[UPDATE with Logs] User interface stops responding on Gigabyte GA-P55M-UD2

[fmillion]

First off, here are my specifications.

• Gigabyte GA-P55M-UD2 motherboard, with latest (F11) BIOS
  
• Intel Core i5-750 CPU
  
• nVidia GeForce 7300LE PCI-X graphics
  
• WD 1.5TB Caviar Green SATA Hard drive
  
• 4GB OCZ DDR3 RAM
  

 

I used the [url="http://www.insanelymac.com/forum/topic/279450-why-insanelymac-does-not-support-tonymacx86/"]#####[/url] method to get OSX installed, then manually installed Chameleon and Kexts as follows:

 

• The enabler for the ALC888B audio chip
  
• NVEnabler for the graphics card
  
• SATA Injector
  
• Realtek R1000 driver for Ethernet
  
• A custom DSDT file from these forums for my specific motherboard
  
• A custom smbios.plist and boot plist, making the machine recognized as a Mac Pro and configuring Chameleon.
  

 

(note: I did have some issues with Chameleon and the 1.5tb drive even with RC4, I was able to get by this by using the older EFI partition method for installing Chameleon to the 200MB EFI partition created by Disk utility.)

 

The problem I am having is that, unpredictably and randomly, the entire user interface of the Mac slowly grinds to a halt. It often happens after opening a new window or opening a new app, and most often occurs only after at least two days of uptime.

 

Whenever performig the given action, the beachball appears, and the app in question stops responding completely to any interaction. Other apps still appear to work at this point. If I force-quit the offending app, the Finder "joins the crowd" and it too becomes beachball-hung. Other apps generally will still function for a half a minute or so, and slowly the entire UI grinds to a complete halt and nothing will work at all. The mouse is still active (as either the beachball or the normal pointer) but no other UI commands (force quit, Cmd+Tab, etc.) will respond at all. The only way to get the system back is using the hard reset button on the case.

 

This is not a kernel panic or memory error or something similar, because 1) I have tested the system RAM with Memtest and 2) The core system still seems responsive. For example, if Garageband is open, and I play notes on the MIDI keyboard attached to the computer, the notes will still play, indicating that the software synthesizer is still active and working.

 

After the UI freeze I am no longer able, however, to access the system remotely (for example, SSH or Samba). This means that when this occurs I have absolutely no way to attempt to retrieve any meaningful log information!

 

Without any further details I cannot proceed any further on my own with trying to figure this out, so I'm wondering if anyone has had similar experiences and if possibly some hardware I'm using may be the culprit?

 

Thanks in advance,

 

FM

[fmillion]

Still exeriencing the same problem...I am going to attempt copying the entire 1.5TB drive over to a 500GB drive with CCC, and will run with that as the boot drive for a few days to see if it solves the problem...

 

Also, I am not sure I even need that SATA injector since this is a regular Intel SATA controller which Disl Utility even sees as such, so I might also do away with that to see if it helps..

 

I'm starting to think it may be a SATA issue, as a similar temporary freeze can also occur at times such as opening Disk utility. If Sata is freezing, that would explain why I have no meaningful log data in the logs after a reboot and am unable to do anything - GarageBand might still have samples cached in RAM to explain why it can still function.

 

Comments?

 

FM

[fmillion]

I have replaced the 1.5TB disk with a 500GB disk, and have been running for about 2 days.

 

I had the UI freeze again, but this time it wasn't as drastic, and I was actually able to launch Activity Monitor and Console.

 

I attached the results - I originally tried to launch Terminal, which hardlocked. After that, I tried to force quit it, and as I expected, Finder hardlocked.. but I was still abl eto use the Dock, and launch Activity Monitor and Console. I sampled the Terminal app which seems stuck on "getattr" and also included the most recent entries in the console log.

 

Of particular interest are the following, which I've snipped out of the attached log:

 

10/31/10 11:27:56 PM	com.apple.launchd.peruser.501[140]	(com.apple.Finder[146]) Clean job failed to exit 1 second after receiving SIGKILL.
...
10/31/10 11:30:26 PM	com.apple.launchd.peruser.501[140]	(com.apple.Finder[146]) Job was clean, killed and has not exited after 1 second. Moving on.
10/31/10 11:30:26 PM	com.apple.launchd.peruser.501[140]	(com.apple.Finder[146]) Job was clean, killed and has not exited after 1 second. Moving on.
...
10/31/10 11:30:34 PM	com.apple.launchd.peruser.501[140]	(com.apple.ReportCrash) Falling back to default Mach exception handler. Could not find: com.apple.ReportCrash.Self
10/31/10 11:30:34 PM	com.apple.launchd.peruser.501[140]	(com.apple.Finder[146]) Job was clean, killed and has not exited after 1 second. Moving on.
10/31/10 11:30:34 PM	com.apple.launchd.peruser.501[140]	(com.apple.Finder[146]) Job was clean, killed and has not exited after 1 second. Moving on.
...
10/31/10 11:30:37 PM	diskarbitrationd[16]	coreservicesd [65]:17667 not responding.
...
10/31/10 11:30:42 PM	KernelEventAgent[38]	tid 00000000 received event(s) VQ_LOWDISK, VQ_VERYLOWDISK (516)
...
10/31/10 11:30:46 PM	com.apple.launchd.peruser.501[140]	([0x0-0x23c23c].com.apple.Terminal[5133]) Exit timeout elapsed (20 seconds). Killing
10/31/10 11:30:46 PM	com.apple.launchd.peruser.501[140]	(com.apple.Finder[146]) Job was clean, killed and has not exited after 1 second. Moving on.
...
10/31/10 11:31:03 PM	diskarbitrationd[16]	coreservicesd [65]:17667 not responding.
10/31/10 11:31:06 PM	diskarbitrationd[16]	coreservicesd [65]:17667 not responding.
...

 

Some of this may be expected, but some of it definitely doesn't look good - in particular "coreservicesd not responding".

 

Since still have an active UI, I was able to actually reboot the machine "properly". I'm going to just wait again and see if this occurs again, but this time I was at least able to grab some log data.

 

Any more help?

 

FM

system.log_Sample.rtf

Sample_of_Terminal.txt

[fmillion]

So far no further crashes seem to have occurred, so I'm crossing my fingers that this whole thing was a SATA issue of some sort. It was combined with the loss of the DVD drive on the DATA bus, which appears to be specific to the drive in question. I plan to try tossing OS X onto my AMD box to do some testing there - in a different environment, on a different chipset - to see if it is actually the drive, or an interaction between it and this particular box...

 

Will post here again if more occurs.

 

FM