"All your drives are belong to..." (Rant)



  • This is something which has bugged me for years, Windows Update downloads hotfixes and wotnot into SoftwareDistribution right...

    Then it runs them which means they get extracted.
    Some of them probably involve Windows Installer.

    I've never bothered to find out which is to blame but I continually find it annoying to find folders with names like F:\4c8ad71cdf3c3031a87bb6c56e lying about in the root of drives

    The drive where the operating system is installed, okay maybe I can understand that, it is the lazy and untidy equivalent of "Sharon" from admin who saves everything in the root of drive C: but at least the operating can claim some right to work with files on the drive where it is installed (even if it should use TEMP for TEMP files).


    What gets me is that it uses any drive it likes the look of, often one with lots of free space and it doesn't really care if that drive belongs to it or not.


    This then combines with something that used to be another pet hate of mine, write caching on removeable drives although I noticed recently Microsoft do now tend to treat my USB devices as such and not cache writes to them, using my removeable drives as if they are scratch space for Microsoft to perform operating system updates has bought back a similar problem, I unplug a drive I am not using and something breaks, in this case whatever service pack or update was running at the time.


    I know its not a biggy, it is made slightly worst by the fact I'm re-installing an ancient laptop for someone and as such every update takes some time.


    And this all brings me to two other rants which lead on from the point that whatever update is creating these folders does not delete them when it is done.


    Microsoft have been for many years now treating hard drive space like its free, they have features which aren't joined up as if developed by different companies (left hand/right hand) and they often end up causing conflicts or quite literally, duplication.
    Download update into SoftwareDistrubtion (1 copy) complete with the installer which is same for every hotfix but must be downloaded (although they have fixed that now), extract to some random folder (2), create a system restore point (3), apply fix and backup files to $NtUninstall...$ (4), if windows update was used keep a copy of the original package tucked away in Installer (5)

    Then, and this is just a perfect example, I had two windows installs on partitions on the same drive who could see each other's file systems. Now admittedly one of them was Vista so I never expected it to end well. Both decided system restore should be on by default and should track changes on the other drive...

    You do the math ... it was only a matter of switches between operating systems before one of them backed up the other complete with the changes the other had backed up of itself before it became an exponential runaway train.


    To get slightly more back to the original point, another two things spring to mind all of a sudden!
    thumbs.db - Its okay, I like thumbs.db, I mean if you are going to have a thumbnail feature it makes sense to keep it in the folder so that it moves when you move the folder rather have to be recreated ... but is it really more important than the actual work its thumbnailing.

    Create a folder, maybe I just call it 'New Folder', start moving photos into it. Rename it - Can't, being accessed. Okay so I call the folder what I want first, then I want to move it onto the network share where it shall live once I have completed making my changes locally - Can't do that either, being accessed.
    Basically the message is 'Pretty thumbnails are actually more important that letting the user get on with work.


    And finally, desktop.ini. I write a CD, I close the CD! But still windows thinks, you know what I'm going to auto create a desktop.ini (and maybe a thumbs.db too!) and keep remind you there are files to be written.



  • The thing I don't get is Windows XP used to have thumbnails that came up just as quick (if not quicker) than Vista, 7, or 8, and did so without any thumbs.db file. And now OSX does the same thing. Kubuntu doesn't, but the thumbnails are also dogshit slow, so I wouldn't call it a gain.



  • This is why a 32 GB Surface has about 10 floppy disks' worth of space available.

    You end up with crap sitting in SoftwareDistribution, C:\Windows\Installer, temp folders in the root of whatever drive the installer picks randomly... They seem to keep all this crap around so that you can install hotfixes to switch to an LDR branch for certain binaries and have the newer already-installed system updates reapply LDR patches as necessary. Which is all stupid. How about just redownloading files as needed when the user wants to install hotfixes, as opposed to keeping local mirrors of all your branch builds "just in case"?



  • More than any of those, thumbs.db is what annoys me. I currently have a working copy checked out from SVN that has a folder of PNGs, so naturally there's a thumbs.db in there. Except it doesn't show, even with hidden and system files shown. I can't delete it, even using cmd.exe and the exact path (since it won't show in Explorer). But everytime I try to commit my working copy it wants to add thumbs.db to the repo.



  • @Master Chief said:

    Windows XP used to have thumbnails [...] without any thumbs.db
     

    What are you saying?

