HELP - Kernel Panic - Memory related?

Posted:
in Genius Bar edited January 2014
So I've had a handful of kernel panics since I bought my new Mac Pro just over a month ago. They all are identical:



Wed Mar 7 09:33:21 2007

panic(cpu 0 caller 0x00776CD0): Uncorrectable Fbd memory error detected. ferr = 30000800 , nerr == 00000000

....(more)



Anyhow, I've gone online, and have tried a number of checks including Apple Hardware Test (which as I've read isn't foolproof in finding bad memory). I've run memtest both via the GUI and in single-user mode (via applejack). I started to think that maybe I didn't have bad memory and removed some extensions (Logitech driver - so I'm just using raw USB support, and I've removed my PowerMate). I'm now running in SafeBoot mode, but I can only handle that for so long since it prevents me from VPNing into my office. I can't reliably cause the problem. I've called AppleCare, but all they did was make me run through all the same tests I had done on my own, and of course, no kernel panic. They then basically said just leave my MacPro in SafeBoot mode until it panics and then bring it in to my local Apple store.



This is getting ridiculous; these random un-reproduceable panics are getting old. I've spent a major chunk of change on this box.



So now I'm down to suspecting BOINC (for running things like seti@home, rossetta@home) as the culprit; then again, maybe it isn't the culprit, but rather one of the few programs that push my system hard enough to make this elusive memory problem rear it's ugly head?



Does anyone have any suggestions/help?



Thanks for reading, hopefully someone out there can help me...

Comments

  • Reply 1 of 9
    rob05aurob05au Posts: 348member
    Do you have any Unsanity Haxies installed at all?



    These can at times causes problems.



    Also can you post some more of the Panic log it may show something that could lead to the cure.
  • Reply 2 of 9
    I shouldn't have any haxies as this is a brand new machine, and I did not do the "port another mac to this one". Here are two complete log entries, they initiallly all looked the same, but I just noticed the last couple panics are slightly different in their wording from the first couple, so here is one of each entry that I've received:



    Sun Feb 25 20:14:47 2007

    panic(cpu 0 caller 0x3D71ACD0): Uncorrectable Fbd memory error detected. ferr = 30000800 , nerr == 00000000



    Backtrace, Format - Frame : Return Address (4 potential args on stack)

    0x108cd8 : 0x128d1f (0x3c9540 0x108cfc 0x131df4 0x0)

    0x108d18 : 0x3d71acd0 (0x3d71ce64 0x30000800 0x0 0x8100a4)

    0x108da8 : 0x3d71af64 (0x43a4f00 0x43a4f00 0x2 0x820044)

    0x108e38 : 0x35b4f712 (0x43a4f00 0x0 0x4295e00 0x0)

    0x108e68 : 0x35b4f59b (0x4262700 0x0 0x0 0x1c)

    0x108e88 : 0x35b6231c (0x1c 0x108ef8 0x8 0x35b63dd9)

    0x108eb8 : 0x35b624c6 (0x425f11c 0x17 0x4262e4a 0x4179d20)

    0x108f18 : 0x35b6f05d (0x425c148 0x0 0x412ff058 0xd89)

    0x108f38 : 0x35d1a4fb (0x425c148 0x0 0x4056c80 0x0)

    0x108f68 : 0x35b557c5 (0x4142000 0x0 0x4227680 0x49)

    0x108f88 : 0x35b4ec1f (0x40cfc00 0x0 0x4227680 0x49)

    0x108fa8 : 0x3b9d21 (0x4237600 0x0 0x4237880 0x49)

    0x108fe8 : 0x19aa20 (0x4edeee0 0x4edeee0 0x19ba76 0x57c000) No mapping exists for frame pointer

    Backtrace terminated-invalid frame pointer 0xb007ff18

    Kernel loadable modules in backtrace (with dependencies):

    com.apple.driver.AppleMCEDriver(1.1f1)@0x3d717000

    dependency: com.apple.iokit.IOACPIFamily(1.2.0)@0x35afa000

    com.apple.driver.AppleAPIC(1.2.0)@0x35d19000

    com.apple.driver.AppleACPIPlatform(1.0.5)@0x35b4d0 00

    dependency: com.apple.iokit.IOPCIFamily(2.1)@0x35ae4000

    dependency: com.apple.iokit.IOACPIFamily(1.2.0)@0x35afa000



    Kernel version:

