Benches macOS

14

Predictable pattern of kernel panics - can you ...

apple Verified

Imported from the Apple Support Forums via the Wayback Machine 🔗

Hi. Long time listener, first time caller.

I installed 10.5 on my PB as soon as it became available and have had an issue with kernel panics ever since. The panics arrive in waves of ~12-15 at a time over the two to three days (~36-48 hours of uptime) immediately subsequent to my having booted from a volume other than the 10.5.2 system on my PB's internal HD. When the panics have run their course, everything returns to normal, and no problems of any kind arise until after I've temporarily booted into another volume.

I have experienced these waves of kps after booting from a 10.5.2 volume on an external drive; two different 10.4.11 volumes on two different external drives; the 10.5 retail install disc; a brand new DiskWarrior 4.1 CD, and the AHT volume on the (replacement) Install and Restore disc that came with my computer. (None of the utilities I've used to attempt diagnose my issue have returned any errors or warnings about the state of my hardware [incl. RAM and AEBS] or my main boot volume.)

I've attached the latest panic report below. It is identical to all of my PanicReporter entries for the past 6 months. I can see that this may be AirPort (driver) related, but if you can offer a more incisive reading of the report than that, I'm definitely interested in what you have to say.

Thu May 22 12:32:18 2008

Unresolved kernel trap(cpu 0): 0x300 - Data access DAR=0x000000000000000C PC=0x00000000002A9684
Latest crash info for cpu 0:
Exception state (sv=0x249d8c80)
PC=0x002A9684; MSR=0x00009030; DAR=0x0000000C; DSISR=0x40000000; LR=0x3014576C; R1=0x1E573C00; XCP=0x0000000C (0x300 - Data access)
Backtrace:
0x301456A0 0x30155BB0 0x300F5EC4 0x00356AC8 0x00355D2C 0x00355DE0
0x000B05D4
Kernel loadable modules in backtrace (with dependencies):
com.apple.driver.AirPortBrcm43xx(312.46.51)@0x300ec000->0x30218fff
dependency: com.apple.iokit.IO80211Family(211.1)@0x2498b000
dependency: com.apple.iokit.IOPCIFamily(2.4.1)@0x1e5f8000
dependency: com.apple.iokit.IONetworkingFamily(1.6.0)@0x24482000
Proceeding back via exception chain:
Exception state (sv=0x249d8c80)
previously dumped as "Latest" state. skipping...
Exception state (sv=0x2469a780)
PC=0x00000000; MSR=0x0000D030; DAR=0x00000000; DSISR=0x00000000; LR=0x00000000; R1=0x00000000; XCP=0x00000000 (Unknown)

BSD process name corresponding to current thread: kernel_task

Mac OS version:
9C7010

Kernel version:
Darwin Kernel Version 9.2.2: Tue Mar 4 21:23:43 PST 2008; root:xnu-1228.4.31~1/RELEASE_PPC
System model name: PowerBook5,1
panic(cpu 0 caller 0xFFFF0003): 0x300 - Data access
Latest stack backtrace for cpu 0:
Backtrace:
0x0009B478 0x0009BE1C 0x00029DD4 0x000AF210 0x000B2A78
Proceeding back via exception chain:
Exception state (sv=0x249d8c80)
PC=0x002A9684; MSR=0x00009030; DAR=0x0000000C; DSISR=0x40000000; LR=0x3014576C; R1=0x1E573C00; XCP=0x0000000C (0x300 - Data access)
Backtrace:
0x301456A0 0x30155BB0 0x300F5EC4 0x00356AC8 0x00355D2C 0x00355DE0
0x000B05D4
Kernel loadable modules in backtrace (with dependencies):
com.apple.driver.AirPortBrcm43xx(312.46.51)@0x300ec000->0x30218fff
dependency: com.apple.iokit.IO80211Family(211.1)@0x2498b000
dependency: com.apple.iokit.IOPCIFamily(2.4.1)@0x1e5f8000
dependency: com.apple.iokit.IONetworkingFamily(1.6.0)@0x24482000
Exception state (sv=0x2469a780)
PC=0x00000000; MSR=0x0000D030; DAR=0x00000000; DSISR=0x00000000; LR=0x00000000; R1=0x00000000; XCP=0x00000000 (Unknown)

BSD process name corresponding to current threa`

Thank you.

14 replies

Sign in or join Beachball Help to reply.

  1. 0

    Apple Support Community member

    Imported from the Apple Support Forums via the Wayback Machine 🔗

    Although not always the most direct path, I have always been a firm believer
    In the Chinese Box Theorem in trying to zero in on a solution to a problem.
    Simply put, you can always find the “Lion in the Desert” by continually dividing
    the desert in half, deciding which half the lion is in, dividing that half in half, and so on. Have you tried physically changing, turning off, disabling or isolating anything in your pB?

    Being an L-4, is it safe to assume you’ve run this by the Lounge?
    If so, please relate to us what can or can’t be ruled out in trying to narrow down the root cause?

    I’ve also realized over the years that the most difficult problems to isolate are those with more than one possible cause, such as older Airport cards, drivers, and indexing. For example, is it possible your kps stop occurring when Spotlight is done crunching?

