Ugly OSX Snow Leopard Freeze - at a loss.
January 3, 2010 11:54 AM Subscribe
Bizarre OSX Snow leopard problem. MBP unibody, 3.06 core2 duo. 8 GB ram. 500 Gig 5400 RPM drive. OSX 10.6.2
Starting about four weeks ago, my system slowly spins out of control - dock/finder/force quit become some of the first apps to fall. This ends in a freeze - UI responsive (mouse, some window movement) with the eventual need to force restart (holding down the power button) It's so bad that I'm constantly running the Console and Activity Monitor to see if I can find some clues.
It has been intermittent - with no fixed pattern of Apps running. It seems to take at least 3-4 hours for it to occur - but has gone longer than 19 at least one.
Initially, a some of the restarts resulted in the system thinking it needed a firmware update. This hasn't continued, but it was a very strange symptom.
I ran Techtool's full set of tests: Ram, drive integrity, Smart status. Passed.
Big obvious things: Permissions, Disk Util, Disk Warrior. FSCK. Onyx (and cleared all the major caches.) Font book's font check and Keychain's first aid.
When I took my system to my local dealer, we pulled my drive (as they give me a replacement.) The problems happened on the replacement system (with my drive in it.)
The dealer ran Apple's full suite of tests (including a couple of overnight tests) and still nothing; We added my HD back and it still passed all Apple tests.
OS Repairs that have failed:
I've tried a combo update.
I've done an Archive & Install.
I've tried a Clean install and moved over key parts of my user- documents, music and photos. *I added back my preferences & some key Application support items.* I star this because it leads me to believe that it's related to my specific user.
We replaced my HD: and I did a time machine restore. Since none of the above system updates made a difference - I felt it was better to work on my full system.
Things I've removed:
Orbicule/Undercover, VMWare/Fusion, Glims, Little Snitch.
Jumpcut, evernote
Blueharvest, Pinpoint, Mouse Locator.
I think it's specific to my user - but I can't really be sure; the problem is very much intermittent. Just creating another user won't necessarily show the problem - the system has run over 19 hours without a crash. I'm really resistant to this because it doesn't show where the problem came from - and I'm concerned that if I reload everything that it'll just occur again.
I thought it might be related to iTunes, my iPhone, Adium, Safari or Sleep, as each of these seem to have triggered it.
I thought it might be temperature related and ran iStat Server - while occasionally the temperatures have been high; all are in spec.
I pulled corrupt extensions (that Onyx wouldn't read) and items that mdimporter (spotlight) had trouble with (mostly some IMAP attachments)
I've spent all day running/quitting/sleeping the machine trying to CAUSE the problem. I can't seem to - and I'm at a complete loss.
It has been intermittent - with no fixed pattern of Apps running. It seems to take at least 3-4 hours for it to occur - but has gone longer than 19 at least one.
Initially, a some of the restarts resulted in the system thinking it needed a firmware update. This hasn't continued, but it was a very strange symptom.
I ran Techtool's full set of tests: Ram, drive integrity, Smart status. Passed.
Big obvious things: Permissions, Disk Util, Disk Warrior. FSCK. Onyx (and cleared all the major caches.) Font book's font check and Keychain's first aid.
When I took my system to my local dealer, we pulled my drive (as they give me a replacement.) The problems happened on the replacement system (with my drive in it.)
The dealer ran Apple's full suite of tests (including a couple of overnight tests) and still nothing; We added my HD back and it still passed all Apple tests.
OS Repairs that have failed:
I've tried a combo update.
I've done an Archive & Install.
I've tried a Clean install and moved over key parts of my user- documents, music and photos. *I added back my preferences & some key Application support items.* I star this because it leads me to believe that it's related to my specific user.
We replaced my HD: and I did a time machine restore. Since none of the above system updates made a difference - I felt it was better to work on my full system.
Things I've removed:
Orbicule/Undercover, VMWare/Fusion, Glims, Little Snitch.
