need help troubleshooting kernel panics on a PBG4

[ChrisBrightwell]

I have a friend w/ a PowerBook G4 (15", 1.5 GHz, no multi-touch trackpad) that's getting lots of kernel panics. The Genius Bar diagnosed it as a bad HDD, so I swapped the HDD for them. It didn't solve the problem.

I've been trying to grok the panic log files, but it makes no sense to me. I'm hoping that someone here can help me out.

The log files are here:
http://dimwell.net/temp/panic_logs/

In my experience, most kernel panics are due to bad hardware or bad drivers. I don't have the machine in my possession, so I can't do a lot of troubleshooting right away. I copied these log files, hoping I'd be able to figure something out from there.

Any help or other guidance would be hugely appreciated. Thanks.

EDIT: I'll be picking up the machine on Tuesday, then running a full scan of TechTool (via an AppleCare disc) on it. I don't have any spare RAM to swap in or out, but I'll try to systematically eliminate that as a culprit, as well.

 

[cluthz]

Seems like all the logs have this error:

panic(cpu 0 caller 0x0024CA10): "hfs_swap_HFSPlusBTInternalNode: unrecognized catalog record type (0x0042; record #23)\n"@/SourceCache/xnu/xnu-1228.12.14/bsd/hfs/hfs_endian.c:702

(I haven't read all, but most of them)

Which kinda says it's a HD related problem.
When you changed the HD, did you clone the old one?

 

[ChrisBrightwell]

When you changed the HD, did you clone the old one?

I did. I was planning to run fsck and AppleJack on the filesystem.

 

[ChrisBrightwell]

Picked up the PBG4 yesterday, finally ran the included Apple Hardware Test tonight. The result?

***ERROR*CODE***ERROR*CODE***
2MEM/4/4:SODIMM1/J25UPPER

***ERROR*CODE***ERROR*CODE***

My next course of action is to remove the upper DIMM and re-run the test. It's a Crucial DIMM, so it should have a lifetime warranty. Any other suggestions?

EDIT: I pulled the third-party RAM (Kingston, not Crucial) and the machine wouldn't boot. I swapped the Kingston RAM into the slot previously occupied by the Apple-branded RAM and the machine booted. Re-ran the AHT and had no errors. I'm going to try the same test with the Kingston in the other slot, then again with both RAM modules installed (but in swapped locations relative to their original positions).