    It sounds like you're saying that XP didn't use thumbs.db at all,  which is incorrect.



  • @dhromed said:

    @Master Chief said:

    Windows XP used to have thumbnails [...] without any thumbs.db
     

    What are you saying?

    It sounds like you're saying that XP didn't use thumbs.db at all,  which is incorrect.


    Did it? Huh. Never noticed them before Vista. Noted.


  • Discourse touched me in a no-no place

    @mott555 said:

    More than any of those, thumbs.db is what annoys me. I currently have a working copy checked out from SVN that has a folder of PNGs, so naturally there's a thumbs.db in there. Except it doesn't show, even with hidden and system files shown. I can't delete it, even using cmd.exe and the exact path (since it won't show in Explorer). But everytime I try to commit my working copy it wants to add thumbs.db to the repo.
    svn propset svn:ignore thumbs.db



  • @Master Chief said:

    Did it? Huh. Never noticed them before Vista. Noted.
     

    Ok. Weird though, since they were just as hidden in Vista+ as in XP.



  • @dhromed said:

    [Filed under: Thursday

    Your calendar seems to be running a day slow.  Better wind it some more.

     



  • @EncoreSpod said:

    I've never bothered to find out which is to blame but I continually find it annoying to find folders with names like F:\4c8ad71cdf3c3031a87bb6c56e lying about in the root of drives
     

    Windows extracts to the drive that has the most free space, which often is an external or data drive. I know, they could do something less retarded such as actually using %temp% and checking how much space is left and estimating how much is needed... but isn't that just too much work?



  • @db2 said:

    This is why a 32 GB Surface has about 10 floppy disks' worth of space available.

    You end up with crap sitting in SoftwareDistribution, C:\Windows\Installer, temp folders in the root of whatever drive the installer picks randomly... They seem to keep all this crap around so that you can install hotfixes to switch to an LDR branch for certain binaries and have the newer already-installed system updates reapply LDR patches as necessary. Which is all stupid. How about just redownloading files as needed when the user wants to install hotfixes, as opposed to keeping local mirrors of all your branch builds "just in case"?

    What I like in Windows 8 is that it's possible to let Windows reinstall itself (there is a button in the Settings page). No need for a OEM recovery partition or for a Ghost image when you simply want to reset a machine before you give it to your neighbor's cute daughter (she's 18 btw). It's all there, built-in, for the mere price of using a shitload of GBs on the expensive SSD.

    Clearly the problem is the size of the SSD. I think Microsoft is on the right path: include the features you need and let the hardware catch up.



  • @Ronald said:

    @db2 said:

    This is why a 32 GB Surface has about 10 floppy disks' worth of space available.

    You end up with crap sitting in SoftwareDistribution, C:\Windows\Installer, temp folders in the root of whatever drive the installer picks randomly... They seem to keep all this crap around so that you can install hotfixes to switch to an LDR branch for certain binaries and have the newer already-installed system updates reapply LDR patches as necessary. Which is all stupid. How about just redownloading files as needed when the user wants to install hotfixes, as opposed to keeping local mirrors of all your branch builds "just in case"?

    What I like in Windows 8 is that it's possible to let Windows reinstall itself (there is a button in the Settings page). No need for a OEM recovery partition or for a Ghost image when you simply want to reset a machine before you give it to your neighbor's cute daughter (she's 18 btw). It's all there, built-in, for the mere price of using a shitload of GBs on the expensive SSD.

    Clearly the problem is the size of the SSD. I think Microsoft is on the right path: include the features you need and let the hardware catch up.

    And if the install partition gets corrupted... Well, we'll let you sort that out. Go buy more install disks.



  • @dhromed said:

    @Master Chief said:

    Did it? Huh. Never noticed them before Vista. Noted.
     

    Ok. Weird though, since they were just as hidden in Vista+ as in XP.

     

    And are just as much of a pain. I needed Unlocker under XP and Win7 clients to scrub some directories that apparently were in use by Explorer wanting to update thumbnails of images I was about to delete.

    And why does the view not get preserved upon renaming folder? It seems thumbs.db needs to be regenerated each time folders are moved.

     



  • @Master Chief said:

    @dhromed said:

    @Master Chief said:

    Windows XP used to have thumbnails [...] without any thumbs.db
     

    What are you saying?