Jumpcut, evernote
Blueharvest, Pinpoint, Mouse Locator.
I think it's specific to my user - but I can't really be sure; the problem is very much intermittent. Just creating another user won't necessarily show the problem - the system has run over 19 hours without a crash. I'm really resistant to this because it doesn't show where the problem came from - and I'm concerned that if I reload everything that it'll just occur again.
I thought it might be related to iTunes, my iPhone, Adium, Safari or Sleep, as each of these seem to have triggered it.
I thought it might be temperature related and ran iStat Server - while occasionally the temperatures have been high; all are in spec.
I pulled corrupt extensions (that Onyx wouldn't read) and items that mdimporter (spotlight) had trouble with (mostly some IMAP attachments)
I've spent all day running/quitting/sleeping the machine trying to CAUSE the problem. I can't seem to - and I'm at a complete loss.
Response by poster: jedicus - The Hard drive has already been replaced.
posted by Towelie at 12:09 PM on January 3, 2010
posted by Towelie at 12:09 PM on January 3, 2010
By the way, my advice assumes that 'we replaced my HD' means you put your old drive back into your system, not that you put a new drive in. Which is the case?
posted by jedicus at 12:10 PM on January 3, 2010
posted by jedicus at 12:10 PM on January 3, 2010
Response by poster: It's the other way. The Hard drive was replaced with a new one at an Apple Store.
posted by Towelie at 12:11 PM on January 3, 2010
posted by Towelie at 12:11 PM on January 3, 2010
Okay, let's get a few things out of the way:
- when you think it's trying to run a firmware update (you see the status bar at startup), this is simply a new progress bar in Snow Leopard that is showing you that fsck is occurring (which is occurring because you had to force a shutdown).
- TechTool is practically worthless. It does almost nothing that can't be done with built-in tools, Apple-supplied tools (Hardware Tests, etc.), or with better tools available (like DiskWarrior). I began avoiding it a long time ago and haven't used it in years.
- You are correct that if the problem persisted despite moving your HD to another machine, then the problem is almost certainly not hardware and is probably software-related.
You haven't mentioned if you have any peripherals attached to this machine. In my experience, I've seen flaky hardware peripherals (namely a failing HD) cause this kind of behavior. So, does this problem occur when hardware peripherals are attached?
You might also attack this problem by continuing what you're doing by analyzing Console, Activity Viewer, and logs, but you might also run a program called fseventer that's going to show you file system events (reading/writing to your HD) and see if you can spot a correlation between the problem occurring and a file being read from, written to.
I'd also replace the RAM with known-good RAM. All of it.
posted by mrbarrett.com at 12:32 PM on January 3, 2010
- when you think it's trying to run a firmware update (you see the status bar at startup), this is simply a new progress bar in Snow Leopard that is showing you that fsck is occurring (which is occurring because you had to force a shutdown).
- TechTool is practically worthless. It does almost nothing that can't be done with built-in tools, Apple-supplied tools (Hardware Tests, etc.), or with better tools available (like DiskWarrior). I began avoiding it a long time ago and haven't used it in years.
- You are correct that if the problem persisted despite moving your HD to another machine, then the problem is almost certainly not hardware and is probably software-related.
You haven't mentioned if you have any peripherals attached to this machine. In my experience, I've seen flaky hardware peripherals (namely a failing HD) cause this kind of behavior. So, does this problem occur when hardware peripherals are attached?
You might also attack this problem by continuing what you're doing by analyzing Console, Activity Viewer, and logs, but you might also run a program called fseventer that's going to show you file system events (reading/writing to your HD) and see if you can spot a correlation between the problem occurring and a file being read from, written to.
I'd also replace the RAM with known-good RAM. All of it.
posted by mrbarrett.com at 12:32 PM on January 3, 2010
Response by poster: I've had this crash with and without other hardware attached.
Most of the time, when I'm traveling, it's without other hardware attached - and it still occurs I've had the crash on a different system (with my hard drive) - doesn't that rule out RAM?
