Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Microcuts

macrumors member
Original poster
Aug 12, 2009
33
0
Hey all,

Since a few weeks my MBP (Mid 2010 15", Lion) is experiencing kernel panics, at seemingly random times. I've been through all the guides to solve these, but it's really hard to troubleshoot when it's not structurally happening.

Some observations:
- Not tied to the use of a specific application (although I doubt that's even possible with panics).
- OS X sometimes heavily lags shortly before it happens.
- No hardware changes were made before the panics started happening.
- Also happens with no external devices connected.
- It seems to be happening increasingly often.
- Added: There are also random "freezes" (only mouse moving), not being labeled as kernel panics, but forcing me to do a hard shut down.

When the Mac wouldn't start yesterday after a panic, I started to panic myself :). It turned out to be a corrupted directory on the system drive, which was luckily solved using the wonderful DiskWarrior software by rebuilding the directory.

With kernel panics it's usually a hardware problem (correct me if I'm wrong). Does the corrupted drive directory mean it's a problem with the drive, or could it still be anything else? It's a non-standard drive: Crucial C300 SSD with latest firmware. I also have third-party RAM installed (100% compatible), which hasn't given any trouble after installing it a long time ago.

A few reports (also containing system info): 1, 2, 3, 4, 5, 6

What are your recommendations to troubleshoot any further? If more info is needed, ask away! A lot of thanks for all the help!
 
Last edited:
Backup and do a clean install? :(

At least you'll rule out any software/OS related causes.

True. But judging by the sad smiley you know it's not a fun thing to do...
How can it be so hard to find the problem causing a kernel panic :(.

Edit: I could try using safe mode for some time, to see if it would still happen. Would that rule out software related causes too? Because it only uses the default system files which are the same as on a clean install, right?
 
Last edited:
Just had another kernel panic, and this time it was obviously triggered by starting my parallels virtual machine. It was starting at first, but after a few seconds the Windows 7 VM showed the scary blue screen of death, after which the Mac kernel panicked too.

If I am correct Parallels uses many forms of kernel extensions, which could be causing the problems. I am for now deinstalling Parallels while keeping my VM backed up, to see and try if the problems are gone.
 
True. But judging by the sad smiley you know it's not a fun thing to do...

You can use also an USB device instead of your internal HDD.

Edit: I could try using safe mode for some time, to see if it would still happen. Would that rule out software related causes too? Because it only uses the default system files which are the same as on a clean install, right?

Yes.

Your problem is probably the RAM and/or Parallels. Try the Apple Hardware Test (AHT).
 

Hadn't been able to try that so far, because I needed the OS dvd. Done it now, gave me Ram errors: 4MEM/9/40000000. I tried switching/reseating the Ram in the slots, without result. After removing the chips one at the time it's now obvious one of the chips is faulty. Will contact supplier for replacement.

Does this mean it wasn't Parallels' fault? Is it safe to be using it again?
Also, the SSD seems to be fine, so it must have been the RAM corrupting the drive directory.
Thanks for all the advice!
 
I had another crash, with only the good RAM installed. Not with the kernel panic notice overlay, but a freeze with only the mouse still moving. Could this have another cause? Parallels is still uninstalled.
 
Looks like multiple issues now....SSD? RAM. Hard to say for sure. It's looking like a fresh install may be the only way to answer this.

Even the extended AHT doesn't show any errors on my RAM. So I don't think the other module is faulty too. Is there any good software around for checking my SSD? I've already used DiskWarrior which did not find any hardware problems. Also Techtool Pro 5, which scans for bad blocks (is that even relevant for SSD's?) and checks the SMART status of the drive. No problems whatsoever.

There are no logs of this last crash by the way. But this type of crash has never given me any reports, only the ones with the kernel panic overlay. Maybe it has a totally different cause, but how to find it?
 
