's still limited. I'm not talking about large shops or redundancy (which is always nice to have). But rather a question of usage. Like I said before, it's not like you can guess exactly how much load they'll be putting on the servers... Even for a very small shop, Active Directory + File Shares + Sharepoint + IIS + SQL Server + Exchange is *really* pushing it (exchange alone is a pretty heavy app that's very RAM hungry). If they're going to send any amount of mail, want decent speeds, host a "real" website and all (a very likely scenario), one server just won't cut it. Ideally, it would be more than 2 servers indeed... But 2 might be the bare minimum requirement (again, no one can reliably guess how much load they'd put on servers based on so little infos - at least if we knew the type of business or something...) There's that possibility that 2000$ worth of software *might* barely suffice on the short term, but that's a bit of a gamble... I'd rather be on the safe side, and regardless, seemingly they can't afford half that, and hardware and all will cost a lot more too, so mostly irrelevant. I'm also not expecting just any shop with perhaps no IT department to just install/setup/configure/secure active directory, exchange, sharepoint, IIS, MSSQL and all by themselves with no previous experience, training or anything like that...

As fizban2 mentionned, the OS doesn't have version control built-in. From your previous posts, I thought you had special version control software to install on it (wasn't very clear). Sharepoint is a good option for office documents, otherwise (BTW, 4000$ + 70/CAL or so)... There are tons of systems for source code (think of CVS/SVN and all the others), but for just any files, I've never really seen anything (haven't looked extensively though). I don't know why you were even expecting things like this... And most of the versionning systems I've seen were somewhat complex. You might find software that does this, and it just might cost a lot too.

Well, if ~1000$ is too much, you might as well just forget about the whole thing really, as there will be lots extra cost unexpected or that you're not planning onto... From server hardware and UPS to other/misc licensing costs, network equipment, man-hours spent working on it (install/maintenance/etc), user/IT dept training or books, extra hardware/licenses if you end up needing more services/power than expected (i.e. the box might not be able to handle mail server + file server + domain controller and all by itself, let alone + web server / database too) and such... And buying SBS is by FAR the cheapest way (Win 2003 standard is like 1000$ + CALs, then exchange is like another 700$ + CALs again, etc) - and you're not even getting the "nice" version of SBS even (I would *definitely* want the premium ed but it's 1500$ instead). 10 CALs will likely not be enough depending on exact setup, best case scenario you'll be one user or one client away from needing another 5 pack (there's 2 types of CALs to choose from) - and I don't know where you're getting CALs this cheap (I'm tempted to think you're mistakenly looking at upgrade CALs - a 5 pack of CALs is more like 450$, so you're already over 1200$)

Why wait for Blu-Ray? -The format just may die i.e. lose to HD DVD - good ole format war yet again -It'll be like a couple of years at least before the writers are half-decently priced -And even then, 4 layer media may take more than 2 years to be somewhat cost effective (assuming Blu-Ray wins)... By that time, 400GB drives or more will be mainsteam. By the time we get decently priced 200GB media, people will have drives over 1TB or such (just look at the new 750GB seagate - in a couple years it'll be quite affordable and there will already be bigger ones). -and by "wait", you make it sound like he shouldn't backup until then or something... When it's available and cost effective (if it ever does), one should consider using it, but meanwhile it's irrelevant. One still needs backups. If you want to wait forever for some expensive technology, then you might as well wait for holographic storage like InPhase's, which will have large capacities (200GB to 1.6TB per disc) i.e. the smallest holographic disc will have as much capacity as Blu-Ray's maximum theorical capacity. Personally, I have no plans to buy a Beta-Ray drive, ever, at any cost. [edit - Delprat posted while I was writing this] "Magnetic storage is far more stable"... why ? The magnetic "data" on the platters is very very stable - you could leave it alone for MANY years with no problem. 'coz you think there's no "error correction bits" ? if that were the case, HD will always send random bytes ! You're misunderstanding. We're not talking about CRC-like error checking for transmission, or to verify if data blocks were valid - DVDs use those error correction bits to fix stuff that wasn't recorded properly (they even measure things like this i.e. BEL and such). All discs have loads of PI errors and some PO errors - no burn is ever "perfect", and cheap discs are quite bad (increasing burning speed only makes things worse - and who wants to burn slow?) HDs don't rely on that type of "error correction". Imperfect burns, combined with dyes that change over time (especially discs using organic dye - most discs do), and all the previously mentionned factors (temp, humidity, etc), imperfect manufacturing (very common - improper sealing, imperfections in plastic, etc), and some disc aging problems (like pin holes in some discs, plastic becoming cloudy on others, oxydization, etc), and with everyday wear (scatches and dust)... I've seen hundreds of discs gone bad... And for a reason. Plus, magnetic storage, as the name suggest, is subject to magnetic sensivity Not so. The drive's in a metal box (faraday cage). @SecretNinja: A ait2 drive alone will cost hundreds... Enough to buy a mega-huge RAID5 array... and the price of tapes ($/GB) isn't much lower than HDs (unless your stuff really compresses well perhaps)

