Jump to content

Mr Snrub

Patron
  • Posts

    765
  • Joined

  • Last visited

  • Donations

    0.00 USD 
  • Country

    Sweden

Posts posted by Mr Snrub

  1. On 64-bit Windows there is no /3GB switch, applications have 2GB user-mode virtual address space if they are not compiled with /LARGEADDRESSAWARE, and 4GB if they are.

    So yep, that sounds like the ceiling your 32-bit apps are hitting.

    Native 64-bit apps would have a theoretical virtual address space of 8TB.

    Check out MSDN and Wikipedia for info.

  2. I tried this out. I installed it, and it ended up hosing up my explorer where it would crash every now and then.
    Given that explorer.exe and iexplore.exe share DLLs and addons in either will be loaded in the other, I would guess an extension problem.

    I had IE7Pro installed at first and there were quirks with the search box flashing and IE8 crashed once - given that the main features I like of IE7Pro are in IE8, I uninstalled it and had no problems after that.

  3. Prefetch is designed to be a list of binaries that individual applications will be requesting (immediately, or very shortly) after they are launched - by using these tiny files the OS can order its disk reads and have the files in memory before they are requested, removing latency from the application launch.

    (Superfetch is a further improvement that looks at what applications you typically launch, and prefetches the necessary files before the icon is even double-clicked.)

    So I would doubt removing prefetch files or disabling Superfetch could improve performance, but possibly the opposite.

    Superfetch only uses idle time to do its lower priority I/O, so even if it reads parts of files into cache and does not use them, it's not impacting user experience or delaying system services that want to use the disk.

    As for a community source (hence independent from Microsoft) regarding the prefetcher: http://en.wikipedia.org/wiki/Prefetcher

    A second myth is that the user should delete the prefetch folder contents to speed up the computer. If this is done, Windows will need to re-create all the prefetch files again, thereby slowing down Windows during boot and program starts until the prefetch files are created.
  4. Are you sure it's not a monitor setting? Maximizing your windows shouldn't "chop" any of your borders, just extend all of them to the edge of the screen...unless I'm not fully understading your problem.
    The maximize button is slightly different to "set the dimensions of this window to the same as the screen resolution" - the window loses its edge controls for resizing and moving (and on Vista it turns off the opacity feature for that window).

    I guess the OP is talking about the small number of pixels making up the window border which are used for resizing it as they are not required - to my knowledge there's no built-in way to change the maximize behaviour (it works like that for efficiency, as it no longer needs to worry about repainting windows behind it - hence why opacity also turns off in Vista).

  5. I'm not sure what creating an image of the hard disk or volume is meant to achieve differently - the drivers will still be incorrect or missing for the new hardware, leaving the system unbootable.
    Obviously you have little experience with imaging drives then. In the process i'm thinking of one root out drivers and anything identifying what the system was, tho keeping anything else, it quickly gets very complicated if one should start detailing what should/can be kept in order to re-use system.
    Actually, though it's been a few years I have used Ghost extensively as a method of creating "restore-inside-20-minutes" images along with a network boot floppy disk - I've not used imaging techs for migrating existing OS's to dissimilar hardware though, due to the problems described already.

    The process you are talking about sounds more like a sysprep than anything to do with imaging specifically, and as you said not practical for a single client machine.

  6. However that half open program, seems to be able to create updated security catalog files for patched file. Which i dont know if nlite does, tho one would not need it anyway, the security cat file i mean.

    That makes the file look legal to MS and it doesn't update it every time you go to Windows Update.

    It stops WFP from removing the file, but if a GDR hotfix is released for TCPIP.SYS then Windows Update would still replace it (and the .cat file).

    (And as Windows Updates are security and severe stability hotfixes only, it would not be smart to avoid using it because of chance of regressing a hack.)