    It sounds like you're saying that XP didn't use thumbs.db at all,  which is incorrect.


    Did it? Huh. Never noticed them before Vista. Noted.

    Thumbs.db has even been in use since Win95/98 with IE4 1 and the Mac OS X equivalent .DS_Store has been in use since at least 20032

    So they are nothing new, and OSX doesn't "now" do the same thing, it has done that for years.



  • @dtech said:

    @Master Chief said:
    @dhromed said:

    @Master Chief said:

    Windows XP used to have thumbnails [...] without any thumbs.db
     

    What are you saying?

    It sounds like you're saying that XP didn't use thumbs.db at all,  which is incorrect.


    Did it? Huh. Never noticed them before Vista. Noted.

    Thumbs.db has even been in use since Win95/98 with IE4 1 and the Mac OS X equivalent .DS_Store has been in use since at least 20032

    So they are nothing new, and OSX doesn't "now" do the same thing, it has done that for years.

    In that case they worked a lot better in XP and previously. I didn't know they existed until Vista, and that was because they were causing havoc with my folder icons and whatnot.


  • @Master Chief said:

    I didn't know they existed until Vista, and that was because they were causing havoc with my folder icons and whatnot.
    In XP, the only time Explorer would generate thumbnails is if you put a window into Thumbnail View, which was the default only for My Pictures and its subfolders. I believe Vista got a little more aggressive about thumbnailing everything it could find.



  • @flabdablet said:

    @Master Chief said:
    I didn't know they existed until Vista, and that was because they were causing havoc with my folder icons and whatnot.
    In XP, the only time Explorer would generate thumbnails is if you put a window into Thumbnail View, which was the default only for My Pictures and its subfolders. I believe Vista got a little more aggressive about thumbnailing everything it could find.

    .docx? That's an image format, right?



  • @EncoreSpod said:

    using my removeable drives as if they are scratch space for Microsoft to perform operating system updates has bought back a similar problem, I unplug a drive I am not using and something breaks

    Next time, unmount it first, maybe?

    @Master Chief said:

    I didn't know they existed until Vista

    Unlike every non-Windows-user who's ever received a directory full of images from a Windows user.



  • @Gurth said:

    Unlike every non-Windows-user who's ever received a directory full of images from a Windows user.

    Macs are way more annoying, tending to clutter up file systems with those ._filename files. One for every file! Plus the .DS_Store files.



  • @EncoreSpod said:

    I've never bothered to find out which is to blame but I continually find it annoying to find folders with names like F:\4c8ad71cdf3c3031a87bb6c56e lying about in the root of drives
    Generally .NET updates, for what it's worth.



  • @Zemm said:

    Macs are way more annoying, tending to clutter up file systems with those ._filename files. One for every file! Plus the .DS_Store files.

    I know, though luckily only on non-HFS+ filesystems. Very annoying when you need to put files onto a memory card for media players that choke on files they can't actually read, though. (Hint: make sure the directory isn't showing in a Finder window and then do rm ._* in it.)



  • @Gurth said:

    @Zemm said:
    Macs are way more annoying, tending to clutter up file systems with those ._filename files. One for every file! Plus the .DS_Store files.

    I know, though luckily only on non-HFS+ filesystems. Very annoying when you need to put files onto a memory card for media players that choke on files they can't actually read, though. (Hint: make sure the directory isn't showing in a Finder window and then do rm ._* in it.)

    Can we, as a world, agree to put temporary files in the system's temporary folder and delete them when we're done with them? I'd be okay with explorer/finder/whatever being a little slower once per reboot per folder if it meant not having tons of shit thrown at my stuff.



  • @dtech said:

    Thumbs.db has even been in use since Win95/98 with IE4 …

    I don't recall having that in Windows 2000 — I seem to recall that it used Alternate Data Streams, which for me seems to be the most logical option. I think I may have seen Win2k use both — it may have only used thumbs.db on FAT32 or something, not really sure. I've had Windows 8 throw a strop many times over my trying to move a folder; since thumbs.db is gone, I have no idea how Explorer can still lock itself out of its own folders.

    Windows 8 no longer remembers thumbnails permanently, at least, not for me. I have 54 MB of thumbnail cache (not a whole lot) but thumbnail data is never committed to disc. Since I have an HDD, I get to watch Explorer sit there extracting thumbnails manually, slowly, one by one. It will cache them in RAM as long as it can, but not to disc. Pretty lame.