Making backups of large amounts of data is a PITA... Current optical discs are way too small. A backup of a single 200GB drive is getting close to a full spindle of DVDs... No matter how cheap the plastic could be, that's just too much discs to swap - to create new backups every now and then (wasting a spindle every couple of months? no thanks!) and to restore if you ever need to. Too time consuming. The main option I see is more HDs. RAID0 is indeed a bit risky for data you can't afford to lose. RAID1 (mirrorred pair of HDs) would be secure, but halving your space sucks... So ideally you go for RAID5. Four 300GB SATA drives is not a whole lot of $ anymore (in some places the 400GBs are becoming cheap, but here they still cost as much as two 300GBs for one). One drive for parity, the other for data, so 4x 300GB would give you 900GB of pretty safe storage (any one drive can die and your data's safe). Including a 4 port SATA controller and tax and all, it should still be around 50 cents/GB. Preferably on UPS or at least a good surge supressor (don't wanna fry all drives if something happens). More of a "safe storage" thing than a backup really... Works great for my video server (all in mpeg4 - same deal for music collection). No worries of losing everything, but even if something happened, it's not the end of the world either. The other popular option is backing up on a external drive. Lots of people like this. But I'm not big on it... The drives are no cheaper than in the RAID5 scenario, and the enclosures only add to the cost (and most cheap enclosures suck and will die of heat - especially those with without fans - I'll link to some reviews if you need some "evidence" of that - I've given up on RMA'ing mine for another one that would last a whole month). And you gotta find a place to put all the external drives eventually... Which are also more prone to be dropped or something (from being manipulated/handled more). Either ways, it's far more convenient than swapping discs all day every month or whatever (50 discs x 10 minutes each or more) - just need to start copying files (a few seconds) or it can even be completely automated - no need to even worry about it anymore. Personally, I take the risk of leaving everything on RAID5. It's not ridiculously expensive anymore for lots of space. You get fast speeds, it's always online and all. For the few things that just CAN'T be lost, then those are kept on a smaller RAID1 from two smaller HDs (think family photos), and are also burnt on 3 taiyo yuden DVDs (two for me, and one "off-site" i.e. sent to my dad - he likes to have them too, and also for in case of fire). Taiyo Yuden are good discs. Don't use cheap stuff for important backups, you'll regret it the day you need it (things like organic dye and cheap manufacturing quality will have likely destroyed the data partly - seen it happen so many times) HD DVD/Blu-Ray/Holographic isn't there yet. Writers @ 1000$ and 25GB blanks for 60$/ea? No thanks! That's FAR more $/GB than any HD will cost you (would you buy a 25GB HD for 60$? me neither). Transfer speeds are still slow-ish. You'll still have to swap discs (backuping a 200GB drive with 25GB discs is "only" 8 discs - now, backuping several RAID arrays i.e. > 1TB? Spindles of 'em... They're not keeping up with HD size increases... And scratching/losing one disc will mean losing tons of information (kinda scary). Price isn't there and won't be for a while, and you'll still end up swapping discs. No thanks. DL: I disagree. HDs are far less likely to be corrupted (optical discs need error correction bits - not a perfect process by any measure). Magnetic storage is far more stable. DVDs aren't so "solid" unless they're pressed (pits and lands vs altered dye). Dye problems, light/atm pressure/humidity/etc, imperfect manufacturing, reflective top layer lifting off, you name it - it'll scrap your discs (I've seen hundreds of discs that "turned bad" over time - kept in a CD wallet in a cool place, no scratches or anything, and not cheapest brand either). And DVDs are very easy to scratch - a very common occurence of data loss... I'll take HDs over DVDs anytime. And over the years, I've never seen a single HD die because of a PSU problem or such - not even once, and we've got thousands of PCs at work... You're more likely to just drop an external HD on the floor instead. My 2 cents...