  7. I install windows xp on my PC.

    Now I want to use this installed windows on another pc but when i attach this hard drive on other pc this is not running on this PC how can i use this window on other PC

    :unsure:

    It is not supported to even do a system restore from one hardware platform to another that is different, and moving the hard disk is effectively trying to do that.

    Also, if the XP x64 installation uses an OEM license then you are not legally entitled to move it to another machine, the license lies with the mainboard on the original computer.

    I'm not sure what creating an image of the hard disk or volume is meant to achieve differently - the drivers will still be incorrect or missing for the new hardware, leaving the system unbootable.

  8. and to Mr Snrub, im sorry bud i cannot find what you are looking for, my computer skills and knowledge is minimal and i am a noob in this field so to speak.
    What about the non-technical questions at the top?

    I find quite often when dealing with customers who open support cases that the "extra" information they don't think to provide initially can give clues as to the root cause of problems.

    There is a huge difference between a system that has been running stable for over a year and one that has just been (re)built - especially if a recent change such as added hardware or drivers being upgraded has been made, for example.

    So having an idea as to whether your system suddenly started having the problem "out of the blue" with no hardware or software changes would imply a hardware fault - conversely if some anti-virus or firewall product, for example, was installed shortly before the first bugcheck then it would be the thing to rule out.

    Fresh builds that are not stable are naturally the trickiest to resolve - there is no history to indicate whether the problem is provoked by software or hardware (sometimes you can get lucky by reproducing the problem in a specific way, but you have already indicated it is random).

    I would recommend that anyone running Vista should have SP1 applied, unless they have some specific software that is known to have problems with it.

    Did you follow the link to the instructions for configuring Windows to produce a complete memory dump instead of minidumps, for the next crash?

    That might give more information to show if the issue has a consistent root cause or not...

  9. Dual 24" widescreen LCDs were probably the biggest investment on my main rig at home, followed by the graphics card - when buying a new setup I tend to go for the "price break point" CPU as it's easily upgraded later (when the price drops further, so it's still not the most expensive component), and RAM is dirt cheap, but the graphics card I tend to go towards the (reasonable) upper end of the scale.

  10. I tried connecting the ethernet cable from "Cable Modem" to the DI-514 router at

    1) WAN port --> same IP in the two PCs

    2) One of the other four ports --> different IP in the two PCs

    ...Can I someway have different IPs with case #1? It would help me connect another Laptop and at the same time have different IPs in all machines.

    Nope, this is the scenario I first described, a home broadband NAT router is designed to have 1 public IP address and assign individual private IP addresses for the internal clients - the traffic from all clients on the Internet is seen to come from the 1 IP address.
  11. Not necessarily a problem - the startup & shutdown times are monitored for consistency, if it takes "longer than normal" then it might trigger these kind of diagnostic events - but for situations where hotfixes are installed or hardware is added, for example, the times may extend quite a bit for a one-off: the OS doesn't check for such exceptions and so reports it.

    The easiest logs to look at are under Event Viewer / Windows Logs : Application & System

    Any answers to the questions I posed?

    This might provide a bit of background to the root cause, or at least flesh out the problem a little...

  12. NAT takes care of that (and they get unique LAN addresses). From the info provided, I think that the problem is on the LAN side. If the router isn't indeed crippled (I've seen some crippled that ISPs give).
    It depends on whether the clients were getting public or private IPs directly from the cable modem, and what they get now - if it was an IP conflict I would expect only 1 of the 2 clients to even work, which is why I assumed the OP was talking about public IPs.

    If the cable modem and both clients are connected to LAN ports, it should act as a switch as before - does that help?

    That would make the network infrastructure identical as to how it was with the switch in place of the router (with the addition of wireless clients), so should resolve your problem.

  13. This is just a minidump, so maybe not enough to make a useful analysis - have you followed the procedure for generating complele memory dumps for future crashes?

    Has the system been stable for a while, and this problem just started, or was it recently built (or upgraded maybe)?

    Any overlocking in place at all?

    Windows Updates installed up to date?

