RAM causes kernel panics
 

I have one of the latest PB 15" running the G4, the one with HD display.

It comes with 512MB RAM by Hynix.
I added 512MB from Corsair (http://www.corsair.com/corsair/value...tml#ddr2sodimm PC2-4200) and worked fine for more than 2 months.
Then I started to experience kernel panics.
I ran the memtest and crashed a couple of times, which confirmed the problem. I removed the Hynix memory and worked like a charm for about a month now.

Now I purchased 2 new SO-DIMM's from a Corsair reselles (the same as the original 512MB DIMM) that are 1GB each. Same as the 512MB.
I installed them last night and today it has already panic'd twice!

This sounds a lot like it's not the RAM (maybe not even the original Hynix), but maybe the PB mainboard. Maybe the RAM SLOT.

Any other idea or suspect? I will run memory tests tonight, but I am pretty sure it's the memory. System Information recognizes 2GB of memory.

I'm not saying this is your issues but you might want to take at look at this Apple Doc.

Thank you for your reply. I am certainly not elegible for this replacement as my computer was purcharsed in October and was just released (so I believe manufactured AFTER April 2005) and also the serial number is higher W854XXX.
Also while I fear the problem is similar (as it seems to happen when I have both slots filled) the symptoms are different. The computer recognizes the 2GB, performances are GREAT, it just freezes! :eek:

Another update, so that maybe if someone else has the same problem knows.

I searched the web and found this:
http://lowermemoryslot.editkid.com/

Apparently MANY people all over the world has similar problems!

Also there a guy who says that the problem appeared after upgrading to 10.4.5. Guess what? I can't remember the exact dates (as it was about a month ago) but it seems like it was about the same time the problem was born.
Last night I ran memtest on a 1GB SO-DIMM, ALL NIGHT (10pm-9am) all ran smoothly, it completed about 10 full tests.
This morning it froze again after less than 2 hours.

24 hours with 512MB Corsair Value Select DDR2 without a problem.

Now I reinstalled 1GB and ran hardware test 4 times (extended test) all successfull.

I will let it run all night and see if tomorrow morning is crashed.

Update
 

I took my PB to an Apple center to get it repaired.
I also brought all my memories, 1 original from hynix (512MB), another 1 from Corsair (512MB) and 2 1GB Corsair memories.

They called me about 1 week later to ask more specific description of my problem. I described it and the guy told me that he has already encountered similar problems with similar powerbooks. He would change my logic board, but did not know if it would effectively solve the problem. Apparently sometimes it was solved, some other times it wasn't.

I got it back in the time they told me, about 2 weeks.

I restored all my backups now and after less than 4 hours of use it already crashed twice.

I am going to call again tomorrow, but this is not funny.

You should check the RAM with 'memtest':
http://www.memtestosx.org/
It is reputed to do a much more thorough test than the Apple Hardware test disk.

I ran memtest on every single DDR I have all successful.
The apple support center did the same on another powerbook.

All memories work fine. The problem is NOT present if I have only 512MB! If I install one of the 1GB memories crashes. If I install 2x512MB crashes. Each of the two DDR of 512MB work fine when alone.

Did you check a single RAM module in both the upper and the lower slot ?

Did the Apple Center really swap the MoBo ?

One of the slots causing problems is a well-known problem.
The inability to use both slots simultaneously is not.

I think I did, but I have to admit it was more than a month ago. I am SURE I tested the Corsair 512MB on each slot and I remember memtest crashing with the 2 512MB RAM's.

I can't swear on anything else other than "I tested them all".

How do I check?

I can tell you that 1 512MB RAM works smoothly on each single slot. I tested them using the computer at work for 1-2 weeks without ever stopping it and using it 8+ hours per day.

As I think I wrote, I also tried a single 1GB RAM in the same slot that worked for DAYS with a 512MB.

Honestly I also started to suspect that it's a problem in the management of memory above 512MB, but I don't know how to confirm this. Also, it's not an application crashing, the system freezes. Sometimes just "stops" some others turns gray and asks me to press the start button.

Very disappointing. :(

Not that I'm too cheap to buy a subscription to macfixit, but I would like to be sure they have a solution. It seems like they recently posted an article with a solution to a problem that seems a lot like mine.
If anyone is subscribed and could check this article I would be grateful:
http://www.macfixit.com/article.php?...60530080506457

I would just like if it suggests a definitive fix for the latest PB 15" 1.67Ghz. I am outside of the warranty extension from Apple, also because I am still within the year after the purchase.

Thanks

could i add this please?i wonder if it is ram related
i have a powerbook 1,67 with 1,5GB RAM and 10.4.7.

when working with Photoshop CS2 for example with big files, or other things, like burning with Toast, i get kernel panics.