    As for the OP, what's more annoying about those installer directories dumped at the root of a random drive is that, often, they're not cleaned up, and they're locked down so that you can't remove them manually without a custody fight against the OS. (UAC and its ilk cause a lot of problems: it destroyed all the workarounds for permitting non-admin users to install fonts (since Windows still has no per-user font installation), and it also prevents you from deleting corrupt print jobs, while Microsoft have still failed to add a single ounce of troubleshooting to the Print Spooler service. Print Spooler will still crash on load endlessly if there's a bad job in the queue, without ever twigging that there's a problem, and offering no official way to ever resolve it. How hard would it actually be to have a flag that detects that the spooler crashed when loading a print job, and throw it away? Otherwise, you have to open up the spooler directory and remove the files by hand, except you can't any more.)



  • @Zemm said:

    @Gurth said:
    Unlike every non-Windows-user who's ever received a directory full of images from a Windows user.

    Macs are way more annoying, tending to clutter up file systems with those ._filename files. One for every file! Plus the .DS_Store files.

    And what about those lousy ZIP files sent by web designers that contain a clone of the entire archive tree under __MACOSX. This is a lot worse than a stranded thumbs.db.



  • @Ronald said:

    @Zemm said:
    @Gurth said:
    Unlike every non-Windows-user who's ever received a directory full of images from a Windows user.

    Macs are way more annoying, tending to clutter up file systems with those ._filename files. One for every file! Plus the .DS_Store files.

    And what about those lousy ZIP files sent by web designers that contain a clone of the entire archive tree under __MACOSX. This is a lot worse than a stranded thumbs.db.

    What's the "rationale" for that "feature" even existing? So they can advertise that their archiver program can handle bigger archives with the same input?


  • ♿ (Parody)

    @Ben L. said:

    What's the "rationale" for that "feature" even existing? So they can advertise that their archiver program can handle bigger archives with the same input?

    It made a lot more sense before the RDF started to fade.



  • The __MACOSX directory and ._* files are used for the same purpose: for storing metadata and the "resource fork" of files on filesystems that don't support said features, which would make it backwards compatible with the ancient OS (7? 6? Older?). Still stupid when those features are no longer properly used! One of the first tasks I was involved with at my current job was modifying a script to handle these files gracefully. (It searched the uploaded zip file for all *.csv. It didn't have much input validation so some binary garbage ended up in the database) I thought it was a bit late to be doing that in 2012, but anyway.



  • @Zemm said:

    One of the first tasks I was involved with at my current job was modifying a script to handle these files gracefully.

    What did you do to deserve that? Sodomized the DocuCentre during your orientation day? Parked your car in the CEO's spot? Asked to the new receptionist "what happened to the cute one"? Requested to get a Macbook instead when they gave you a Thinkpad?



  • @Ben L. said:

    Can we, as a world, agree to put temporary files in the system's temporary folder and delete them when we're done with them?

    The Mac's ._* files aren't temps, though. They're resource forks. If I copy My Wonderful Essay from a native Mac filesystem to some other filesystem that doesn't support multiple data streams per file, the data fork ends up in My Wonderful Essay and the resource fork in ._My Wonderful Essay.

    Macs, being somewhat POSIX-compliant, won't show you the ._ file by default If you use a Mac to do normal file handling operations like move, copy and delete to the visible data fork file, it will shadow the same operation on the hidden resource fork file so they always travel together. Move a data fork file back to a native Mac filesystem, and the Mac will put it in the data fork and put the corresponding ._ file in the resource fork. So from a Mac user's point of view, the ._ files essentially give you the ability to work transparently with non-native filesystems.

    So when you plug the same USB stick into a Windows box and find it littered with mysterious ._ files, you really don't want to delete them - unless you're sure you don't need what's in the resource fork. Sometimes all that's in there is fiile type and creator application metadata, which you can generally work around the lack of; sometimes there's important stuff like fonts and icons.



  • @flabdablet said:

    @Ben L. said:
    Can we, as a world, agree to put temporary files in the system's temporary folder and delete them when we're done with them?