    I notice SP1 is not installed - any reason?

    Am I right in assuming the mainboard is an Asus P5E?

    I see the BIOS version is 0203, but I can't get much joy from the Asus website to see what the most recent is - often the BIOS updates relate to CPU support and I see you have a quad-core which is guaranteed to be "kinda recent".

    Windows Vista Kernel Version 6000 MP (4 procs) Free x64

    Product: WinNt, suite: TerminalServer SingleUserTS Personal

    Built by: 6000.16584.amd64fre.vista_gdr.071023-1545

    Kernel base = 0xfffff800`01c00000 PsLoadedModuleList = 0xfffff800`01d9af70

    Debug session time: Sat Aug 30 17:59:19.878 2008 (GMT+2)

    System Uptime: 0 days 0:25:55.624

    3: kd> !sysinfo machineid

    Machine ID Information [From Smbios 2.4, DMIVersion 36, Size=2655]

    BiosMajorRelease = 8

    BiosMinorRelease = 12

    BiosVendor = American Megatrends Inc.

    BiosVersion = 0203

    BiosReleaseDate = 10/11/2007

    SystemManufacturer = System manufacturer

    SystemProductName = P5E

    SystemFamily = To Be Filled By O.E.M.

    SystemVersion = System Version

    SystemSKU = To Be Filled By O.E.M.

    BaseBoardManufacturer = ASUSTeK Computer INC.

    BaseBoardProduct = P5E

    BaseBoardVersion = Rev 1.xx

    3: kd> !sysinfo cpuinfo

    [CPU Information]

    ~MHz = REG_DWORD 2405

    Component Information = REG_BINARY 0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0

    Configuration Data = REG_FULL_RESOURCE_DESCRIPTOR ff,ff,ff,ff,ff,ff,ff,ff,0,0,0,0,0,0,0,0

    Identifier = REG_SZ EM64T Family 6 Model 15 Stepping 11

    ProcessorNameString = REG_SZ Intel® Core2 Quad CPU Q6600 @ 2.40GHz

    Update Signature = REG_BINARY 0,0,0,0,b3,0,0,0

    Update Status = REG_DWORD 6

    VendorIdentifier = REG_SZ GenuineIntel

    MSR8B = REG_QWORD b300000000

    IRQL_NOT_LESS_OR_EQUAL (a)

    An attempt was made to access a pageable (or completely invalid) address at an interrupt request level (IRQL) that is too high.

    This is usually caused by drivers using improper addresses.

    If a kernel debugger is available get the stack backtrace.

    Arguments:

    Arg1: 0000000000000028, memory referenced

    Arg2: 0000000000000002, IRQL

    Arg3: 0000000000000000, bitfield :

    bit 0 : value 0 = read operation, 1 = write operation

    bit 3 : value 0 = not an execute operation, 1 = execute operation (only on chips which support this level of status)

    Arg4: fffff80001c8e3b0, address which referenced memory

    TRAP_FRAME: fffff980113dc5e0 -- (.trap 0xfffff980113dc5e0)

    3: kd> .trap 0xfffff980113dc5e0

    3: kd> r

    Last set context:

    rax=0000000000000000 rbx=fffff80001c00000 rcx=0000000000000000

    rdx=0000000000000000 rsi=fffff80001d69200 rdi=000000000000000a

    rip=fffff80001c8e3b0 rsp=fffff980113dc778 rbp=0000000000000000

    r8=fffff980113dc7c0 r9=fffff88003aef600 r10=fffffa80079661f0

    r11=fffffa8007966120 r12=0000000000000000 r13=0000000000000000

    r14=0000000000000000 r15=0000000000000000

    iopl=0 nv up ei ng nz na pe nc

    cs=0010 ss=0018 ds=0000 es=0000 fs=0000 gs=0000 efl=00010282

    nt!MiFindNodeOrParent:

    fffff800`01c8e3b0 48f7412800ffffff test qword ptr [rcx+28h],0FFFFFFFFFFFFFF00h ds:00000000`00000028=????????????????

