cluberti

I do not believe this list is entirely inclusive, although it's possible there was a release of new keys after the download deadline extension (perhaps to trace those that got ahold of the beta after the extension).

Unfortunately, that's not as easy. Registry editing requires ACLs and perms set on the registry and the filesystem that would make the user, basically, an administrator for all intents and purposes. It might be better to change the account type temporarily, make the fix, then switch it back.

It's hard to say for certain, because this is kernel nonpaged pool corruption, but it looks like either pool was corrupted BEFORE this .dll loaded, or loading this .dll caused the corruption: // The crash: 1: kd> kb10 ChildEBP RetAddr Args to Child eb3283b0 804fe827 0000008e c0000005 eda15b75 nt!KeBugCheckEx+0x1b eb328778 80542095 eb328794 00000000 eb3287e8 nt!KiDispatchException+0x3b1 eb3287e0 80542046 eb328be0 eda15b75 badb0d00 nt!CommonDispatchException+0x4d eb328800 8054b95f e342f870 00000000 eb328820 nt!KiExceptionExit+0x18a eb328be0 805d00bc 84c78058 00000184 eb328c20 nt!ExFreePool+0xf eb328c00 805b1421 84c78058 00000184 eb328c20 nt!PsCallImageNotifyRoutines+0x36 eb328c48 805b1efe 84c084b8 03780000 eb328d18 nt!MiMapViewOfImageSection+0x4c1 eb328ca4 805b22c3 00000004 84e688d0 eb328d18 nt!MmMapViewOfSection+0x13c eb328d34 8054162c 000006d0 ffffffff 0012cd34 nt!NtMapViewOfSection+0x2bd eb328d34 7c90e4f4 000006d0 ffffffff 0012cd34 nt!KiFastCallEntry+0xfc 0012cc58 7c90d50c 7c91bd03 000006d0 ffffffff ntdll!KiFastSystemCallRet 0012cc5c 7c91bd03 000006d0 ffffffff 0012cd34 ntdll!NtMapViewOfSection+0xc 0012cd50 7c91624a 00143808 0012cddc 0012d304 ntdll!LdrpMapDll+0x330 0012d010 7c9164b3 00000000 00143808 0012d304 ntdll!LdrpLoadDll+0x1e9 0012d2b8 7c801bbd 00143808 0012d304 0012d2e4 ntdll!LdrLoadDll+0x230 0012d320 791b5cc3 03b84ad8 00000000 00000009 KERNEL32!LoadLibraryExW+0x18e // What we loaded right before the crash: 1: kd> .frame 5 05 eb328c00 805b1421 nt!PsCallImageNotifyRoutines+0x36 1: kd> dc 84c78058 84c78058 00f800ac e267a830 00000000 00000000 ....0.g......... 84c78068 00000000 00000000 00000000 00040001 ................ 84c78078 00000000 84c7807c 84c7807c 00040000 ....|...|....... 84c78088 00000000 84c7808c 84c7808c 00000000 ................ 84c78098 00010013 00000000 0a040001 53646156 ............VadS 84c780a8 0000ba10 0000bb0f 84cdfa18 84d64910 .............I.. 84c780b8 8458a2b8 84000018 0a060004 ee657645 ..X.........Eve. 84c780c8 00000001 00000001 853c2980 00000000 .........)<..... 1: kd> du e267a830 e267a830 "\Program Files\ATI Technologies\" e267a870 "ATI.ACE\CLI.Aspect.DisplaysColou" e267a8b0 "r2.Graphics.Shared.dll" I'd consider a complete *uninstall* of the ATI graphics package, and reinstall from ATI.

Use runas from the command line, would be my suggestion.

I'm not talking about a Live CD here, or firing up Qemu for a VM to run Linux from inside Windows (VHD boot is *not* a virtual machine) - what are you speaking of, and where's the documentation? I cannot find any, nor do I remember ever finding this in BeOS, specifically, either. I'd like to see some documentation if I'm wrong.