I'm running fseventer (I've used it a long time ago, thanks for the link.) What (if anything) am I looking for with it?)
posted by Towelie at 12:42 PM on January 3, 2010
Most of the time, when I'm traveling, it's without other hardware attached - and it still occurs I've had the crash on a different system (with my hard drive) - doesn't that rule out RAM?
I'm running fseventer (I've used it a long time ago, thanks for the link.) What (if anything) am I looking for with it?)
posted by Towelie at 12:42 PM on January 3, 2010
If it's intermittent, the problem's most likely not going to appear on your tests, unless you're running them immediately after one of these crashes. Temperature would be my guess as well. Have you tried running a few Google Videos simultaneously and leaving the fans in nuclear mode to see if that brings it on?
Do your fans ever go into nuclear mode (if you've got a Mac laptop and Flash, they should be for sure)?
posted by bonaldi at 1:08 PM on January 3, 2010
Do your fans ever go into nuclear mode (if you've got a Mac laptop and Flash, they should be for sure)?
posted by bonaldi at 1:08 PM on January 3, 2010
Response by poster: Boom. It just happened again. One other thought - DNS services (new browser windows, mail retrieval) seem to fail/be an early indicator.
posted by Towelie at 1:08 PM on January 3, 2010
posted by Towelie at 1:08 PM on January 3, 2010
Response by poster: bonaldi- as far as stress test: I ran yes > /dev/null as detailed here and downloaded and tried CPU Test . Even with the fans at 6000RPM it was stable; since it' just crashed, I check the temps - nothing crazy; they were <3000 prior to the crash.
posted by Towelie at 1:12 PM on January 3, 2010
posted by Towelie at 1:12 PM on January 3, 2010
Does the problem happen connected via both wired and wireless?
posted by mrbarrett.com at 1:24 PM on January 3, 2010
posted by mrbarrett.com at 1:24 PM on January 3, 2010
Response by poster: I haven't tried a wired connection (although I could right now…) Any thoughts to what that might be causing a wireless problem?
posted by Towelie at 1:26 PM on January 3, 2010
posted by Towelie at 1:26 PM on January 3, 2010
Corrupted prefs in /Library/Preferences/SystemConfiguration
Or possibly a failing Airport card.
posted by mrbarrett.com at 1:29 PM on January 3, 2010
Or possibly a failing Airport card.
posted by mrbarrett.com at 1:29 PM on January 3, 2010
Response by poster: Trashed the whole system config folder.
posted by Towelie at 1:44 PM on January 3, 2010
posted by Towelie at 1:44 PM on January 3, 2010
Are you using Firefox? Recent versions (anything newer than 3.5.3) lock up several Macs I use in essentially the manner you describe -- I've finally just downgraded to 3.5.3 and turned off automatic version updates, and that seems to have resolved it...
posted by genehack at 2:37 PM on January 3, 2010
posted by genehack at 2:37 PM on January 3, 2010
Here is a generally comprehensive procedure that I have used for troubleshooting Mac issues:
A. What happens if you set up a new user account and log into that for a while, just to test how things work out?
If performing step A does not show the same symptoms, some item in your home folder's Library folder is causing issues. Migrate your personal documents to the freshly-made account. You can migrate some of your preferences in groups to see if one or some of them are causing issues. You're done.
B. If your laptop shows the same symptoms, if you have an external drive, install OS X on there, booting a fresh OS installation from that. This leaves your hard drive data untouched for testing purposes.
If, in performing step B, the laptop shows the same symptoms, there may be either a non-hard-drive-related hardware issue (e.g. bad Airport card or component on the motherboard) or a bug with the version of OS X you are using. More likely a non-hard-drive-related hardware issue, though. Call AppleCare or visit the Apple Store for hardware support. You're generally done, as far as what you can do yourself.