    3: kd> kv

    *** Stack trace for last set context - .thread/.cxr resets it

    Child-SP RetAddr : Args to Child : Call Site

    fffff980`113dc778 fffff800`01c7cfe7 : 00000000`00000000 fffff800`01cd503a 42506650`01c00000 fffffa80`04ded000 : nt!MiFindNodeOrParent

    fffff980`113dc780 fffff800`01cd576e : 00000000`00000000 00000000`000007ff 00000000`000018c0 fffff800`01d3d2a4 : nt!MiLocateAddressInTree+0x17

    fffff980`113dc7b0 fffff800`01ce44fe : 00000000`00054ad8 fffffa80`04dedbc8 fffffa80`0785d060 00000000`00000000 : nt!MiIdentifyPfn+0x77b

    fffff980`113dc850 fffff800`01fb3965 : fffffa80`04ded000 fffff980`113dcca0 fffff980`113dc918 00000000`00000000 : nt!MmQueryPfnList+0x13e

    fffff980`113dc890 fffff800`01e38c9c : 00000000`00000000 00000000`00000000 fffffa80`04ded000 00000000`00000001 : nt!PfpPfnPrioRequest+0x115

    fffff980`113dc8e0 fffff800`01ec06ac : 00000000`00000000 00000000`00000000 00000000`0222e940 fffff980`047d7501 : nt!PfQuerySuperfetchInformation+0x1db

    fffff980`113dc950 fffff800`01c4d733 : fffffa80`0785d060 00000000`042faff0 00000000`03d023e0 00000000`00000000 : nt!NtQuerySystemInformation+0x11aa

    fffff980`113dcc20 00000000`770d05da : 00000000`00000000 00000000`00000000 00000000`00000000 00000000`00000000 : nt!KiSystemServiceCopyEnd+0x13 (TrapFrame @ fffff980`113dcc20)

    00000000`0222e898 00000000`00000000 : 00000000`00000000 00000000`00000000 00000000`00000000 00000000`00000000 : 0x770d05da

    Crash occurred because CPU register RCX contained zero when it was meant to point to a memory address.

    RCX was populated by RAX, which is also zero, if we scan back a little bit - but once we've jumped into a couple of functions it's not reliable to make assumptions on the register contents.

    The STOP code here was 0xA, not 0x3B - which might indicate the bugchecks are different (possibly random) which could imply a hardware (or overheating/overclocking) problem.

    If you take a look in the System event log are there a bunch of BugCheck event 1001 entries corresponding with each crash?

    (Tip: Click on the column heading Source and wait until the "Sorting..." indicator has gone, then the events are sorted in alphabetical order.)

    If you find there are a collection of these, what does it say the bugcheck code was for each? (No need to record the parameters, just the single main code like 0xA for the one above.)

  14. I think it's a NAT router that doesn't support more than 1 public IP address - I couldn't see a way to even disable NAT, though I only went through the emulator briefly.

    If the switch is behind the cable modem and the clients are connected to that and get unique public IP addresses, what purpose does the router provide?

    (Typically home broadband routers allow multiple clients to access the Internet when only 1 public IP is available.)

  15. Strange, I just copied a minidump to my desktop and 7-zip was able to archive it from there... I did the following:

    Start / Computer

    Browse to C:\Windows\Mindump - click OK on prompt to get access

    Select multiple minidump*.dmp files, right-click and select Copy

    Show desktop, right-click and select Paste - files are now copied

    With desktop copies of files selected, right-click and go to "7-zip / Add to archive..."

    Did you MOVE the files by mistake, instead of COPYING?

    If I try to create an archive of the minidumps in the C:\Windows\Minidump folder, then 7-zip reports an access denied error similar to your screenshot.

