Submit Hint Search The Forums LinksStatsPollsHeadlinesRSS
14,000 hints and counting!


Click here to return to the 'sleep coerces 'lost' internal dvd drives (on laptop) to re-appear | USB non-ASAS drives R flaky' hint
The following comments are owned by whoever posted them. This site is not responsible for what they say.
sleep coerces 'lost' internal dvd drives (on laptop) to re-appear | USB non-ASAS drives R flaky
Authored by: zahadum on Aug 07, '07 06:32:52PM

i havent tried the sleep option - because i am scared to death i will suffer a one of notorious hang-on-wake situations - so i will have to wait until i dont have anything important enqueued to test this hint.

however, i can confirm that the 'disappearing' dvd problem occurs on ppc mini's as well - among other macs; the problem can also be one of 'lock-out- where the device seems to be registered in disk util / sys profile / finder / etc .... but cant be accessed.

but a restart is not always required.

for external drives, re-connecting the usb cable will often force the usb system driver / disk arbitration mechanism to reinitialize itself.

however, there is another related problem that effects both internal & external drives and does absolutely require a restart: sometimes the dvd drive stutters; video playback is halting and jittery; and data access is atrocious (a 45 _minute_ RIP can take 10-20 *HOURS*) ... especially if two instances of MTR are running, with different user privileges - as the wonderful sudo tool Pseudo allows one to do.

often this access problem seems to be a resource contention conflict when two or more apps try to access the same device (eg dvd player; mactheripper; handbrake - sometimes even just momentary polling will set off the lock-up ... this cascade of death is just one of the many, special 'gotchas' that osx inflicts on us) ... the indispensable telemetry tool fseventer (sweet!) shows that there is a loop back to a file (dvd.css, i think) which is part of the jam.

side-note: obviously apple has failed to design a graceful IO arbitration API .. however this should not be a surprise since it is part of the same pattern that cut R&D in half when Jobs returned to apple ... it is but more fall-out from the bone-headed decision to ditch os/x's original, sturdy driver model - built in objectiveC - in favour of the ludicrously brittle C++ api - which consequently deprives of us *true* plug'n'play ... it took more than a decade for linux kiddies to catchup with the elegance of OSF/2 - which died out when compaq killed the DEC Alpha; whilst in the same timeframe, apple just threw away the only other (commercially viable) robust driver model for no corresponding benefit. ALL HAIL THE GENUIS STEVE JOBS.

as indicated, this jam up can effect internal & external drives; and sometimes re-initializing the (usb) external device will not set things right - only a restart will clear the decks ....

this is profoundly intrusive if one has other mission-critical work which is mid-way through some heavy-duty crunching (that dont arent created with the ability to serialize an intermediate pause/resume mode --- I AM LOOKING AT YOU, COREDATA!).

so, one is forced to do with out dvd drive access for a day or more while waiting for a production run to finish ... or else one is forced to endure long RIP's (did i mention 10-20 hours?!) as the only way to offload the data on to a dasd for smooth playback.

this penalty is clearly unacceptable.

apple should be ashamed at shipping a platform with first-order gremlins throughout the system: driver disasters (the dvd lock-up; wobbly usb bluetooth; etc etc); network nightmares (the cascade of death); and filesystem fandango's (the famous err-36) .... not to mention a Finder that is STILL _broken_ after 10 years!!

in addition to the appalling lack of basic engineering practice at apple - WTF happened to UML?! - there is an equally shoddy approach to QA at apple (no surprised that SDL isnt not SOP!).

my guess is that this dvd jamming situation is NOT the (singular) result of not using a drive which isnt ASAS (apple shipped & supported) - the 'lost' internal drive is a good clue about the disappearance being endemic!

my guess is that apple just doesnt give a damn ...

want more proof of apple's systemic indifference to device management? .... just look at how pathetic the telemetry is for disk utility when trying to unmount a recalcitrant drive: you will receive a vague & useless message 'device cant be unmounted because it is in use by another app' .... the culprit is process NEVER named (so that it can be wrestled with); nor is the _mistakenly_ unclosed file identified (again, just a hint would be sufficient to help manual intervention); nor is there an over-ride command to force a dismount (perhaps with an authentication required to give the less sophisticated user a chance to rethink the consequences).

Telemetry and os/x are an oxymoron .... and the exception proves the rule! - it is unfathomable, though most welcome, that XRAY is in leopard's tool-chain (alas, while DTrace did make it over from solaris, the other gem from SUN -ZFS - kinda sorta got left behind, perhaps part of the overall schedule slip of leopard).

leopard should have been the occasion to fix all this shabbiness: osx needs a central device manager thatis as coherent as the Launchd approach is to process management (i should be able coerce any device to be unloaded - thinking along the same lines as what the dismount kext command _hints_ at being able to do ... again, sadly, linux actually has a full implemtation but osx does not).

but that aint gonna happen ... cuz ... apple just doesnt give a damn.

---
mailto:osxinfo _at_ yahoo.ca



[ Reply to This | # ]