C. If that doesn't show the same symptoms, it is probable that the operating system installation on your hard drive is faulty in some way — or the hard drive itself — and that you should try, in this order, to determine which:
1. Back up your personal documents and Library folder, reformat the hard drive, reinstall the operating system and recreate a home folder. If the problem does not reappear, you're done.
2. If #1 doesn't cause the issue to go away, replace the hard drive and reinstall the OS, replace your personal data, etc.
This seems like a lot of steps, but you may only have to go to a certain point, depending on the issue. I had used this general process for eight years and it worked every time for correctly and efficiently isolating the problem.
Above all else, back up your data before doing anything. A good and recent backup will save you many tears.
Good luck!
posted by Blazecock Pileon at 2:51 PM on January 3, 2010 [1 favorite]
A. What happens if you set up a new user account and log into that for a while, just to test how things work out?
If performing step A does not show the same symptoms, some item in your home folder's Library folder is causing issues. Migrate your personal documents to the freshly-made account. You can migrate some of your preferences in groups to see if one or some of them are causing issues. You're done.
B. If your laptop shows the same symptoms, if you have an external drive, install OS X on there, booting a fresh OS installation from that. This leaves your hard drive data untouched for testing purposes.
If, in performing step B, the laptop shows the same symptoms, there may be either a non-hard-drive-related hardware issue (e.g. bad Airport card or component on the motherboard) or a bug with the version of OS X you are using. More likely a non-hard-drive-related hardware issue, though. Call AppleCare or visit the Apple Store for hardware support. You're generally done, as far as what you can do yourself.
C. If that doesn't show the same symptoms, it is probable that the operating system installation on your hard drive is faulty in some way — or the hard drive itself — and that you should try, in this order, to determine which:
1. Back up your personal documents and Library folder, reformat the hard drive, reinstall the operating system and recreate a home folder. If the problem does not reappear, you're done.
2. If #1 doesn't cause the issue to go away, replace the hard drive and reinstall the OS, replace your personal data, etc.
This seems like a lot of steps, but you may only have to go to a certain point, depending on the issue. I had used this general process for eight years and it worked every time for correctly and efficiently isolating the problem.
Above all else, back up your data before doing anything. A good and recent backup will save you many tears.
Good luck!
posted by Blazecock Pileon at 2:51 PM on January 3, 2010 [1 favorite]
Response by poster: Genehack - yeah using the latest version. I'll have to look into it.
Blazecok - A) not a solution - just a direction
B) tested on different hardware
C) I have moved from a home folder.
and replacing my data isnt' a great idea.
posted by Towelie at 4:51 PM on January 3, 2010
Blazecok - A) not a solution - just a direction
B) tested on different hardware
C) I have moved from a home folder.
and replacing my data isnt' a great idea.
posted by Towelie at 4:51 PM on January 3, 2010
A) not a solution - just a direction
I'm not sure what this means. Did you try it? A new user account has its own home folder. If there is a preference, font, cached item or other application support file in your home folder's Library folder that is causing instability, a new account will fix the problem.
B) tested on different hardware
I'm not sure what this means. If you boot your laptop from an external hard drive, you remove the hard drive from the equation and can determine if there is a problem with the logic board or other hardware.
C) I have moved from a home folder.
To what?
and replacing my data isnt' a great idea
You don't have to replace it, just back it up and move it to a new account or to a new hard drive.
Everything I've mentioned is fairly standard troubleshooting. If you contact AppleCare or an Apple Store, you will need to do some of the same things.
posted by Blazecock Pileon at 7:35 PM on January 3, 2010
I'm not sure what this means. Did you try it? A new user account has its own home folder. If there is a preference, font, cached item or other application support file in your home folder's Library folder that is causing instability, a new account will fix the problem.
B) tested on different hardware
I'm not sure what this means. If you boot your laptop from an external hard drive, you remove the hard drive from the equation and can determine if there is a problem with the logic board or other hardware.
C) I have moved from a home folder.
