Troubleshooting Mac OS X 10.2.5: Date bug really resolved?; FireWire problems; More on USB relation to kernel panics; more

Date bug really resolved? We have begun receiving reports indicating that Mac OS X 10.2.5 may not have resolved the 1969 date reset bug under every operating circumstance. Posts are also starting to appear in our forums and other discussion arenas confirming the same. MacFixIt reader Jim Jagielski writes:

"Mac OS X 10.2.5 was supposed to fix the problem on some systems where the clock was reset to 1969/1970 whenever the system was powered down. My Dual Power Mac G4 MDD 867 was one of those affected. However, since upgrading to Mac OS X 10.2.5, I see no change at all. From the Terminal:

While it is unlikely that Jagielski's battery was drained during natural use, we previously reported that Mac OS X 10.2.4 may have actually acted as a catalyst for battery power decay, meaning that some systems may actually need to have the small PRAM battery replaced.

We are awaiting further information, but please let us know if you are experiencing a similar issue. It should be noted that the overwhelming majority of reports indicate that Mac OS X 10.2.5 fully resolves this issue.

FireWire problems With each incremental update to Mac OS X, it seems at least a handful of users have sudden loss of FireWire functionality. The new 10.2.5 release is no exception. Reports indicating inability to recognize external FireWire hard drives and other devices have started coming in at a faster pace as more users apply the latest update.

Some users have discussed doing a "deep hardware reset" as a possible resolution to the issue. This involves removing a user-installable RAM module (if you have any additional RAM modules separate from the 128 MB required by Mac OS X) and restarting, or by pressing the PMU (power management unit) reset button.

More on USB relation to kernel panics We are continuing to investigate widespread kernel panics occurring after installing Mac OS X 10.2.5, which we originally reported the day after its release. A note from Joachim Schaller adds credence to our previous speculation that the problem is related to the USB I/O system, and offering another suggestion for USB hub users experiencing the problem:

"In Single User Mode, messages appeared on screen containing strings like 'appleusbio' and 'OHCI' timing failure notices. Two status LEDs on my Belkin 7-port USB Hub were off. The LEDs indicate power failures on connected USB devices like my home banking card reader and the Wacom Intuos. The USB failure messages continued until I unplugged the failing USB devices. The problem was gone."

AFP File transfers corrupted We have received corroborated reports of corruption in large AFP file transfers after upgrading to Mac OS X 10.2.5. Carsten Kobusch writes:

"This happens accessing RealBasic-Files on the server or while creating StuffIt archives on the server volume. I just checked the files with BBEdit and the server-side file contains large blocks of zeros or sometimes only a few changed bytes compared to an archive created locally. I reproduced the problem with Helios (old and new) and xServe.

As of now, the only solution for users experiencing this problem seems to be reversion to Mac OS X 10.2.4.

Resources