Marc Sira <[EMAIL PROTECTED]> replied to:
>> I've found that OS 8.6 is EXCEPTIONALLY sensitive to ram and disk
>> directory integrity. I have a G3 tower with 512mb ram. Everything ran
>> fine on 8.5.1 (which is quite fault tolerant) but I had repeated
>> system bus errors after upgrading to 8.6. Careful checking narrowed
>> this to a bad 64mb ram module (1 of 8) which ran faultless under
>> 8.5.1. Each module was individually tested with TT2.1.1 and isolated
>> bootup.
>
> It's a bit facile to say "8.5.1 is fault tolerant and 8.6 is picky".
> More likely a coincidence, in that something terribly important
> happened to load into the bad RAM on your 8.6 install, while the
> earlier install had (not exactly fortunately in that it masks a serious
> problem) put something much less crucial there. This would likely be
> particular to your machine model and extension set, since those affect
> what gets loaded and when. My own policy is to test the hell out of
> any new memory (including that in a new machine) as soon as I get it
> (that also precludes turning off the POST RAM test right away, near-useless
> though it is).
Actually, I run several differently configured systems on multiple drives
and partitions. "Extensions off" also failed with all OS 8.6 bootup
attempts, also from different partitions. These would change the load order
and ram usage considerably so your theory is as good/bad/facile as mine ;).
There were also several others who suddenly had OS 8.6 upgrade problems on
desktops on another digest, but too few to draw conclusions--hence I
speculate. I also keep a "clean" installed system without 3rd party
extensions on a separate partition to
reference all trials.
>
> Your problem sounds like it was a badly corrupted power manager, though
> it's always tough to say in retrospect. It's even possible that the
> 2400/3400 has a design flaw that allows the power management firmware
> to get into a (soft!) state from which it sometimes can't extricate
> itself - thus our dead motherboards aren't really dead, they just need
> some extraordinary measure to reset them that we can't supply. But that
> too is only a theory (not a particularly pretty one, since odds are
> every 2400 would eventually reach that state, and no fix would be
> permanent). And it doesn't particularly jibe with my experience, where
> the problems seemed to be progressive, so take it very much with a grain
> of salt.
>
>> --Screen freezes while online (and probably offline) can severely
>> corrupt PRAM and disk directories.
>
> It's important to note that these two things aren't particularly
> related. Directory corruption is pretty simple, and results entirely
> from the OS being interrupted before it can write out some cached data
> belonging to the directory. PRAM corruption is a bit more nebulous, and
> power manager corruption more bizarre still.
Correct, these are not normally closely related but there's a first time for
every bad coincidence. Someone else's computer failed completely doing the
same thing. My point in the post was to give indications, procedures and
encouragement to others for persistence in trying to bring a 2400 back from
the dead. I recall from a few digests back that Tim Seufert also stressed
the necessity of multiple power manager resets to revive recalcitrant 2400s.
>
>> --The 2400 perhaps demands a cleaner disk directory than other PBs,
>> irrespective of 603e or G3 CPU, but this may be specific to OS 8.6
>> (remembering that a clean install of 8.6 is not possible for OS 8.5
>> updaters and may have partial influence)
>
> Sorry, but this is absurd. :) I think you may be reaching an
> understandable but hasty conclusion from a single experience here, but
> the bottom line is that this theory doesn't work. I'm not picking on
> you here, but I do see a lot of misguided theories in Mac troubleshooting,
> and it doesn't always help people in similar situations to have too many
> half-formed notions kicking around.
Absurd it may be, but, as you can see, I'm speculating openly and inviting
opinions/explanations to better understand the problem. The fact is the
problem hard drive would boot both a G3-3400 directly and a G3-300 tower (in
SCSI mode) but not the 2400 (either directly installed or in SCSI mode).
That is absurd and defies logic. Apple recommends clean installs whenever
possible, and with OS 8.6 updater this is not possible. Any good
ideas/explanations why a directory will boot on certain computers and not
others?
---
Sidney Ho
[EMAIL PROTECTED]
----------
Duo/2400 List, The friendliest place on the Net!
A listserv for users and fans of Mac subportables.
FAQ at <http://www.themacintoshguy.com/lists/DuoListFAQ.shtml>
Be sure to visit Mac2400! <http://www.sineware.com/mac2400>
Dr. Bott LLC | Get a LapBottom with new �GRIPPIER� gripper pads!
| http://www.drbott.com/prod/lapbot.html $23.00
Midwest Mac Parts <http://www.midwestmac.com> | After-market parts
for Macs. Fix your ticking dock for $59 w/ trade-in of PS. 888-356-1104
PowerBook Guy is | Click here!
Everything PowerBook! | http://www.powerbookguy.com