    As good as a bump anyway Tuttle, good luck.

    Joe

  2. 0

    Apple Support Community member

    Imported from the Apple Support Forums via the Wayback Machine 🔗

    Hi, Joe. Good thoughts, as always.

    Being an L-4, is it safe to assume you’ve run this by the Lounge?

    Actually, no. As you know, there are many knowledgeable members at every reputation level here, so it is in my best interest not to limit the possibility of my receiving a useful answer from the larger pool by depending on such a limited, somewhat arbitrarily selected group--many of whom I like very much--whose skills mainly reside outside of my immediate needs.

    ... the most difficult problems to isolate are those with more than one possible cause, such as older Airport cards, drivers, and indexing. For example, is it possible your kps stop occurring when Spotlight is done crunching?

    Agreed.

    I have no reason to believe that Spotlight is involved as all external devices are excepted from indexing, so no indexing is occurring when I boot back into my main volume. AirPort card may be a possibility, but as this box is two years out of (extended) warranty, and a new MBP is coming my way at the next revision, I'd like to rule out software/driver causes before I drop any dough on hardware tests/repairs. In any event, kps occur whether or not I'm online and whether AirPort is on or off.

    Have you tried physically changing, turning off, disabling or isolating anything in your pB?

    Yes, everything I can think of, which is why I'm here.... ; )

    (Thanks for the bump if not the fix.)

  3. 0

    Apple Support Community member

    Imported from the Apple Support Forums via the Wayback Machine 🔗

    Tuttle,
    1. Do you have any third party wireless routers?
    2. Is your Powerbook at least 867 Mhz?
    3. Is all your RAM to spec and not exceeding spec?
    4. Have you looked over my FAQ*:

    http://www.macmaps.com/kernelpanic.html

    - * Links to my pages may give me compensation.

  4. 0

    Apple Support Community member

    Imported from the Apple Support Forums via the Wayback Machine 🔗

    a brody - I have explicitly asked for help in interpreting the panic report I posted above. If you are unable to answer my specific question, then please move on to other topics containing issues you are qualified to help with.

    Additionally, you have asked three questions of me which have already been answered in my posts above and in my profile below.

    As to your fourth question, I'm not interested in reading your (or Dr. Smoke's) FAQs (again) as they do not provide any guidance beyond the data points I've already collected (and shared) based on my own considerable experience with OS X and Mac hardware.

    Thanks anyway, and good luck to you!

  5. 0

    Apple Support Community member

    Imported from the Apple Support Forums via the Wayback Machine 🔗

    The panic clearly states that it has to do with airport.
    If you bothered to read my FAQ, you'd also learn that panics can inaccurately report what the actual problem is, especially if other hardware issues are there.
    If you bothered to read my FAQ, you'd learn some driver issues may happen because of directory failure.
    My FAQ also directs you how to backup your data, before testing for directory issues, and how to run a hardware test to test for the immediate possible RAM issues. You still have not told me if any third party routers are available, or answered any of my other questions directly.
    Please don't be saucy with me. I'm trying to help. If you can't see it, I'm sorry, but I can't spell out everything over and over and over and over and over again. I'd be happy to clarify any point on the FAQ though if you don't understand it.

    Message was edited by: a brody

  6. 0

    Apple Support Community member

    Imported from the Apple Support Forums via the Wayback Machine 🔗

    I'd be happy to clarify any point on the FAQ though if you don't understand it.

    Oh, please. :-

  7. 0

    Apple Support Community member

    Imported from the Apple Support Forums via the Wayback Machine 🔗

    I'm serious Tuttle.

  8. 0

    Apple Support Community member

    Imported from the Apple Support Forums via the Wayback Machine 🔗

    .

    Message was edited by: Tuttle

  9. 0

    Apple Support Community member

    Imported from the Apple Support Forums via the Wayback Machine 🔗

    How serious are you about wanting help?

    Kernel panics can happen because of varying causes. Logfiles while helpful are often not the final telltale reason of why kernel panics happen.

    If the kernel is somehow seeing an issue it can't resolve, it will panic at the state that it is in, while reading from it, which is not necessarily the actual place where a problem exists. I.e. if there is memory that is bad, and kernel needs more memory to cache information over airport, then it is entirely possible for the logfile to say that airport caused the problem, when the memory is at fault.

    If the directory is not well kept, the fact that the directory and the boot block are closely tied together could lead to kernel panic, which is resolved when a new directory is built.

    The fact that hardware tests can miss bad RAM when it happens means we have to eliminate potential causes in an orderly manner:

    1. Backup your data, in case some troubleshooting step causes more harm than good.

    2. Verify all drivers are up to date for the system, were not transfered over from other incompatible systems via Target Disk Mode or Migration Assistant.

    3. If all are up to date for the system, the hard drive isn't too full, verify and repair the directory.

    4. If the directory appears to be OK, eliminate all possible causes for hardware issues one by one, third party first with firmware that needs to be compatible, and then Apple's.

    5. If none of the above appear to be a problem, try erase and installing the operating system. If kernel panics persist with just Apple software and hardware, see if it is when a specific hardware is accessed or not.