""Invalid Memory Access at
%SRR0: 05610d04
%SRR1: 00003030""

thanks

You should check the RAM with 'memtest':
http://www.memtestosx.org/

i checked it and it didnt find something.

i got more panics every day.
does it say its cause of airport here?:

"panic.log:

"Sat Jul 29 22:13:16 2006


Unresolved kernel trap(cpu 0): 0x300 - Data access DAR=0x0000000005003660 PC=0x0000000000029248
Latest crash info for cpu 0:
Exception state (sv=0x3F7E7280)
PC=0x00029248; MSR=0x00009030; DAR=0x05003660; DSISR=0x40000000; LR=0x000291C0; R1=0x2219BE10; XCP=0x0000000C (0x300 - Data access)
Backtrace:
0x000291C0 0x000233AC 0x000AC0AC 0x6B657926
Proceeding back via exception chain:
Exception state (sv=0x3F7E7280)
previously dumped as "Latest" state. skipping...
Exception state (sv=0x480CD000)
PC=0x9000B268; MSR=0x0200F030; DAR=0xA36D2758; DSISR=0x40000000; LR=0x9000B1BC; R1=0xBFFFF5E0; XCP=0x00000030 (0xC00 - System call)

Kernel version:
Darwin Kernel Version 8.7.0: Fri May 26 15:20:53 PDT 2006; root:xnu-792.6.76.obj~1/RELEASE_PPC
panic(cpu 0 caller 0xFFFF0003): 0x300 - Data access
Latest stack backtrace for cpu 0:
Backtrace:
0x00095718 0x00095C30 0x0002683C 0x000A8404 0x000ABD80
Proceeding back via exception chain:
Exception state (sv=0x3F7E7280)
PC=0x00029248; MSR=0x00009030; DAR=0x05003660; DSISR=0x40000000; LR=0x000291C0; R1=0x2219BE10; XCP=0x0000000C (0x300 - Data access)
Backtrace:
0x000291C0 0x000233AC 0x000AC0AC 0x6B657926
Exception state (sv=0x480CD000)
PC=0x9000B268; MSR=0x0200F030; DAR=0xA36D2758; DSISR=0x40000000; LR=0x9000B1BC; R1=0xBFFFF5E0; XCP=0x00000030 (0xC00 - System call)

Kernel version:
Darwin Kernel Version 8.7.0: Fri May 26 15:20:53 PDT 2006; root:xnu-792.6.76.obj~1/RELEASE_PPC"


and


""Tue Aug 1 15:07:26 2006


Unresolved kernel trap(cpu 0): 0x700 - Program DAR=0x00000000E7295000 PC=0x0000000000718AC4
Latest crash info for cpu 0:
Exception state (sv=0x38BD2500)
PC=0x00718AC4; MSR=0x00089030; DAR=0xE7295000; DSISR=0x42000000; LR=0x0071A6F4; R1=0x21F63D20; XCP=0x0000001C (0x700 - Program)
Backtrace:
0x0071A6F4 0x0072F07C 0x002D0D00 0x002CFBC8 0x000A9914
Kernel loadable modules in backtrace (with dependencies):
com.apple.iokit.AppleAirPort2(404.2)@0x6 f0000
dependency: com.apple.iokit.IONetworkingFamily(1.5.0 )@0x5be000
dependency: com.apple.iokit.IOPCIFamily(1.7)@0x45d00 0
Proceeding back via exception chain:
Exception state (sv=0x38BD2500)
previously dumped as "Latest" state. skipping...
Exception state (sv=0x3E427780)
PC=0x00000000; MSR=0x0000D030; DAR=0x00000000; DSISR=0x00000000; LR=0x00000000; R1=0x00000000; XCP=0x00000000 (Unknown)

Kernel version:
Darwin Kernel Version 8.7.0: Fri May 26 15:20:53 PDT 2006; root:xnu-792.6.76.obj~1/RELEASE_PPC
panic(cpu 0 caller 0xFFFF0007): 0x700 - Program
Latest stack backtrace for cpu 0:
Backtrace:
0x00095718 0x00095C30 0x0002683C 0x000A8404 0x000ABD80
Proceeding back via exception chain:
Exception state (sv=0x38BD2500)
PC=0x00718AC4; MSR=0x00089030; DAR=0xE7295000; DSISR=0x42000000; LR=0x0071A6F4; R1=0x21F63D20; XCP=0x0000001C (0x700 - Program)
Backtrace:
0x0071A6F4 0x0072F07C 0x002D0D00 0x002CFBC8 0x000A9914
Kernel loadable modules in backtrace (with dependencies):
com.apple.iokit.AppleAirPort2(404.2)@0x6 f0000
dependency: com.apple.iokit.IONetworkingFamily(1.5.0 )@0x5be000
dependency: com.apple.iokit.IOPCIFamily(1.7)@0x45d00 0
Exception state (sv=0x3E427780)
PC=0x00000000; MSR=0x0000D030; DAR=0x00000000; DSISR=0x00000000; LR=0x00000000; R1=0x00000000; XCP=0x00000000 (Unknown)

Kernel version:
Darwin Kernel Version 8.7.0: Fri May 26 15:20:53 PDT 2006; root:xnu-792.6.76.obj~1/RELEASE_PPC""


thanks

Did you do a full test? This usually takes several hours or overnight. It is best to do it from single-user mode to test the maximum amount of RAM.