You can very easily do that using vbscript (not the best pick like I mentionned before, but it gets the job done... mostly?) First you enumerate your domain (sounds like you're using a NT domain by what you said), i.e. using GetObject("WinNT://DomainNameHere") then filter it, i.e. .Filter = Array("Computer") Then ping 'em ("ping.exe -n 1 ComputerName"), and if responsive: GetObject("winmgmts:\\ComputerName\root\cimv2") [might need to impersonate] then exec your WMI queries against the classes you want (they're all documented VERY well on MSDN) .ExecQuery("Select * from Win32_WhateverClassYouWantJustLookitUpOnMSDN",,48) and then read the properties you want, and finally write all that text to a CSV file (since you want an excel sheet) It's easier than it may sound. What you write to the CSV file (from the pic you posted): -first colum is just the name of the PC you're polling from (that you got from the WinNT:// query) -IP address (of first NIC polled...) like such -mfg and product name usually are empty... but you can still enumerate them to see if they return anything, like this -then OS info -processor info Having IP of first NIC only might not be a big deal, but here you'd only write the first CPU to your CSV file, and no mention if it has HT, is DualCore or anything either (there are ways to do this, i.e. Win32_Processor.ProcessorId returns EAX & EDX registers of CPUID called with EAX=1 - bitmasking does the trick; few apps bother with this unfortunately, even the commercial ones!) -enumerate RAM -logged on user (checkin time? sounds like the datatime at which it was scanned) You just need to lookup what the CSV format is like (comma separated... find one and look at it in notepad) to see what you need to write to your output file. I don't have a pre-written script to do all this for you unfortunately (don't have too much free time either). If you can't write it yourself, or can't adapt one, or that no one finds enough spare time to write one for you, then your best bet is likely to find an app that outputs tons of infos and select just the one you need from it.