Don't be crazy, they were not. QNX is the only OS to have something similar, and it's not the same. We're not talking about a kernel image, or running a kernel image off of a non-native filesystem - this is an entire "volume" inside a VHD, that is bootable from another OS natively without a Virtual machine. It's not quite the same - and also, we're talking about Windows here, not a fringe OS - this brings virtual disk booting to the masses (well, once Win7 releases).

No, not necessarily. Make sure you've followed these instructions for creating a memory dump first. There are valid reasons why a crash wouldn't produce a dump file (for example, the paging file not being on the same partition/volume as the \Windows directory).

I think that depends on their contract with Microsoft - if they're eligible, I'd say yes, but it's always best to check with your MS Rep first just in case.

Also, if he wants to compare SQL, it would be better to compare the platforms it actually runs on - Server 2003 vs Server 2008 vs Server 2008 R2 beta. That would have been a far more interesting test indeed.

I think you're probably right - at MSFN, anything more than scripting is pretty much above and beyond what people come here for (that's mostly questions about Windows and associated Microsoft apps, and hacking said apps judging from the apps in the member apps section). I'd be interested to know, either publicly or in a PM, what dev forums you are speaking of to see if they're the same 2 or 3 I normally visit .

That is true - but I think it goes without saying that those of us who do this for a living don't need a forum for help on it, either. I asked for source because I wanted to see how gun did it, not because it's inherently useful. Also, being a programming section, putting a tool without source (as it appeared to slip gun's mind, so no worries) is kind of useless. Everyone should just lighten up a bit.

But where's the adventure of writing an entire app just to stop and start services?

Very nice - care to post the source for this little app?

The OP specifically said NO FLAMES!!! lol

Depends on the ISP, although you may be right. If it's just a modem, however...

If it's really important to you, use the Send Feedback links to file this.

Yes, you could conceivably create a folder structure as the "start menu" used to be, and then add that folder as a toolbar to the superbar (I've done this, just to see that it can be done). It's not quite the same, but it does work. By default, search uses contextual search - it searches indexed locations (including the start menu's contents) for applications, files, and text inside files. And for B, yes the order is basically as you see, and unfortunately this is coded into the shell .dlls, so to change this would involve rewriting the .dll files (there are no reg keys or other settings you can modify, currently, to change this behavior - it is similar to what you see in the IE address bar with regards to laying out favorites vs previously-typed URL history, etc).

You only need a router if you don't have enough public IP addresses (given to your modem by your ISP for handing out to your clients). If your ISP only gives you one public IP, you will need a device that can do network address translation and can act as a gateway (what you would understand as a router).

So noted. Credit given where credit is due.

It seems Gradius's guide was posted a little over 24 hours after aviko's (in polish). Gradius should be giving credit (at the very least) to aviko and link to his original posts, otherwise I will edit the post.

It seems that the post made by aviko yesterday (here points to two links that appear (from my very basic grasp of polski) to be what Gradius2 has posted. Can anyone confirm the published date on either of these two URLs, as in, were they up on the internet before January 18th 2009?

If you feel your drive is fast (even though the HDD tests say otherwise), then you need to file a bug with both Microsoft and Seagate to get it fixed. Just because a drive can seek or spin fast doesn't mean it's actually going to *perform* fast. Edit: Also, it does seem that the default jumper set up is indeed .... poorly chosen (thanks to accessdenied's keen eyes ).

Pattern found, problem solved.

JatinBeniwal- violation of forum rule 1a on a beta OS? Not very smart. Banned.

Normally I wouldn't touch this, as I generally find Win9x old and archaic, but what you say is absolutely absurd. Windows 98's *minimum* requirement was 16MB of RAM, and the recommended was 24MB. It isn't until you get up around 1GB that the Win9x design starts to show it's age, so having the same problems at 64MB of RAM and 320MB of RAM indicate that it's not a system memory problem at all. I can't believe I'm doing this, but I have to side with most Win9x guys here - 320MB of RAM is nothing for Win98 to handle, and your statement just shows your lack of understanding of the problem.