Slow shutdown being universal or not depends on how we define a slow shutdown. My rMBP shuts down in 20-25 sec or so but I don't remember my macs ever shutting down instantly with 10.8 or 10.7 or 10.6 or 10.5. Sometimes I get 4 sec shutdowns on the same machine but I don't expect them every time just because it is possible.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Apple Seeds Build 12E40 of OS X Beta 10.8.4 Developers
- Thread starter MacRumors
- Start date
- Sort by reaction score
That's a real pity. Because it's the one thing above anything else that would dramatically speed up Safari. It would indeed be "snappier" if it cached content correctly.
For me 10.8.1. was the fastest shutdown. It would shutdown in about 6 seconds. Now it takes about 20, which was the same for 10.8, 10.8.2 and 10.8.3.
Careful, now...
Once upon a time, a man put some toast in his toaster. It took a minute to do its thing. And it provided lovely nice crunchy toast.
But the man got impatient with all the hanging about and decided it would be a great idea to hit the toaster with a sledgehammer after 30 seconds, so he could get his toast out a bit sooner. Heck, if the toaster could switch itself off after 1 minute, why couldn't he?
Success! The toaster even 'worked' on a few occasions after that.
The man didn't really understand why the toaster was taking so long. He knew it was something to do with heat, or network connectivity, or iCloud synchronisation, or the type of bread, or maybe even something else far too technical to worry about. But he hit it anyway, because he really was a very impatient man. And the toast looked and tasted sort of OK. Toast version 10.8 should be exactly like Toast version 10.6, shouldn't it? I mean, heaven forbid, the toaster wouldn't be doing things all day long that make life a bit quicker, that need a bit of extra sorting out when the Toast button is hit, would it?
But one day, a piece of toast shot out and hit him in the eye with a message on it saying that 'file permissions on toast/donotbashme' were incorrect, and the filing system needed to be repaired.
"Oh, bugger!" said the man. "I need to hit it a bit sooner next time."
I'd love if it was silky smooth by now, I was simply saying that it isn't the lack of updates.
Nice parable. We who have slow shut down times (and boot up times too)with 10.8 want our new toaster to make the same quality toast, in the same amount of time, as our old toaster, preferably without having to rewire the outlet.
I don't think that is expecting too much, is it?
Wow, get of your arrogant high-horse. Yes, there's a lot of irrational insecurity going around especially among not so tech-savvy Mac users, but this is not it. Slow shutdown phenomenon starting with X.8.2 is a well-researched problem that many people hoped would be fixed with 8.3. I administer about 60 Macs at a university, and they all have the same problem with different hardware setups. People say it is a change in Mac OS X event handling that came about in 8.2. On machines where I installed 8.3 shutdown times were cut in half, but still longer that 8.0 or 8.1.
You are funny. You really are. Tell me why, exactly, is it a "problem" for computers to take as long as they need to in order to shutdown properly? You have intimate knowledge, I suppose, as to what the shutdown code is doing in the operating system. Well, do you?
----------
Read the bit about "not understanding what's going on" again. It's not the same toast.
"The shutdown is slow"
"Then don't shut it down"
Brilliant solution. Next thing it will be is "X program takes forever to open up" (it could be Safari or calendar or mail) and the solution will be "then don't shut it down ever".
"Then don't shut it down"
Brilliant solution. Next thing it will be is "X program takes forever to open up" (it could be Safari or calendar or mail) and the solution will be "then don't shut it down ever".
Why does Apple kill the processes at 20 seconds rather than let them "take as long as they need in order to shutdown properly"? Here's a hint, because if a process or app has hung, it might never shut down properly without the OS force quitting it. Apple figures 20 seconds is a reasonable time to wait for a response before killing it.
Now, if your console log shows that every single time you shutdown certain tasks are crashing and requiring them to be killed, then waiting 19 extra seconds is useless since the records show they will not respond within the given time interval anyway.
I'm all for letting systems take as long as they need to shut down properly, and that is exactly why I'd like Apple to fix things so when I open up my console there are no error logs. Then we wouldn't have cause to complain about slow shutdowns because the system is doing whatever it is supposed to do. But when there are error logs pointing to specific tasks that are not shutting down properly, and the slow response has been shown to be caused by those specific tasks, you can't tell me there isn't something wrong with them and that everything is normal. Error logs are not "normal" behavior, they are there to help diagnose a problem.
Actually, I went back and reread it, just to amuse myself.
It's the same hardware, apps, and files (let's call that the toast). My old toaster (SL) makes perfectly browned, yummy toast in 2-3 seconds (with a 7200RPM HDD, by the way, not a SDD); the new toaster (ML) takes 30+ seconds to make the same brown, yummy toast.