Well, no need to thank me for those comments (I'd say that's more of a rant than anything). Sorry if I was a bit harsh, but I was in a bad mood (migraines don't help) and I get annoyed at ppl joining to "spam" an URL... There were a couple more yesterday (or was that the day before?), one of which even pointed to spyware... Anyhow. I see it's free, but the "hardware inventory" field is getting pretty cut-throat lately (I know, for I've been looking at the "competitors" in this market lately). Free alone just doesn't cut it. There's ridiculous amounts of apps that do this nowadays - some you gotta pay for (almost nothing to large sums), freeware ones, and even a bunch open source ones are popping up lately. I'd guess it's due to the large amount of ppl getting into WMI lately (just a guess). I'm expecting them to become even more popular quicker... The commercial ones overal are nice, but often are way too expensive for what they do... Kinda. Too expensive for small shops for the most part (like SMS - lotsa $ upfront), and if there's per-seat licensing, it quickly becomes overpriced (like Everest @ 10$/PC * a few thousand PCs). Either ways, there's lots of major competitors in the field that have lots of brand recognition helping their sales... And most companies that buy those big-name commercial apps are unlikely to change their ways - they'll just buy SMS regardless, at any price, no matter what... Those not going for the expensive commercial offerings are likely to settle for something as inexpensive as possible. Having to buy a SQL Server license or an extra Win2k3 license (for IIS) defeats the purpose. MSDE sucks bad. Least you could do is ensure you app supports SQL Server 2005 Express (free and doesn't suck like MSDE), and perhaps other DBs including free ones like PostgreSQL, Oracle & DB2 Express - and perhaps MySQL even though it's not really free (dual licensed), but just because so many places already have it installed and know how it works. XP Pro has IIS indeed, but it's so limited, it's only useful for really small things, and a PITA overall (create a DNS entry for it - just for that app?) and most companies wouldn't put this app on their intranet's IIS (fear of bringing the whole box down, already too high server utilisation, you name it). While I'm no fan of PHP at all (ok, I truly despise it), it has the advantage of running on a 100% free "LAMP stack" i.e. lower licensing costs overall (I've seen sevreal java-based ones too, running on free app servers too). Besides price (TCO - including the price of your and and that of the server software licensing), you still need more things... There's already a bunch of completely free apps out there - the app itself being totally free, the soft it runs onto (web server and database) and all. Some are even open source (just search for "inventory" on sourceforge to see a few)... And they're not necessarily complicated, hard to install, nor suck real bad. You gotta have something better than then (and it's pretty hard to beat totally free + open source really). Many, many shops these days aren't even using any of these solutions - they just use a simple vbscript that does the "polling" (and dumps to DB or text file or such). Again, totally free, and open source kinda... (you can even find lots of such scripts all over the web, and there's countless script resources, including script making tools & wmi tools, script repositories and examples, etc - even MS themselves have a lot about this stuff). No server stuff to install, nothing complicated, etc. Just run the thing... While I'm no huge open source fan (nice to find OSS stuff, but I don't adhere to the "Free" mentality part much), this is one type of app where it tends to matter a lot more than usual (the commercial offerings get away with it as they provide paid support and will release regular bugfixes and all). WMI scanning tends to be error prone. With only a handful of PCs everything seems to go right, but as you can a couple thousand PCs at a time, there's almost always something that will go wrong, no matter how well the app is written or how much error handling you got (some old PCs like to cause trouble every time it seems...). That, and you can be sure that people will want to customize something, or add/enhance something else... It's pretty much a given in this field. Flexibility is required and expected (they can already modify their free vbscripts to do whatever they want). And preferably integrate with other apps and info. So, if you want your app to gain some market share or whatever, you'd kinda have to find a niche in there... Not easy to do. Most people's needs are rather well met. The only way I see one could get people to use their app is to have something fundamentally different in some way. Gotta "think outside the box" and come up with something innovative (easier said than done). There's got to be a reason to use your app over everyone else's... I've made a hardware inventory app too (well, it does a lot more than just that). I was thinking of either starting a micro ISV to sell it, or as a last resort open sourcing it, and I'm somewhat giving up on both ideas... It's a multithreaded C# app (.NET 2.0). IIS required too for the web interface, and it's SQL server centric as well (could be easily ported to almost any DB in minutes though, just never had the need). Instead of having IIS do all the network scanning/pinging/polling and such, I have a desktop PC do all that stuff, IIS is only used to display the data. Kinda typical... Depending on config, it enumerates PCs from either AD, NT domain, or reads from a text file, then uses built-in ICMP ping, WMI-scans the responsive ones... But instead of only dumping straight to DB, I also dump *everything* to a [large] XML file. It's easier to retain all the details that way (most DB schemas will only have total RAM, as being able to have every stick of RAM in there would need a table for RAM sticks and FKs to it and all - kinda ugly and nedlessly heavy especially if you do it for everything in the entire app - one ends up with countless tables and joins, or some n00bs use 4 integer fields, and the app crashes when there are more sticks or such... the XML file is extremely flexible, keep all the info you want in a hierarchical format, parse & modify as you want/need... FAR more powerful, although it gets large quickly). IIS is used to display the polled data as webforms (as well as provide some web services and such - it's part of a larger n-tier soa app, and there's also a winforms "rich" client - which again does a lot more than just this... in fact, this is just an "aside" to pre-fill data and do auditing). About the interface, this is one point where you might need more work than you expect... It's pretty much what gives someone's first impression of your app too (if they don't like it, they won't even try it no matter how good the underlying architecture may be). I'm not proposing you go way overboard with usability studies and the whole 9 yards, but it needs to be good - REAL good. Especially if you web interface is your main interface/selling point... Columns sorted alphabetically don't cut it. At all. Ideally they'd be in some logical order, or order of importance (and some columns could be just considered useless clutter - ideally one can pick and choose which). That's just the starting point. Think of UML "use-case diagrams" - what's the end user going to do? Is having a menu that lets them list "all modems" installed across all PCs really that useful? Perhaps the feature might be useful too, but could belong in a sub-menu as it's not something frequently used? I'm thinking of tasks like "list hardware or software on a PC named XYZ", or perhaps tasks like "find PCs to lifecycle". Common tasks like those (or whatever) should have easy to find shortcuts that require no scrolling (list would ideally be easily customizable by/for customers). I have wizzards for tasks like "find old junk to lifecycle". Also, telling ppl they can just access the database and "have at it" for anything else isn't exactly great either... If they're willing to do this, my guess is they're already using a homebrew vbscript or app, or are using one of the open source apps that they've customized. You want some export features and at least basic reports. No need for complex/fancy business intelligence & data warehousing stuff, but somewhere between that and none... The app should also have some sort of built-in security (AD integration or form authentication or whatever). And that's if the app only does "WMI polling"... And I'm not suggesting it should only do that either (you can throw/integrate anything else useful in there: a more complete "non-computer" asset management part, an ITIL trouble ticket app, a knowledge base, bug tracker, anything goes!). And even if the app was perfect, there's the "getting it known"/marketing part... (and user docs, help files, etc - that's the part I hate most) Wow. Did I just write all this?