    The Mac's ._* files aren't temps, though. They're resource forks. If I copy My Wonderful Essay from a native Mac filesystem to some other filesystem that doesn't support multiple data streams per file, the data fork ends up in My Wonderful Essay and the resource fork in ._My Wonderful Essay.

    Macs, being somewhat POSIX-compliant, won't show you the ._ file by default If you use a Mac to do normal file handling operations like move, copy and delete to the visible data fork file, it will shadow the same operation on the hidden resource fork file so they always travel together. Move a data fork file back to a native Mac filesystem, and the Mac will put it in the data fork and put the corresponding ._ file in the resource fork. So from a Mac user's point of view, the ._ files essentially give you the ability to work transparently with non-native filesystems.

    So when you plug the same USB stick into a Windows box and find it littered with mysterious ._ files, you really don't want to delete them - unless you're sure you don't need what's in the resource fork. Sometimes all that's in there is fiile type and creator application metadata, which you can generally work around the lack of; sometimes there's important stuff like fonts and icons.

    Of course, if you're using NTFS, which actually does support resource forks (called Alternate Data Streams - and yes, even over SMB) - it still scatters .DS_Store and __MACOSX folders everywhere.

     



  • @Daniel Beardsmore said:

    As for the OP, what's more annoying about those installer directories dumped at the root of a random drive is that, often, they're not cleaned up, and they're locked down so that you can't remove them manually without a custody fight against the OS. (UAC and its ilk cause a lot of problems: it destroyed all the workarounds for permitting non-admin users to install fonts (since Windows still has no per-user font installation), and it also prevents you from deleting corrupt print jobs, while Microsoft have still failed to add a single ounce of troubleshooting to the Print Spooler service. Print Spooler will still crash on load endlessly if there's a bad job in the queue, without ever twigging that there's a problem, and offering no official way to ever resolve it. How hard would it actually be to have a flag that detects that the spooler crashed when loading a print job, and throw it away? Otherwise, you have to open up the spooler directory and remove the files by hand, except you can't any more.)

    Don't get me started on removing corrupt print jobs. Print jobs on my parents' computer (windows 7) often get corrupt for a mysterious reason and hang up the spooler completely. The only way to get it resolved is to unplug the printer, reboot, stop the spooler and hope it lets you remove the files from the spooler directory (which 9 out of 10 times works). It would save me a lot of time if I can just tell them to remove the job from the spooler window and try again, instead of doing this procedure myself because it gets to complicated for them.

     On the dumping of installer stuff on root directories, MSSQL does this as well as I discovered when I found I needed to install some components, after I mounted the Truecrypt container with the data I wanted to add to the database. The installer ended up on the root of this container which is on a USB 2.0 thumb drive, making the installation process very slow. This thread reminds me to check whether it cleaned up after itself or not.



  • @Quincy5 said:

    Don't get me started on removing corrupt print jobs. Print jobs on my parents' computer (windows 7) often get corrupt for a mysterious reason and hang up the spooler completely.

    That's with updated drivers, I take it? (I don't know whether changing the print processor to WINPRINT would help there.) Problems like this tend to be related to a specific program, such as Adobe Reader, that might need to be configured to print in a different way.



  • @Quincy5 said:

    It would save me a lot of time if I can just tell them to remove the job from the spooler window and try again, instead of doing this procedure myself because it gets to complicated for them.
    Do they have a HP LaserJet. I don't remember which model exactly, but two clients have the same LaserJet printer, and at both certain jobs would regularly throw the spooler in crash loop. I ended up writing a batch file that cleans the spooler for them:
    @echo off
    if [%1]==[run] goto run
    "%~dp0<a href="http://code.kliu.org/misc/elevate/">elevate.exe" -c -w "%0" run
    goto :eof
    :run
    net stop spooler
    set sysdir=system32
    if not exist "%SystemRoot%%sysdir%\spool\printers" set sysdir=sysnative
    del /y "%SystemRoot%%sysdir%\spool\printers*"
    net start spooler



  • @ender said:


    "%~dp0<a href="http://code.kliu.org/misc/elevate/">elevate.exe" -c -w "%0" run

    THAT'S A VIRUS IT ENDS WITH EXE



  • @Ronald said:

    @ender said:

    "%~dp0\elevate.exe" -c -w "%0" run
    

    THAT'S A VIRUS IT ENDS WITH EXE

     



  • @mott555 said:

    More than any of those, thumbs.db is what annoys me.