    Darwin Kernel Version 8.8.1: Mon Sep 25 19:42:00 PDT 2006; root:xnu-792.13.8.obj~1/RELEASE_I386





    *********



    Wed Mar 7 09:07:39 2007

    panic(cpu 0 caller 0x3D71CCD0): Uncorrectable Fbd memory error detected. ferr = 30000800 , nerr == 00000000



    Backtrace, Format - Frame : Return Address (4 potential args on stack)

    0x108cd8 : 0x128d1f (0x3c9540 0x108cfc 0x131df4 0x0)

    0x108d18 : 0x3d71ccd0 (0x3d71ee64 0x30000800 0x0 0x8100a4)

    0x108da8 : 0x3d71cf64 (0x43b2c80 0x43b2c80 0x2 0x820044)

    0x108e38 : 0x35b51712 (0x43b2c80 0x0 0x4297e00 0x0)

    0x108e68 : 0x35b5159b (0x4264700 0x0 0x0 0x1c)

    0x108e88 : 0x35b6431c (0x1c 0x108ef8 0x8 0x35b65dd9)

    0x108eb8 : 0x35b644c6 (0x426111c 0x17 0x42390ca 0x417bd20)

    0x108f18 : 0x35b7105d (0x425e148 0x0 0x1129cabb 0xcd6)

    0x108f38 : 0x35d1c4fb (0x425e148 0x0 0x4058c80 0x0)

    0x108f68 : 0x35b577c5 (0x4144000 0x0 0x4229680 0x49)

    0x108f88 : 0x35b50c1f (0x40d1c00 0x0 0x4229680 0x49)

    0x108fa8 : 0x3b9d21 (0x4239600 0x0 0x4239880 0x49)

    0x108fe8 : 0x19aa20 (0x258e3e30 0x258e3e30 0x19ba76 0x57c000)

    0x258e3f18 : 0x1a42f5 (0x18 0x1 0x258e3f38 0x6)

    0x258e3f38 : 0x19d871 (0x0 0x0 0x258e3f68 0x206)

    0x258e3f58 : 0x135f23 (0x0 0x0 0x0 0x19d23a) \tBacktrace continues...

    Kernel loadable modules in backtrace (with dependencies):

    com.apple.driver.AppleMCEDriver(1.1f1)@0x3d719000

    dependency: com.apple.iokit.IOACPIFamily(1.2.0)@0x35afc000

    com.apple.driver.AppleAPIC(1.2.0)@0x35d1b000

    com.apple.driver.AppleACPIPlatform(1.0.5)@0x35b4f0 00

    dependency: com.apple.iokit.IOPCIFamily(2.1)@0x35ae6000

    dependency: com.apple.iokit.IOACPIFamily(1.2.0)@0x35afc000



    Kernel version:

    Darwin Kernel Version 8.8.1: Mon Sep 25 19:42:00 PDT 2006; root:xnu-792.13.8.obj~1/RELEASE_I386
  • Reply 3 of 9
    rob05aurob05au Posts: 348member
    This may sound crazy but a mate has had the same thing happen and it was traced back to the latest firmware update not installing correctly for some unknown reason.



    Try removing any extra memory added and do a test without it.



    Also try re-running the latest firmware update and you will more than likely need to do a clean install as this seems to affect some apple supplied drivers which are unable to load correctly.



    Also remove all third party external devices as these can lead to strange problems occuring.



    You could also try re-running the 10.4.9 combo updater which I have always used as it seeems to fix more problems than it causes.
  • Reply 4 of 9
    fuzz_ballfuzz_ball Posts: 390member
    Okay, I've tried a number of things (including the suggestions above, thank you) and I finally came down to one last thing (short of a complete wipe/reinstall): Remove Apple's extensions. I never really thought that Apple's extensions could be the culprit, but now I'm 99.9% sure which one it is: The new Airport Disk extension (for those that have the new airport extreme and have attached an external hard drive to the extreme).



    Since my problem has started, a surefire way to cause the kernel panic to raise its ugly head is to run BOINC and memtest at the same time and it's literally a matter of minutes before my Mac Pro would panic. NOTE: I have had the crash when neither of those programs are running, so they have long since been removed as suspects.



    So I removed this new Apple extension, and low and behold, while running BOINC, memtest, AND playing a game (because I got bored waiting); my system ran for an HOUR and I never got a panic.