You joined to spam this. Nice of you... (looks like you've done this with more than one website too) I fail to see how it falls under "General Discussion". Spam coming from existing/participating community members is one thing, but posting for one's self-centered interests with no intentions of contributing anything besides spamming your URL... It doesn't bring/do anything new (there are FAR more "agentless" WMI-polling solutions out there from what I've seen). Pretty light on features - the only thing it has is a half-decent search feature (and some stuff is downright weird, like the side nav - limited usefulness IMHO, very poor column ordering at best (alphabetically? LOL!) and can't seemingly customize which to show either [it's show a useless "uninstalled" column for CPUs but won't show if it supports HT or such], a bunch of clutter one can't remove, etc). There are MANY competitors in this field (many simple, free, open source, running on free LAMP stacks, etc), you gotta bring something new if you want to succeed at this. Even for free it's not very interesting. Lots of downsides (like, only works with AD - not NT domains, which are still used in many businesses; poor DB support - MSDE is outdated and truly is "teh suck" and MSSQL is overly expensive for such a trivial app, app is not open source, rather limited features, .NET 1.1 instead of 2.0 [no master pages for "branding", etc], etc) Creating a similar app is easy and quick (speaking from experience). Being open source is a major plus in this field (I wouldn't be interested to even try one of those if I couldn't modify it to fit my needs - and this solution would definitely require it).