    I've kicked off huge copy jobs on Windows and returned hours later to find it sat there waiting for me to tell it whether or not to replace thumbs.db.



  • @RTapeLoadingError said:

    @mott555 said:
    More than any of those, thumbs.db is what annoys me.

    I've kicked off huge copy jobs on Windows and returned hours later to find it sat there waiting for me to tell it whether or not to replace thumbs.db.

    That's why God created robocopy.



  • @Ronald said:

    That's why God created robocopy.

    I think you'll find that was Satan …



  • @Ronald said:

    @RTapeLoadingError said:
    @mott555 said:
    More than any of those, thumbs.db is what annoys me.

    I've kicked off huge copy jobs on Windows and returned hours later to find it sat there waiting for me to tell it whether or not to replace thumbs.db.

    That's why God created robocopy.

    That's New Testament touchy-feely God and I've always been an Old Testament fire, brimstone and Windows-Explorer-for-file-copy kind of guy. If Windows Exporer was good enough for Moses then it's good enough for me.

    Religion aside, I'm disappointed that the operating system's standard file copy method (Windows Explorer) isn't a bit smarter. Has anyone anywhere ever wanted a multi-GB file copy to stop part way through so that they make a call on Thumbs.db?!?



  • What's more annoying is a long file copy that almost completes, only to tell you at the very end that "the file name(s) would be too long for the destination folder" followed by a bare file or folder name. No clue whereabouts this is within the folder tree, especially if there are a lot of files or folders with that name, and only a few of them failed to copy. It doesn't look like the MAX_PATH limit is going away any time soon.

    Whoever set that limit deserves some fire and brimstone.


  • Discourse touched me in a no-no place

    @Daniel Beardsmore said:

    It doesn't look like the MAX_PATH limit is going away any time soon.
    Which wouldn't matter too much except it's such a small value on Windows. (The maximum command line length is also quite small, which is occasionally very frustrating.)

    Before you ask, yes, I do have a genuine need for programs with kilobytes of arguments. Alas.



  • It would be nice if, when passing a bunch of files to a Send To shortcut, Windows could chdir() to the enclosing directory and issue local pathnames. Otherwise, you can hit the command-line limit very easily that way. (Though I've not done this for a long time now.)



  • @RTapeLoadingError said:

    @Ronald said:
    @RTapeLoadingError said:
    @mott555 said:
    More than any of those, thumbs.db is what annoys me.

    I've kicked off huge copy jobs on Windows and returned hours later to find it sat there waiting for me to tell it whether or not to replace thumbs.db.

    That's why God created robocopy.

    That's New Testament touchy-feely God and I've always been an Old Testament fire, brimstone and Windows-Explorer-for-file-copy kind of guy. If Windows Exporer was good enough for Moses then it's good enough for me.

    Religion aside, I'm disappointed that the operating system's standard file copy method (Windows Explorer) isn't a bit smarter. Has anyone anywhere ever wanted a multi-GB file copy to stop part way through so that they make a call on Thumbs.db?!?

    You are just like the guy in that joke who refuses to get on the boats during a flood because "god will save him".

    Windows Explorer is an interactive application, if you want to use it for non-interactive operations you are TRWTF. There is a tool in Windows specifically designed for unattended file copy operations and it's called robocopy; if you can't find your way around a command-line STFW there are plenty of robocopy GUIs (one of them is called RichCopy and it has plenty of checkboxes for people who like to click on stuff).



  • @ender said:

    @Quincy5 said:
    It would save me a lot of time if I can just tell them to remove the job from the spooler window and try again, instead of doing this procedure myself because it gets to complicated for them.
    Do they have a HP LaserJet. I don't remember which model exactly, but two clients have the same LaserJet printer, and at both certain jobs would regularly throw the spooler in crash loop. I ended up writing a batch file that cleans the spooler for them:

    @echo off
    if [%1]==[run] goto run
    "%~dp0\elevate.exe" -c -w "%0" run
    goto :eof
    :run
    net stop spooler
    set sysdir=system32
    if not exist "%SystemRoot%\%sysdir%\spool\printers\" set sysdir=sysnative
    del /y "%SystemRoot%\%sysdir%\spool\printers\*"
    net start spooler

     

    They have a HP Deskjet (I don't know the exact type).  I will look into your script; thank you very much.