    As an aside, I've had numerous problems with the new Airport Extreme (as have a number of others, just to to Apple's discussion boards); so in retrospect I'm not surprised this Apple extension is the cause of my panics. If anything changes (like my Mac Pro starts panicing again sans the airport disk extension) I'll update this post. Otherwise, you might want to think twice about running that airport disk extension; I've been lucky that I've never lost any work on my panics, but I'm sure it was a matter of time.
  • Reply 5 of 9
    setiensetien Posts: 1member
    Quote:
    Originally Posted by fuzz_ball View Post


    Okay, I've tried a number of things (including the suggestions above, thank you) and I finally came down to one last thing (short of a complete wipe/reinstall): Remove Apple's extensions. I never really thought that Apple's extensions could be the culprit, but now I'm 99.9% sure which one it is: The new Airport Disk extension (for those that have the new airport extreme and have attached an external hard drive to the extreme).



    I don't have the airport extreme, and I can't seem to find an extension that relates to it.

    What is the name of the extension? Would it be loaded even if I don't have the airport extreme?



    I also get uncorrectable fbd errors with a panic log that looks very much like yours, and I get nothing in the apple hardware test.



    I'll try the tip with the combo updater. I'll also try running boinc to see if I can push my machine to generate the error.
  • Reply 6 of 9
    fuzz_ballfuzz_ball Posts: 390member
    Quote:
    Originally Posted by setien View Post


    I don't have the airport extreme, and I can't seem to find an extension that relates to it.

    What is the name of the extension? Would it be loaded even if I don't have the airport extreme?



    I also get uncorrectable fbd errors with a panic log that looks very much like yours, and I get nothing in the apple hardware test.



    I'll try the tip with the combo updater. I'll also try running boinc to see if I can push my machine to generate the error.



    Launch the Activity application (in your Utilities folder). Click on the sort by name column and then look for "Airport Disk Agent". Select it, and choose "Quit". You should not need to do "Force Quit" on it, as the regular quit seems to work without issue. Then try running all the programs that you typically run when the panic happens and see if your problem goes away. Of course, if you don't have the Airport Disk Agent (ADA), then your problem is different than mine.



    Everytime that little bastard is running, and if I run some application that pushes my system I get a kernel panic every damn time! I assume it got installed with one of the more recent airport updates, I'm not really sure when it appeared, I just know that it causes all my kernel panics.



    I have not found a way to prevent that extension from running, so on the rare occasion I have to reboot my system, the first thing I do is kill it from the Activity application.



    Hope this helps!



    PS - The recent airport update did NOT fix the problem, just delayed it. Rather than my system panicing within 5 minutes with ADA running, it takes somewhere in the neighborhood of 6 to 12 hours, which obviously is still totally unacceptable. Without ADA running though, I go for weeks without ANY problems.
  • Reply 7 of 9
    fuzz_ballfuzz_ball Posts: 390member
    Quote:
    Originally Posted by setien View Post


    I get nothing in the apple hardware test.



    One more thing, I've heard that sometimes you can have a memory problem that apple hardware test won't catch. Go download Applejack (and make sure you choose the download that includes memtest; if no such download exists, then download Rember, which is a GUI front end to memtest and includes memtest, then install Applejack).



    You'll need to read Applejack's readme, because (at least in the version I used) the option to run a full memory test from Applejack is hidden. You have to make a certain key entry to expose the "beta menu" so you can run memtest.



    Oh yeah, you actually run Applejack (and consequently memtest) in single-user mode (means you reboot your system and hold a certain key combination and you end up with a UNIX like prompt instead of your typical friendly OS X interface).



    Try this out too just to make sure you aren't actually suffering from a bad memory issue.
  • Reply 8 of 9
    rob05aurob05au Posts: 348member
    Did this start to happen before or after the 10.4.9 update. The 10.4.9 update has a nasty pre-binding bug which can zero out some important system files and cause no rebooting is the main problem. But I have seen others where memory errors occur due to some other system files either zeroed out or become corrupted.
  • Reply 9 of 9
    fuzz_ballfuzz_ball Posts: 390member
    The problem started before the 10.4.9 release. I state that only because according to my quick Google search, 10.4.9 was release March 13th (or thereabouts). My problem started almost immediately after I purchased my new Mac Pro. My first recorded kernel panic was February 24th.
Sign In or Register to comment.