That's one thing I don't bother unattending. Only have so many VS2005 instances to install, and it's not like we have to reinstall it often or anything (haven't had to reinstall it even once since it's out of beta) That, and every install is always followed with tons of other stuff (various framworks [EntLib, WSE, etc], several TDD-oriented apps [TD.NET, CC.NET, etc], N-Everything [NUnit, NAnt, NDoc, etc], CodeSmith & MyGeneration plus a couple ORMs, Jetbrains apps, visualizers, Consolas and a couple other fonts, DB2 & Oracle "addins", a couple SCM tools, couple XML tools, a few nice components, lots of in-house stuff [snippets, components, templates, etc], some extra web dev stuff [for asp.net apps], and usually tons of other dev stuff like RKs, SDKs, IIS tools, etc) If anything, I'd rather install visual studio by hand, and have the rest of the stuff install unattended (and all the tweaks/customizations). That would save *FAR* more time (and it's way more tedious / error prone than installing VS too)

.

Clamwin is getting more popular lately - lots of ppl are talking about it (in a good way). From what I've seen, the app itself is OK, the problem is with the definitions. From the tests I've seen, it didn't get very good results (can't seem to find the link again), although that could have changed lately too... If you're restricting your choices to free apps, there isn't a whole lot of choice though. AVG, Avast, antivir, bitdefender all have free versions AFAIK. The problem is two fold. Not only you want something a free client that's unobtrusive (like no ads), but being an AV app it also has to have decent detection rates (defs must be good) and hopefully low false positives. Some test results (uses only the paid versions, but they should be even better then what you'd get using their free versions). AVG is 2nd last of all tested, avast is a little better, bitdefender quite good, antivir scores best of all 4. Also, a quick review of free AVs. (not a bad review - AVG is dead last again, and antivir still top) If you only want on-demand scanning though, there is always HouseCall.

If you can't use a USB drive instead (don't know if your BIOS supports booting from USB, can't look it up without a model # either), then there are other solutions, but not too many. The main one would be booting from a floppy, and installing over the network: -Boot from a Win98se floppy that has proper NDIS2 drivers for your network card and all (optionally remove/create partitions/format as desired - reboot as req'd) -map network drive (using "net use" - the command has help, and google is your friend; map to whatever share that has the i386 files - need valid credentials too) -install from that mapped drive (or copy i386 folder to temp FAT32 partition or such if you prefer and install from there)

1) The OS won't prevent that, nor really facilitate it (well, there are some utils that come with it, but they suck). It's somewhat irrelevant. You're looking at mail server software (there are tons of mail servers, for windows or linux or whatever). MS Exchange is pretty more or less the standard app for this under windows (especially because of outlook integration and features like calendaring and such). If you just want mail (and under windows), then there are plenty of other options, like hMilServer (decent, free & open source) and tons of commercial ones. 2) It'll definitely work for that. XP has file permissions (ACLs) indeed (all NT based OS'es using NTFS have that - of course XP's "simple sharing" must be disabled for one's sanity). But that's irrelevant. Using XP as a file server is directly against the EULA (low 10 device CAL limits and such) and probably wouldn't even work -- even 2003 web server ed is crippled (i.e. max 10 UNC connections) so you won't use it as a file server (why let you buy the 400$ web svr ed when they can force you to spend 1000$ for the std ed?) Samba could be an option too (if you have sufficient linux knowledge and can't afford or justify Win2003's costs for a simple/small fileserver) 3) Depends what type of version control you're referring to... If it's made for windows, there should no problems. (It's not like the filesystem has built-in file versionning or such - you need apps to do this). Don't know what for, what app exactly or such (perhaps you had something in mind?) More details needed to answer this really. 4) No problem - as long as it's configured properly (just like for any other OS). Server security is a topic of it's own... Not going to elaborate further (entire books are written on that subject alone) 5) Never heard of acomba before. Looks like a client app. I don't see what it has to do with using a server (perhaps there's a server "back-end" or such... don't know why you'd mention this otherwise). And like fizban2 said, you don't usually just throw up a server like that (regardless of network setup/existing architecture/whatever). No idea what OS you're running on workstations (is it even windows?), if you already have some server stuff running (AD? DCs? existing mail server? etc). If it's a small shop, SBS could be an option too. rjdohnert: care to elaborate on the non-win2k pro client problems? I've never experienced or heard of such problems before. Not sure what you're referring to.