To what?
and replacing my data isnt' a great idea
You don't have to replace it, just back it up and move it to a new account or to a new hard drive.
Everything I've mentioned is fairly standard troubleshooting. If you contact AppleCare or an Apple Store, you will need to do some of the same things.
posted by Blazecock Pileon at 7:35 PM on January 3, 2010
Response by poster: Sorry, my initial post actually talked about each. I must have posted when tired.
a) I've done a new user and moved data over manually. A new user by itself doesn't seem the answer as I can't trigger the problem. The problem actually takes 3+ hours before it shows itself (if it does.) For example, right now I'm running at 13 hours without a crash.
b) It means that when I pulled the hard drive and put it in different hardware, ruled out the motherboard, ram etc.
c) I moved my data to a new home folder and had the same problems.
posted by Towelie at 3:16 AM on January 4, 2010
a) I've done a new user and moved data over manually. A new user by itself doesn't seem the answer as I can't trigger the problem. The problem actually takes 3+ hours before it shows itself (if it does.) For example, right now I'm running at 13 hours without a crash.
b) It means that when I pulled the hard drive and put it in different hardware, ruled out the motherboard, ram etc.
c) I moved my data to a new home folder and had the same problems.
posted by Towelie at 3:16 AM on January 4, 2010
FF 3.5.7 hasn't crashed my MBP, or my wife's unibody MB. For that matter the previous versions didn't crash it either, and I was running everything from prerelease versions of FF 3 up through the current builds on the same machine. While it remains a possibility, I wouldn't expect this to be the main culprit.
posted by caution live frogs at 10:56 AM on January 4, 2010
posted by caution live frogs at 10:56 AM on January 4, 2010
Is this a 15" or 17"? There's a huge thread on Apple's support forums about SATA II hard drives in the unibody MBPs, seems to be bad engineering of the ribbon connector for the hard drive / drop sensor (to park the disk when the laptop falls).
Firmware updated that caused the most problems:
http://support.apple.com/kb/HT3561
Original thread:
http://discussions.apple.com/thread.jspa?threadID=2054387&start=0&tstart=0
Continuation thread:
http://discussions.apple.com/thread.jspa?threadID=2267098
Apple hasn't officially acknowledged the problem, but it might be something else to consider.
posted by ZakDaddy at 2:18 PM on January 4, 2010
Firmware updated that caused the most problems:
http://support.apple.com/kb/HT3561
Original thread:
http://discussions.apple.com/thread.jspa?threadID=2054387&start=0&tstart=0
Continuation thread:
http://discussions.apple.com/thread.jspa?threadID=2267098
Apple hasn't officially acknowledged the problem, but it might be something else to consider.
posted by ZakDaddy at 2:18 PM on January 4, 2010
Best answer: Post apple 2 hr call. Apple engineering is looking at my console log files.
It's my user. We know it's not hardware (due to the switching of the HD in a separate MBP) and the same error occurring. And we swapped for a new HD. And it's not a user preference (as I've pulled all of them.)
It doesn't seem to happen in a new user.
It's something to do with my wireless. About 3 hrs on wireless….boom. System freezes up. If I'm wired…I'm fine. Currently, I'm up 3 days with zero problems.
The 'new user' solution isn't a solution - it's a workaround - and the Apple Tier 2 people agreed; I could just be inducing the problem back after I set everything back up; Hopefully this week I'll be hearing from them about what step(s) they'd like me to do.
posted by Towelie at 8:35 PM on January 10, 2010
It's my user. We know it's not hardware (due to the switching of the HD in a separate MBP) and the same error occurring. And we swapped for a new HD. And it's not a user preference (as I've pulled all of them.)
It doesn't seem to happen in a new user.
It's something to do with my wireless. About 3 hrs on wireless….boom. System freezes up. If I'm wired…I'm fine. Currently, I'm up 3 days with zero problems.