Still crashes happening. This is what the console logs at crashes:

20-01-12 15:17:29,000 kernel: NVDA(OpenGL): Channel exception! exception type = 0x6 = Fifo: Parse Error
20-01-12 15:17:29,000 kernel: IOVendorGLContext::ReportGPURestart
20-01-12 15:17:29,000 kernel: 00000069
20-01-12 15:17:29,609 sandboxd: ([1242]) DumpGPURestart(1242) deny ipc-posix-shm apple.shm.notification_center
20-01-12 15:17:29,611 sandboxd: ([1242]) DumpGPURestart(1242) deny ipc-posix-shm apple.shm.notification_center
20-01-12 15:17:29,612 sandboxd: ([1242]) DumpGPURestart(1242) deny ipc-posix-shm apple.shm.notification_center
20-01-12 15:17:29,613 sandboxd: ([1242]) DumpGPURestart(1242) deny ipc-posix-shm apple.shm.notification_center
20-01-12 15:17:29,615 sandboxd: ([1242]) DumpGPURestart(1242) deny ipc-posix-shm apple.shm.notification_center
20-01-12 15:17:29,623 sandboxd: ([1242]) DumpGPURestart(1242) deny ipc-posix-shm apple.shm.notification_center
20-01-12 15:17:29,627 sandboxd: ([1242]) DumpGPURestart(1242) deny network-outbound /private/var/run/mDNSResponder
20-01-12 15:17:29,628 sandboxd: ([1242]) DumpGPURestart(1242) deny file-read-data /private/etc/hosts
20-01-12 15:17:29,629 sandboxd: ([1242]) DumpGPURestart(1242) deny file-read-data /private/etc/hosts
20-01-12 15:17:29,639 sandboxd: ([1242]) DumpGPURestart(1242) deny ipc-posix-shm apple.shm.notification_center
20-01-12 15:17:29,641 sandboxd: ([1242]) DumpGPURestart(1242) deny ipc-posix-shm apple.shm.notification_center
20-01-12 15:17:31,814 applepushserviced: <APSCourier: 0x7ff71ad0ecf0>: Stream error occurred for <APSTCPStream: 0x7ff71c1155c0>: The operation couldn’t be completed. Connection reset by peer
20-01-12 15:17:33,000 kernel: NVDA(Video): Channel exception! exception type = 0x1a = Fifo: FBISTATE Timeout Error
20-01-12 15:17:49,000 kernel: NVDA(OpenGL): Channel timeout!
20-01-12 15:19:22,885 [0x0-0x56056].com.apple.iTunes: The plugin 'Quartz Composer Visualizer' failed to load because it has the wrong CPU architecture for this version of iTunes. (3585)
20-01-12 15:19:23,992 com.apple.usbmuxd: HandleUSBMuxDictionary client 0x101c004c0-iTunes/com.apple.iTunes using library usbmuxd-263, running usbmuxd-265.1
20-01-12 15:19:30,366 Cloud: handleRequest request: An exception was caught while attempting to create a return message, and will be squashed: [NSMachPort sendBeforeDate:] destination port invalid
20-01-12 15:19:35,737 [0x0-0x58058].com.apple.iTunes: The plugin 'Quartz Composer Visualizer' failed to load because it has the wrong CPU architecture for this version of iTunes. (3585)
20-01-12 15:19:36,845 com.apple.usbmuxd: HandleUSBMuxDictionary client 0x10180bb60-iTunes/com.apple.iTunes using library usbmuxd-263, running usbmuxd-265.1
20-01-12 15:19:46,155 login: USER_PROCESS: 1262 ttys000
20-01-12 15:19:48,834 login: DEAD_PROCESS: 1262 ttys000

Seems to have something to do with the GPU then, doesn't it? I do have the mid-2010 macbook pro which was said to have trouble under Lion (1, 2, updates are installed), but I never had any problems before.

I'm not having the issues the support document describes, so I'm not sure if this is it...
 
Graphics for sure, judging by the report opengl is the issue. If it were me I think I'd do a fresh install....It's not going to go away on it's own I'm afraid.

Will do so asap. If I restart from the recovery partition, can it do a fully clean install, or will it leave files around? I can also format the system partition first from the recovery, right?
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.