Now, my new toaster (ML) has some cool new features on it, including interconnectivity with my microwave (iPhone) and oven (iPad) that allows me to follow how my toast is coming along, by looking at these other devices and let's me see on my toaster how the roast is progessing in the oven, but I still want my toast to be ready in the same amount of time (perfectly brown and yummy, too).
Not only that, but I don't want my new toaster to go to sleep partially through the toasting cycle because it incorrectly thinks there is no toast in the slot (I'm referring to ML's propensity to sleep, even during critical operations like transcoding a file, or burning a disk); my old toaster (SL, if you're still following along) wouldn't sleep until after the toast was finished and browned to perfection.
Unless you mean that the hardware, apps and files are the TOASTER and the OS is the TOAST; or is SL BREAD and ML is a BAGEL, or is it a WAFFLE?!
The point being that, all other things being equal, many of us (not you, so it is not universal) have issues with the boot and shut down times of ML vs SL, along with some other quirks. I don't understand why you are so adamant in telling everyone that it is not ML, it's something else.
Thanks. But I'm not even saying it's a problem. It's a fact, or a phenomenon. No, I don't have intimate knowledge, I just read log files and forum threads, like you do. I assume that Apple changed the timeout policy, and propably for good reasons. Amond the hardware setups I work with, I see no difference in shutdown behavior or log prints depending on hard disc type, graphics card type, Apple hardware or Hackintosh, offline or online machine. So I assume those factors can be ruled out and a changed timeout policy fits the facts from my perspective.
Your toaster metaphor doesn't work here. At least tech-savvy mac users experiencing this phenomenon wouldn't complain for shutdown procedures taking longer if there was a good reason. Nobody complains about procedures having changed from X.6 to X.8 just because of that. Yet the OS not shutting down processes properly is something to take note off.
Again, get off your cock-sure knot-it-all attitude and don't always assume the worst about people you don't know posting on forums.
Your toaster metaphor doesn't work here. At least tech-savvy mac users experiencing this phenomenon wouldn't complain for shutdown procedures taking longer if there was a good reason. Nobody complains about procedures having changed from X.6 to X.8 just because of that. Yet the OS not shutting down processes properly is something to take note off.
Again, get off your cock-sure knot-it-all attitude and don't always assume the worst about people you don't know posting on forums.
Last edited:
I love you for this post. Thank you so much for a clever and well written point!
----------
The general community may appreciate your comments if you would show some mutual respect and decency. Insulting others invalidates any points you may attempt to convey. It irks me when a perfectly informative and civil thread is derailed by immature and disrespectful individuals who use the anonymity of the internet as a shield to deflect personal integrity. The idiom "you can catch more flies with honey than with vinegar" rings especially true in these instances, with all due respect.
OK, you win. "Universal" means it happens on 100% of machines, which may not be the case. It's probably not literally "universal", more like extremely widespread. Still definitely a problem with the OS and not user specific though.
People have actually tracked what the system is doing during shutdown, and things are just hanging and the system gives them time to end before finally killing them off. Nothing useful is happening during that time, it most certainly is a bug. And by your logic, when 10.6 shuts down dramatically faster, are you saying there's something wrong with that? Or somehow both a 4 second and a 20 second shutdown are both "as designed" even though there's no advantage to the slower one?
Read the post from johndoe above.
Fixed that for you.
Yeah, you'd think that there was 10.8.1, 10.8.2, two supplemental updates to 10.8.2, 10.8.3, and now 10.8.4 being tested.
That's only 6 updates in less than a year. They should be up to 10.8.18 or something by now.
need more effective updates...not mumble jumbo updates.
Airport issues still present.
my rmbp still fails to connect to my router after sleep. Only happens on my Belgacom B-Box router. I always have to restart to get back on the internet, this has been sinds 10.8.0
And now that the 2 second shut down is gone, it is very irritating.
my rmbp still fails to connect to my router after sleep. Only happens on my Belgacom B-Box router. I always have to restart to get back on the internet, this has been sinds 10.8.0
And now that the 2 second shut down is gone, it is very irritating.
OK, so which process is it that has the propensity to stall? Let's take it from there.
Sorry, but when people assert that there are universal problems caused by a shutdown bug when they are not aware of a) the true number of affected user, b) what the code is doing and c) extraneous factors, they need to take stock of their own arrogance.
If you really want to 'fix' a slow shutdown, in the first case identify the process that is taking the longest to close and establish if there is any commonality across users.
Of course, it's a lot lot easier just to moan that there's a shutdown bug.
com.apple.coreservices.appleevents
com.apple.securityd
com.apple.launchd
com.apple.coremedia.videodecoder
These are the most common ones.