     



  • @Quincy5 said:

    @ender said:

    @Quincy5 said:
    It would save me a lot of time if I can just tell them to remove the job from the spooler window and try again, instead of doing this procedure myself because it gets to complicated for them.
    Do they have a HP LaserJet. I don't remember which model exactly, but two clients have the same LaserJet printer, and at both certain jobs would regularly throw the spooler in crash loop. I ended up writing a batch file that cleans the spooler for them:

    @echo off
    if [%1]==[run] goto run
    "%~dp0<a href="http://code.kliu.org/misc/elevate/">elevate.exe" -c -w "%0" run
    goto :eof
    :run
    net stop spooler
    set sysdir=system32
    if not exist "%SystemRoot%%sysdir%\spool\printers" set sysdir=sysnative
    del /y "%SystemRoot%%sysdir%\spool\printers*"
    net start spooler

     

    They have a HP Deskjet (I don't know the exact type).  I will look into your script; thank you very much.

     

    Or you could just restart the print spooler service and cancel the job in question in the usual way.



  • @TDWTF123 said:

    Or you could just restart the print spooler service and cancel the job in question in the usual way.

    Not if the bad print job causes Print Spooler to crash on load, you can't. That's something MS never considered.



  • @Quincy5 said:

    @ender said:

    @Quincy5 said:
    It would save me a lot of time if I can just tell them to remove the job from the spooler window and try again, instead of doing this procedure myself because it gets to complicated for them.
    Do they have a HP LaserJet. I don't remember which model exactly, but two clients have the same LaserJet printer, and at both certain jobs would regularly throw the spooler in crash loop. I ended up writing a batch file that cleans the spooler for them:

    @echo off
    if [%1]==[run] goto run
    "%~dp0<a href="http://code.kliu.org/misc/elevate/">elevate.exe" -c -w "%0" run
    goto :eof
    :run
    net stop spooler
    set sysdir=system32
    if not exist "%SystemRoot%%sysdir%\spool\printers" set sysdir=sysnative
    del /y "%SystemRoot%%sysdir%\spool\printers*"
    net start spooler

     

    They have a HP Deskjet (I don't know the exact type).  I will look into your script; thank you very much.

     

    The only thing worse than an HP printer driver is an HP printer driver installer. It will cost you less grief just to buy them a Canon.



  • @Daniel Beardsmore said:

    @TDWTF123 said:
    Or you could just restart the print spooler service and cancel the job in question in the usual way.

    Not if the bad print job causes Print Spooler to crash on load, you can't. That's something MS never considered.

    I'm not sure what you're trying to say there, but yes, you can. The spooler won't try and process the job immediately on load if printing is paused. Shit, I forget the details of all this because it's years since I had to troubleshoot this crap, but I remember seeing the 'solutions' on Google and laughing at the problems people were making for themselves. The real solution's something trivial like pause printing - which you can do even if the print spooler has crashed - restart the service, then cancel the offending print-job.


  • @flabdablet said:

    The only thing worse than an HP printer driver is an HP printer driver installer. It will cost you less grief just to buy them a Canon.
    In my experience, Canons are even worse than HPs.
    @TDWTF123 said:
    I'm not sure what you're trying to say there, but yes, you can.
    No, no you can not. By the time Explorer lists the printers, the spooler has had enough time to start processing jobs and crash again. How do I know? Because I tried to do exactly that - the most I managed was to open the printer's job list, but by then spooler was already dead.
    @TDWTF123 said:
    The real solution's something trivial like pause printing - which you can do even if the print spooler has crashed - restart the service, then cancel the offending print-job.
    And how do you do that when you can't even see the installed printers with the spooler dead (and if you happen to have printer job list open, any command you try will just result in "Error processing request" in the statusbar).


  • Discourse touched me in a no-no place

    @ender said:

    @flabdablet said:
    The only thing worse than an HP printer driver is an HP printer driver installer. It will cost you less grief just to buy them a Canon.
    In my experience, Canons are even worse than HPs.
    Hmm. I've never had problems with Canon printers that wasn't down to a physical issue like a clogged ink reservoir. Were you using a pure printer, or one of the many-things-in-one devices? (I have enough room to have a separate scanner and printer, and allows each to be a good piece of kit instead of being mulched together into a half-assed jack of all trades.)


Log in to reply