Haven't played much with it, but not bad so far (only toyed around their demo web interface so far though - haven't tried the .NET client app). Like Everest, they're not really completely free either - their latest/best version i.e. the one being demoed - is 250€ and otherwise is erm... "donationware". Oh, and donations of 100€+ required even for the free version if you need any support (bugs or whatever). More € again if you want/need customizations... It better not get too expensive, otherwise one might as well get SMS instead. No idea what the old version is like... What features are lacking, etc. You're stuck with some old version which will likely get no updates (expect only paid support) There are a bunch of other downsides besides price too: old and crappy MSDE or SQL Server which hopefully you already have licenses for - and SQL Server only installs on a expensive *server* ed of windows excluding the cheap web server ed, and don't forget CALs - more $; *might* work with express ed but not supported (not mentionned), but NO other DBs are supported. The client needs to be run as part of the login script (I honestly don't care much for that... ran every logon uselessly, relies on client-side dependancies, etc). No idea on how their "rich" (i.e. non-web) client is, there aren't even screenshots available of it, and downloading/intalling the whole thing (client & server) to get an idea of what it's like is too much trouble. Hopefully it has more features than the web version, namely reporting, as the web ed only displays one category of info of a single PC at a time, no exporting features or anything (handy, but clearly lacking basic features). Their web interface needs IIS to run as well, hopefully you already have a Win200x web server (with enough resources left - the scanning part may be somewhat intensive too @ 25 simultaneous scans), otherwise lots more $ (and CALs, etc). While it looks not bad, I'd honestly rather do it myself. Completely free, you have the source code unlike this one (in the language of your choice too), so you can do the required changes, tweaks and bugfixes for free (which you're likely to need/want in the long run). And it doesn't take long to code something basic - perhaps quicker than it would take to install, deploy, integrate, debug, configure and test this commercial solution (need client apps deployed, login scripts modified, some app installed on IIS, new database on SQL Server, etc - lots to do). I would personally rather have a client app that does the scanning (in mosts cases, there are exceptions). Run it on demand, no server part to install/maintain/upgrade later/whatever, no expensive server soft req'd (SQL Server + a couple Win2003 licenses and CALs would be a few thousand $), keep data in a local DB of your choice or XML files or whatever - whatever you want. You pick the DB schema too (haven't peeked at their DB's schema, but I'm kinda scared to). Export as full featured "recursive" XML and then transform to a simple and clean DB schema, do auditing, import in excel/access, whatever you want. Far more flexible IMHO. I've seen a bunch of similar apps too. And while I'm no fan of PHP (more like I despise it) nor MySQL (second worst DB after MS Access IMHO), there are apps like winventory (on sourceforge) which do most of the same, truly cost nothing at all, runs on a totally free LAMP stack like XAMPP (runs on non-server OSes too), etc. It's easy to modify. And reporting wise, it's no worse than the other (i.e. pull raw data off DB if you want something else - same deal). No need for deploying client apps, adding a new app (and delay) to all the logon scripts, etc. If all one wants is hardware details or such, I would say this is sufficient and in pretty much all cases cheaper (and hey, worst case scenario, one could easily modify their scanning vbscript to fit their needs). Should be easy to port to another DB too like PostgreSQL or such (in a matter of minutes), something the other tool won't let you do (don't have the source). It's also an active project, which means you can expect it to get better over time (free updates/improvements/new functions) and should be easy to upgrade (the upgrade to the current version is 250€ - a one time deal, every future upgrade means more $) A quick search on sourceforge alone brings up many more too: inventory, zCI, OCS inventory, etc. I'm sure google has even more to offer... (including non open-source ones, like lansweeper)

