I do pro bono tech support for a local elementary school computer lab filled with 20 G3 iMacs and 4 eMacs. One of the eMacs recently began throwing intermittent kernel panics.
Specs: 1 GHz, USB 2.0, single 512 MB PC2700 stick. mfgd. March 2005.
DiskWarrior 3.0.3 found no problems. Repaired permissions and ran Disk Utility repair. Open Firmware resets were done (set-defaults, reset-nvram, reset-all). Problem still existed.
Then I booted from the Apple Hardware Test disk. It passed the Quick test, but the Extended test was aborted during the Memory test.
There was no error code. Instead, an Open Firmware window popped up. It said AHT was aborting due to a script execution error in AHT on line 11688. Underneath this was a line that said there were memory leaks.
AHT was rerun, and both Quick and Extended tests were aborted with the same message.
The 512 RAM stick was removed from slot 2 and placed in slot 1. Cuda button was pushed. AHT was rerun and the same aborting script execution message happened on both Quick and Extended.
The 512 MB RAM stick was removed and two known-good 256 MB sticks were placed in the slots. Cuda button was pushed. AHT was rerun and both tests were aborted during the memory test portion. The only difference is that the error occurred on Line 11685 both times. The locations of memory leaks changed too but I didn't write those down.
This AHT is Stand Alone Diagnostics V23.0.
My assumption, based on the above, is that the logic board's memory controller is failing. Or is it something else?
Specs: 1 GHz, USB 2.0, single 512 MB PC2700 stick. mfgd. March 2005.
DiskWarrior 3.0.3 found no problems. Repaired permissions and ran Disk Utility repair. Open Firmware resets were done (set-defaults, reset-nvram, reset-all). Problem still existed.
Then I booted from the Apple Hardware Test disk. It passed the Quick test, but the Extended test was aborted during the Memory test.
There was no error code. Instead, an Open Firmware window popped up. It said AHT was aborting due to a script execution error in AHT on line 11688. Underneath this was a line that said there were memory leaks.
AHT was rerun, and both Quick and Extended tests were aborted with the same message.
The 512 RAM stick was removed from slot 2 and placed in slot 1. Cuda button was pushed. AHT was rerun and the same aborting script execution message happened on both Quick and Extended.
The 512 MB RAM stick was removed and two known-good 256 MB sticks were placed in the slots. Cuda button was pushed. AHT was rerun and both tests were aborted during the memory test portion. The only difference is that the error occurred on Line 11685 both times. The locations of memory leaks changed too but I didn't write those down.
This AHT is Stand Alone Diagnostics V23.0.
My assumption, based on the above, is that the logic board's memory controller is failing. Or is it something else?