The 'new user' solution isn't a solution - it's a workaround - and the Apple Tier 2 people agreed; I could just be inducing the problem back after I set everything back up; Hopefully this week I'll be hearing from them about what step(s) they'd like me to do.
posted by Towelie at 8:35 PM on January 10, 2010
WOW
2 friends and I are having the exact same problem with snow leopard! Crash every ~3 on network connection. Out of desperation I completely reinstalled from scratch…and still failure. After a few hours of the phone with Apple they thought it was a Firefox plugin and suggested creating a new user (we tried safari under a new user with no luck)
We think something makes a blocking network call. IN THEORY (encapsulation)- a particular program should not be able to place a system-wide blocking network call.
Consistencies across our machines:
15" macbook pros of various make model all with stock ram & hdd
using wireless
firefox 3.5/3.6
itunes
iphone
adium <> little snitch <>
VERY interesting about it being exclusively a wireless problem. Please keep me updated!!
posted by Zamiang at 9:16 PM on January 21, 2010
2 friends and I are having the exact same problem with snow leopard! Crash every ~3 on network connection. Out of desperation I completely reinstalled from scratch…and still failure. After a few hours of the phone with Apple they thought it was a Firefox plugin and suggested creating a new user (we tried safari under a new user with no luck)
We think something makes a blocking network call. IN THEORY (encapsulation)- a particular program should not be able to place a system-wide blocking network call.
Consistencies across our machines:
15" macbook pros of various make model all with stock ram & hdd
using wireless
firefox 3.5/3.6
itunes
iphone
adium <> little snitch <>
VERY interesting about it being exclusively a wireless problem. Please keep me updated!!
posted by Zamiang at 9:16 PM on January 21, 2010
I've been having this same issue also, and was close to bringing it here. I never would have guessed that it was a wireless problem!
posted by danb at 9:18 PM on February 2, 2010
posted by danb at 9:18 PM on February 2, 2010
All right, I suppose it's not the same issue -- I created a new user account, and got a 45-second beachball within five minutes. I suppose a trip to the Genius Bar is in my future.
Regardless, I'll be interested to hear what Towelie's Apple engineers have to say.
posted by danb at 6:27 PM on February 3, 2010
Regardless, I'll be interested to hear what Towelie's Apple engineers have to say.
posted by danb at 6:27 PM on February 3, 2010
no luck from the genius bar.
i brought my macbook pro (15" 2.4ghz core duo w/ all stock parts) to the apple store and after completely reinstalling from scratch AND replacing the actual wireless card (and the entire display half of my mac). i still get around 3-4 crashes per-day at 3 hour intervals. YAY!
*still get no crashes when plugged directly into wireless or tethering from iphone via bluetooth or usb.
a friend got a new macbook pro with snow leopard and is also having this wireless-related crashing problem.
judging from the lack of activity on posts like this it seems like few people have this problem? can it be?
posted by Zamiang at 4:45 PM on February 16, 2010
i brought my macbook pro (15" 2.4ghz core duo w/ all stock parts) to the apple store and after completely reinstalling from scratch AND replacing the actual wireless card (and the entire display half of my mac). i still get around 3-4 crashes per-day at 3 hour intervals. YAY!
*still get no crashes when plugged directly into wireless or tethering from iphone via bluetooth or usb.
a friend got a new macbook pro with snow leopard and is also having this wireless-related crashing problem.
judging from the lack of activity on posts like this it seems like few people have this problem? can it be?
posted by Zamiang at 4:45 PM on February 16, 2010
This thread is closed to new comments.
It sounds like the problem is your hard drive. Hard drive failures are notoriously difficult to diagnose. Google (which operates man, many hard drives) did a study of hard drive failure rates and patterns [pdf] and found that the usual drive diagnostics are not particularly predictive of drive failure. So the fact that your system passes various tests is, unfortunately, not very meaningful.
So why not replace the hard drive? Use SuperDuper! and a cheap USB drive enclosure to clone your drive on to the replacement, then swap it in.
posted by jedicus at 12:08 PM on January 3, 2010