Jump to content


  • Content Count

  • Joined

  • Last visited

About cluberti

  • Rank
  • Birthday June 27

Contact Methods

  • Website URL
  • ICQ

Profile Information

  • Gender
  • Location
    a device attached to the internet
  • OS
  • Country
    United States
  1. While you've said the ISO is fine, I'd say Windows says something is amiss: C:\Users\Carl>err 80070570# as an HRESULT: Severity: SUCCESS (0), Facility: 0x4c5, Code 0xc7aa# as an HRESULT: Severity: FAILURE (1), Facility: 0x7, Code 0x570# for hex 0x570 / decimal 1392 : ERROR_FILE_CORRUPT winerror.h# The file or directory is corrupted and unreadable.# 1 matches found for "80070570"
  2. As long as it's downloading the updates from the catalog and not shipping them directly with the tool, it's fine. If you redistribute in any way (other than just a tool to download them for yourself), that's where you go into copyright and redist problems.
  3. I've replied to your PM - in the interim, it might be best to post instructions on how to go about doing this with a user's own ISO. Going through the process with Microsoft's legal folks will take time, and there's no guarantee that copyright permission will be granted. Good luck though - it'd be cool if you could swing this. If not, there are lots of ways to do this otherwise.
  4. I am the friend, and I assure you I get it. Hosting Microsoft code (whether that be Windows 7 ISOs, hotfixes, downloadable content, etc) would indeed not be legal. However, the reason I pointed out the updates ISO is because it's not that difficult to integrate updates yourself into your own code and recreate your own "updated" ISO.
  5. You haven't done all the steps necessary to make it work. See this KB article (it's for Vista, but applies to Server 2008 as well): http://support.microsoft.com/kb/942956
  6. This is what perfmon is for - specifically, the memory counters listed under the Process section. In looking at your process explorer output screenshot it would seem to indicate Avast, but that could just be a red herring (working set isn't very high for any one process, and Virtual Size counts both committed and uncommitted (just reserved) memory allocations made by the process. The only true way (short of uninstalling things and putting them back one at a time) to figure out memory leaks is to use perfmon and use the object counters under Process first to see which process you should be investigating, and what type of memory allocations it's using.
  7. Looks like windrvr6.sys is attempting to pass the pointer from rdi+0x34 to ecx - problem is, rdi is 0x0 - this is going to access violate. Since it's a kernel driver, this will cause a bugcheck. I'd get with the makers of windrvr6.sys (http://www.jungo.com/st/windriver_usb_pci_driver_development_software.html) to see what they think of this. You probably want to get a complete dump this time though.
  8. These all appear to be classic memory corruption problems - pages from RAM missing/corrupt causing PFN lookup errors, calldriver failures, Working Set trim failures, PoolTag free failures, and even a memory access violation in csrss.exe that caused one of the dumps. If both sticks work one at a time (and you've tested them both in another known working machine together to verify they don't cause the problems with each other), then you've got a problem with the motherboard in that system and it should be replaced. By the way, that is the motherboard in an EQS M64K9-ALV system from EQS Limited (out of business it seems), and it's an ATI Radeon M200 chipset. http://ati.amd.com/products/certified/eqs-mb.html
  9. You do get what you pay for - the case will be fine, but if you build a machine again in the near future re-consider the Cosmos 1000.
  10. You could always run resmon and make sure it is being flagged as "hardware reserved" - if so, you'll at least know it's the BIOS (and probably your onboard video).
  11. You could use resmon to get a high-level overview of memory usage (and that may be enough), but if not you're gonna have to use perfmon.msc and utilize the Process objects for "<All Instances>" (not "_Total"!!!), specifically the handle count, pool nonpaged bytes, pool paged bytes, private bytes, virtual bytes, working set, and working set - private.
  12. If you want to see a breakdown of memory usage, run resmon (Resource Monitor). As of RC1 it breaks down where all of your installed memory is allocated, even showing what is taken by hardware.
  13. Have those folders ever been on an encrypted volume? It's complaining about a thumbs.db file with an encrypted alternate data stream. So, either they were on a volume that was encrypted, or they're potentially a candidate for scanning for malicious data. In any case, if you can boot into ubuntu with the ntfs-3g package, you will be able to delete it. However, due to the encrypted ADS, you will not be able to whack it from within Windows (and without ntfs-3g, probably not get it from ubuntu either).
  14. This is a known issue with XP x64 (and 2003 x64) if you have S4 enabled in your ACPI BIOS.
  15. Are you able to open explorer and have it stay open for any length of time? Usually if you're getting an error in shdocvw, you have a CLSID file type handler or MIME type issue (shdocvw = SHell DOCument VieWer control). If you can boot up in safe mode, I'd download and run autoruns from sysinternals to disable anything non-Microsoft, and do the same when running ShellExView from nirsoft.net. You might also want to run an sfc /scannow to make sure all of the files on your box are kosher - but yes, shdocvw is crashing but likely because something is trying to load and failing, making it the victim (and hiding the source of the problem).
  • Create New...