  16. Do you have a memory dump file with today's date saved as C:\Windows\MEMORY.DMP ?

    (I think the default for Vista is to produce kernel memory dumps, which was the "beginning dump of physical memory" message you saw at the bottom of the blue screen, but I might be wrong.)

    If you have that file, copy & zip it up, then upload it somewhere like http://skydrive.live.com as a public file and give us the URL.

    If not, then zip & upload any minidump*.dmp files in C:\Windows\Minidump instead - also prepare the system to create complete dumps in the future by following steps 2-3 under "Memory dump of the entire system" in the following post:

    http://www.msfn.org/board/Creating-memory-dumps-t90244.html

  17. To ensure "like for like" comparison of startup times it might be good to define the start & end of the "start" process.

    e.g. from pressing the power button to reaching the user selection screen, or from the end of POST to the desktop appearing, etc.

    Some systems have a large delay during the hardware detection or PXE boot during POST which aren't anything to do with the OS at all.

    For extra "useful comparative information" it might be useful to indicate anti-virus & personal firewall products installed, as my experience has been that these can affect startup times and "responsiveness" of the OS considerably.

  18. i ran the device on the safe mode it was working just fine but when i restarted it. it gave me the msg" that the hard drive s not detected error 0418 or something"

    i researched the error and it means the hard drive needs 2 be replaced..so i figured ok the hard drive maybe damaged by the fall.

    i went and bought a new (WD 160gb SATA)hard drive hoping that it will solve the problem..but no i tried 2 install the Vista and it crashes when the instillation starts..

    XP and ubuntu cant detect the new hard drive..

    First, please don't copy/paste the same question 3 times in the same forum, ressurrecting posts from 3 years ago (I have removed the 2 duplicates).

    I would assume the reason XP & Ubuntu are unable to see the hard disk is that it is a SATA drive and they have no in-box drivers - however if the Vista installation fails then it might indicate a hardware problem elsewhere.

    More details on exactly when the Vista install crashes and what is displayed on the screen may be useful - e.g. does it get to the drive/partition selection screen and start copying files, then reboot during this stage, or get stuck in a reboot loop after the first restart?

  19. I like the InPrivate browsing mode which doesn't bother to cache any information about browsed pages - very handy if borrowing someone's PC briefly without relogging.

    The real SLD is also highlighted in the address bar, which is great for visual security (further protection against phishing).

    The find-as-you-type feature I love - this was pretty much the only reason I installed IE7Pro, so that's gone from my system now.

    Opening a new tab also allows you to reopen tabs closed in the current session, which is nice as I asked for that feature in 7.0 and missed the boat.

    I'm liking this a lot so far :)

  20. You could try using CACLS and RD in a batch file, something like this:

    cacls "C:\Documents and Settings\<profilename>" /T /G <yourdomainname\yourusername>:F
    rd /s /q "C:\Documents and Settings\<profilename>"

    The first line should recursively grant full access to the profile folder and everything inside it for your user account.

    The second should silently remove the directory and all its contents.

    I've not tested this at all, so check it manually on a non-important server first :)

  21. It sounds like the laptop has a SATA controller and therefore disk, and there is no in-box driver for the controller on the XP CD (bear in mind the OS came out years before SATA existed, so it's not unusual).

    You would need to find the SATA controller driver for Windows XP for the laptop on the manufacturer's website, and copy it to a floppy disk.

    From the comment about an external floppy disk drive, I assume they mean the laptop has no internal floppy drive?

    Then you would need to boot from the XP CD and hit F6 when prompted and insert the floppy disk with the driver present - this will allow Setup to see the hard disk and write to it.

    However, if you are not very techy then you might want to reconsider the manual OS downgrade - you may have issues finding drivers for the rest of the hardware (display, network, chipset, touchpad peripherals typically).

    It would be much better if the vendor could supply the laptop with the OS downgraded (complete with appropriate recovery CD), if they provide that service.

    [Moving to XP forum as this is not related to Vista.]

×
×
  • Create New...