The above reference to com.apple.iokit.AppleAirPort2 in the panic report might indicate a problem with the Airport card (or possibly with its driver software).

thanks!

after a long time. (switching ram modules)
i didnt have any kernel panics at all.

today was my first since 4 month

could i learn how to read this log today?

"Sat Oct 21 16:12:10 2006


Unresolved kernel trap(cpu 0): 0x300 - Data access DAR=0x0000000000000008 PC=0x00000000000A3FA4
Latest crash info for cpu 0:
Exception state (sv=0x38783500)
PC=0x000A3FA4; MSR=0x00009030; DAR=0x00000008; DSISR=0x40000000; LR=0x000711F4; R1=0x2220BCF0; XCP=0x0000000C (0x300 - Data access)
Backtrace:
0x00070648 0x00062584 0x000A825C 0x000AB780
Proceeding back via exception chain:
Exception state (sv=0x38783500)
previously dumped as "Latest" state. skipping...
Exception state (sv=0x384B6780)
PC=0x907BFCCC; MSR=0x0200F930; DAR=0x92AC67F8; DSISR=0x40000000; LR=0x907BFB7C; R1=0xBFFFE5A0; XCP=0x0000000C (0x300 - Data access)

Kernel version:
Darwin Kernel Version 8.8.0: Fri Sep 8 17:18:57 PDT 2006; root:xnu-792.12.6.obj~1/RELEASE_PPC
panic(cpu 0 caller 0xFFFF0003): 0x300 - Data access
Latest stack backtrace for cpu 0:
Backtrace:
0x00095138 0x00095650 0x00026898 0x000A7E04 0x000AB780
Proceeding back via exception chain:
Exception state (sv=0x38783500)
PC=0x000A3FA4; MSR=0x00009030; DAR=0x00000008; DSISR=0x40000000; LR=0x000711F4; R1=0x2220BCF0; XCP=0x0000000C (0x300 - Data access)
Backtrace:
0x00070648 0x00062584 0x000A825C 0x000AB780
Exception state (sv=0x384B6780)
PC=0x907BFCCC; MSR=0x0200F930; DAR=0x92AC67F8; DSISR=0x40000000; LR=0x907BFB7C; R1=0xBFFFE5A0; XCP=0x0000000C (0x300 - Data access)
"

There's not much in the way of clues easily discernible from that panic.log
I would guess that is what you'd expect if the underlying problem is a RAM failure since RAM failures produce more or less random symptoms as the programs (including the kernel) don't have any solid ground to stand on.

If you want to spend more time with this, see the Apple doc on kernel panics:
http://developer.apple.com/technotes/tn2002/tn2063.html

thanks
i see

isnt there an easier and fast way of testing ram than with apple dvd or this small app that needs about 6 or 12h to test?

i need my pb every day. so i cant stay only in single user boot mode to test the whole day. i would appreciate a solid test solution, where i could work same time with my pb. any idea?

thanks

In order to test the RAM, it can't be being used by any other application, so any fully effective test will necessarily preclude using your Mac for other things.
The best RAM tester I know of is "memtest" which is I presume what you are referring to.
Why can't you let "memtest" run overnight?

You can't get a good test of your memory, if you are also using that same memory space for your system, and the apps that you might be using at the time. The docs that come with Memtest cover that.

Run the test overnight, when you are not working ?

If you get a failure, then it's easy to decide to replace the memory.
Even if you get NO failures during a memory test, you still do not know (for certain) if the memory is good in the long term.

The ultimate memory 'test' is to replace your existing memory with known-good memory, and do your best to duplicate the kernel panic conditions, if you can. Get your replacement memory from a reliable source, such as www.crucial.com

good idea
i will let do it overnight
i once tried overnight i guess. long time ago.
then it froze and i never used it again (memtest)
if its freezing i cant read out which is out of order, i guess
i have the apple 512MB inside and additionally 1GB from Samsung.

i once orderer another 1GB RAM to get 2GB alltogether
i had to send it back after 2 month as i had too much freezes,
maybe the slots where the RAM reside are defective? how to test that please?
i dont understand, if they are defective, why can i work with it for 6 month without
any problems or panics?

maybe my airport is causing the panics. i heard that the airport software can cause panics.

Check this Apple service page. Your Powerbook could have this issue. The affected serial numbers are part way down the page.
http://docs.info.apple.com/article.html?artnum=303173
From that article

ah. thanks great!

i also know the problem now:
"about this mac" shows only 1GB RAM.
but i have 1,5 GB installed. this means the 512MB isnt recognized,
before i had the 1GB ram into this slot and it only showed 512MB in "about this mac"

so. its the slot is corrupt yes?
but apple lists only serials between W8503xxxxxx through W8518xxxxxx
and i have W8540....

so i dont get it repaird? i have still warranty.

what would you do now?

thx

Call Apple and have it taken in for service.