    6. If nothing seems to trigger it specifically, RAM is the most likely cause.

  10. 0

    Apple Support Community member

    Imported from the Apple Support Forums via the Wayback Machine 🔗

    Because this post contains information that may be useful to others later on, I will not ask that it be removed as I have some of your previous posts in this topic.

    Notes:

    Panics do indeed occur in various situations, but as you suggest, I cannot at every moment account for the status of background processes. As I've said, however, the data in the panic reports I have amassed over the past six months are (virtually) identical. And as I pointed out more than once, these spates of panics occur in predictable, repeatable cycles.

    AHT in Extended, loop mode over 24 hours has not detected a RAM issue, though I am aware that this does not necessarily rule a RAM issue out.

    As mentioned above, neither DU's Disk Repair nor DiskWarrior have solved this issue.

    No data was transferred to this present installation via TDM or Migration Assistant. It was an E&I onto a freshly formatted volume. There is very little third-party software installed (BBEdit; CSSEdit; an FTP client and various browsers). Other major third-party software exists on 10.4.11 volumes because I am too cheap to upgrade many of these products when older versions still work well in the previous OS.

    There is no third-party hardware attached to this machine beyond a variety of HDs (all OWC; Hitachi), and there is no firmware to update for any other peripherals being used, however sporadically.

    This problem has presented itself through a variety of installation options, including a simple upgrade from 10.4.10; an A&I over a 10.4.10 volume; E&Is both using and not using Migration Assistant (with A&Is following both), and cloning an E&I on an external volume back to a newly formatted volume on my PB.

    Could be the RAM, but the symptoms (and the log files) tell me that more likely it's an I/O issue having something to do with AirPort, which brings you and I full circle, I guess.

    Thanks again!

  11. 0

    Apple Support Community member

    Imported from the Apple Support Forums via the Wayback Machine 🔗

    Tuttle, did you see this very lengthy thread in the Developer Forum? There are others as well, including a discussion at the Azureus web site:

    http://discussions.apple.com/thread.jspa?messageID=5716769

    http://forum.vuze.com/message.jspa?messageID=38328

    At some even longer threads involving KPs and Airport drivers in Tiger:

    http://episteme.arstechnica.com/eve/forums/a/tpc/f/8300945231/m/411008906831

    That one includes several links to other discussions, including 10 pages of reports here in AD.
    Francine


    Francine
    Schwieder

  12. 0

    Apple Support Community member

    Imported from the Apple Support Forums via the Wayback Machine 🔗

    Thanks, Francine.

    ... did you see this very lengthy thread in the Developer Forum?

    I'm culling through those posts now.

    A few things: Everyone who's posted about this here and at ars seems to be using the Atheros chipset which is much newer than I have. And, while I use Azureus, it's not installed on the volume at issue, and I haven't had a kp using my FTP client (Captain FTP). I also see that the reports posted in that topic all contain the line,

    panic(cpu 0 caller 0x002C0FE4): "m_free: freeing an already freed mbuf"@/SourceCache/xnu/xnu-1228.0.2/bsd/kern/uipc_mbuf.c:3574

    or similar, while mine shows,

    panic(cpu 0 caller 0xFFFF0003): 0x300 - Data access

    which may actually be a RAM issue, right?

    WRT the ars topic, I see this:

    Pulling the IO80211Family kext from 10.4.9 will fix it*; remember to chown it to root:wheel if you do.

    *disclaimer: this is definitely not any sort of supported configuration, and may do other bad things. It's worked fine for me though.

    but I have had the same problem from 10.5-10.5.2, so there probably isn't anywhere from which to pull a usable kext...

    Now I'm off to read through http://discussions.apple.com/thread.jspa?threadID=986785&tstart=0

    I appreciate your help...

  13. 0

    Apple Support Community member

    Imported from the Apple Support Forums via the Wayback Machine 🔗

    You know, one thing I wonder is if, when you boot from the external drive, some setting gets re-set in PRAM (I'm a bit foggy about exactly what things are kept in PRAM, but I know some settings get carried over, for lack of a better word, that are made when you are in one system, then when you boot into another system that setting is still there). And if the thing that got reset keeps causing a kernel panic until it finally gets set properly for Leopard on the internal drive. Yeah, I know, that is a pretty vague theory, but it would be easy to test: after having booted from your external drive, try resetting the PRAM before rebooting to the internal drive and see what happens. Your problem is very strange and I didn't see anything that really looked the same--all I took away from the various discussions of Airport and kernel panics is that the driver is evidently not all that great, reliablity-wise.
    Francine


    Francine
    Schwieder

  14. 0

    Apple Support Community member

    Imported from the Apple Support Forums via the Wayback Machine 🔗

    ... some settings get carried over, for lack of a better word, that are made when you are in one system, then when you boot into another system that setting is still there ... try resetting the PRAM before rebooting to the internal drive and see what happens....

    LOL - That's a thought I had, too, and I will try the PRAM reset after I boot into a 10.4.11 volume to use Photoshop 7 tomorrow fully realizing that if it doesn't work, I'll have to spend the entirety of my Memorial Day weekend recovering from kernel panics.... ; )