There is no known/documented way to slipsteam it in MSSQL 2000 unfortunately (supposedly due to the fact it uses an IS installer). That, and I don't see why you would use such an old SP (unless your app will NOT work with SP4, which I highly doubt, having never seen such a scenario before. You'd hope they'd have updated it to work with SP4 otherwise). SQL Server 2005 is likely to get slipstream support eventually, but AFAIK SP1 is not slipstreamable (haven't even bothered trying, in good old SQL Server tradition)

Nothing against everest (nice app), but I wouldn't necessarily go that way. 20$/PC isn't too expensive, but as your network gets bigger, that starts to get like a fair amount of $$ (even with 50% rebates, if we included every PC in our organization across country with no rebates, we'd hit mid-6 figures... ouch! That's a lot for simple functionnality, really.) But if everest also does other things you need, it could be worth the price tag... Mind you, every admin utility out there seemingly does this (like Hyena). If all you want is simple OS infos, perhaps some config and such (free disk space, installed apps, etc), but mainly hardware specs... It's something trivial to do using simple scripting. You could write a full blown app to do this (I made a multithreaded C# app that does this and more), but it's not necessary. Plain old VBScript can do the job easily (the worst part of it is really VBScript's very poor error handling - no try/catch blocks here can really bite you in the a** when a remote PC acts up...) Just look up the various WMI classes you need at MSDN. There are tons of sample scripts/code out there (MSDN has a pretty good scripting center too), and there are even utilities that will create most if not all the code for you (Scriptomatic, WMI code creator, etc). Google could find you lots of such scripts (ppl have been doing this for years). The script itself is rather simple and straightforward: -enumerate PCs (using LDAP:// provider for ADSI, or WinNT:// provider for NT domains...) -ping PC (continue if pings OK, if doesn't resolve or timeout then don't) -run your WMI queries againt the remote PC (need admin rights on remote PCs) -log results to text/csv/xml file, update database, whatever you want WMI is *way* under-utilized (especially by tech folks like here, and most admins I see). It's pretty powerful/handy/time-saving. It's not just "read-only" - you can change settings on remote PCs and such with it, or spawn processes remotely and such - which is another option in your case. You could use a script to run devcon.exe on remote PCs, which might give you enough infos. If you decide to use scripting, there's a free script debugger (the tricky part is, you gotta run your scripts followed with the //X switch - yes, 2 slashes - and it'll launch it for ya, if you're lazy just make a registry entry to add a contextual "debug" menu entry when you right click vbs files). There's lots of stuff out there you can copy/paste from or just use for inspiration. Lots of ppl could help you with this (just ask away if you need help...) If you decide against scripting and would rather make a real app, you have endless possibilities of what to use. Personally, I use C# mainly, but lots of folks also use VB.Net or other languages. Perhaps the learning curve might be a bit higher (if you're not a programmer, don't know OOP concepts, don't know the frameworks, etc), but it's FAR more powerful and flexible (VBScript doesn't compare at all). If you wanna try it that way, Microsoft Visual C# Express is free and quite good (System.Management is what you're looking for here) If you look around enough, you might even find pre-made apps that do all or most of this already (on sourceforge perhaps). Google's your friend... Either ways, a little knowledge and a little sweat goes a long way. For the price of everest we could have hired a programmer for like 10 years, whereas you can whip up half decent in a few hours even if you're not a programmer (chances are doing the paperwork/going thru corporate red-tape, etc would take longer - to buy the pre-made solution i.e. MORE work!) It's tailored to your needs, expandable, reusable, customizable, you have the source, no licensing issues (no need to buy more licenses as you buy more PCs either, or even the bother of having to count licenses every now and then - waste of time), you can fix bugs yourself right now in a few minutes (instead of hope for some company to get around to it and release a new build in a couple of months), ultimate flexibility/interoperability (reuse data elsewhere, interact with anything, whatever), etc... And even if you were no programmer at all you could still hire someone to code something like that for quite cheap (how much one can charge for a easy 15 minute job of mainly copy and paste needing only a few tweaks that requires no real coding knowledge?)