• Updated 2023-07-12: Hello, Guest! Welcome back, and be sure to check out this follow-up post about our outage a week or so ago.

Determining the causes of seemingly random freezes

jessenator

Well-known member
I'm wondering if there's a way to get a more precise level of information on random system/mouse freezes that don't coincide with a bomb/error message.

I saw one of Netfreak's videos where they did that for Copland, using a Powerbook as the dump location for whatever would happen on the Copalnd-end machine. Is there a process laid out there that documents what a lay user would need in order to diagnose a seemingly error-less crash?

E.g. with my Quadra 700, I was having some wiring issues with my fan power splitter, so while I wait for it to arrive I've been using a SCSI2SD (so the Molex can be solely dedicated to the fan), which may end up being the final destination for it. When I did the simple copy drive contents from the spinner and copied those to the SCSI2SD, the system would give me a strange error I'd never seen, something about an unresolved alias or something I wasn't able to get a 1:1 search result for on Google anyway. But thinking that wasn't a good idea anyway, I went back to a clean slate and put 7.1.1 on the SD card, just to weed that copy crap out of the equation.

However, after the computer has been on a while even my fresh install will have freezes, and it's never the same operation (that I'm doing) or the same time. I've run the Apple Personal Diagnostic on loop (checking logic board and RAM) almost 10 times in a row with no ill effect. I'll go to restart the machine, and it hangs after loading the Finder. I'll hit the reset and it hangs at startup. I'll hit it again and it boots fine, until I go to move a window. I'll even start up from a FloppyEmu image (APD Emergency image) and it'll hang on boot.   <-That's just one example series, so my assumption is that it's a hardware problem, not a software one, hence my desire to get a better understanding of what is triggering that error under the hood.

In further detail, my 700 is overclocked to 33 MHz: a fully qualified MC68040RC40 CPU in the slot, and a 66.666 MHz clock oscillator in a socket. But up until now (swapping out the drive), I've had it on and operating for quite long periods of time with no ill effects that I've noticed. I have a fan mounted atop the heatsink pulling air through it for some bonus thermal management, even though the CPU can handle 40. But maybe it really is the OC, I don't know.

 
Last edited by a moderator:

jessenator

Well-known member
 you tried Macsbug, I assume?
Now here's where my noobishness will truly shine. I tried using the interrupt/programmers button but I didn't get anything to display. Was I doing that correctly? Is there another way to work MacsBug?

 

Crutch

Well-known member
I see now you said you had a mouse freeze, meaning you can’t move the mouse at all right? Is this happening every time? Often that indicates a hardware-level issue and Macsbug might not work.  (In that case, almost nothing will work really ... ). I had similar issues to what you describe with a Turbo040 installed in one of my SE/30s in the past, never could figure it out entirely — seemed to be heat related (after a while the thing would run hot, and issues would occur randomly).

But yes to activate Macsbug you should just be able to hit the interrupt switch, assuming one is installed.  If Macsbug isn’t installed, that should bring up MicroBug.  Either way, if nothing happens, the Mac is past the point of diagnosis and must be reset.

If you CAN get Macsbug up, the next question will be to see what chunk of code is running at the time of the crash.  “WH PC” may tell you, or if not do a stack crawl with “SC6”.  The Macsbug reference linked above describes these commands - I actually think it’s one of the best Apple technical docs of the time, with great examples and even a basic overview of 68k assembly, but it does require patience to learn.

 
Last edited by a moderator:
Top