OS X Mavericks kernel_task eating 100% CPU | Hacker News Reader

OS X Mavericks kernel_task eating 100% CPU

discussions.apple.com

73 pointsEpicEng107 comments

I had a strange problem appear after I spilled tea on my 2010 MBP last year. Continually, kernel_task was using an entire core (150-250% in top) and no amount of reboots could fix this. The built-in hardware diagnostic indicated it was not compatible with my Mac, and I could find nothing else that apparently was wrong. My computer was running very slowly though, and the system load was always over 4.

Finally I found through research that this is somehow related to power management. Apparently the system will have the CPU loop simple operations to prevent the processor from doing actual work and heating up. Sure enough, disabling power management by hacking a kernel extension made the kernel_task problem go away. My Mac has run perfectly for three months ever since I made that fix, after being nearly disabled for a year.

I remember having this same problem on a hackintosh. Installing the NullPowerManagement kext was the usual fix back then since DSDT patches were brand new and not well known by the community. If that kext is still available and compatible with 10.9 it's probably a good way to handle this until they fix it.

Damn, my 2009 Macbook has been running very slowly for the past 4 months or so and no amount of rebooting would resolve the problem. I was actually puzzled when it slowed down to a crawl becoming basically useless and would even restart with the same slowness. Could you elaborate on your fix?

See my comment above - it might help.

Thank you.

Also try running sudo fs_usage in Terminal. Every line printed is a disk read/write and since you would be on an older machine with slower I/O you need to reduce the number of lines printed.

One useful trick is to add folders to Spotlight's Privacy section e.g. source code, downloads, torrent locations, game files etc.

This is the post that provided the fix that worked for me: http://www.rdoxenham.com/?p=259

Had the exact same problem (minus the spilled tea, but I'm often running my computer very hot so I'm not surprised) on OS 10.8 on a 2011 (?) MBP, and did exactly this to fix it. I verified that the problem was with some sort of IntelACPI... [sic] module using a DTrace script (been a while but I can try to find it if anyone wants... basically that module was using up a large amount of CPU whenever there was load, such as when I launched a game. Long story short, the step-by-step to fix it is here:

http://www.rdoxenham.com/?p=259 (cached: http://webcache.googleusercontent.com/search?q=cache:UPanIg_... )

Been working like a charm for about 13 months now, and it's survived multiple minor 10.8.x upgrades since then. YMMV may vary with 10.9, and since they supposedly changed power management, I wouldn't touch that upgrade with a [Mac OS] 10 foot pole.

The fix for me was to simply delete the power profile from my MacBook model in a kext config directory. Everything was fine until I upgraded to 10.8.5, and then to my system would not boot at all – feeling with the kernel panic very early in booth that had something to do with AppleIntelCpuPowerManagement.kext.

I completely reinstalled and install Mavericks, and then repeated the original fix. It worked fine and my system is doing great running Mavericks.

Uh, "feeling with the kernel panic very early in booth" meaning, boot failed with a kernel panic.

I was kind of confused by what I read about power management having been changed in Mavericks, but it turned out that whatever is relevant to this issue has not changed. As I noted the same solution fixed the same problem.

Any other company that would DARE completely ignore the cries for help of its consumers would be publicly shamed and their reputation would suffer.

Yet, not a single word in this thread about how silent they tend to be about issues effecting thousands of users (which is a common trend with them)...

I wonder how they manage to pull that off, it's magic.

Really? Name one non open-source company that will jump on forum threads and fix it, rather than just putting it in their internal bug tracker to investigate.

I have a Logitech gaming keyboard. Their drivers are shockingly bad and freeze up every so often requiring me to re-plug the keyboard to get them to to work again. This is mentioned in the forums for at least 6 months, no fix has been announced.

I'm not defending Apple here, they should at least comment on their own forums, but they're not alone.

Back in the day, when there was no internet (only the arpanet) and Unix systems shipped with 10 manuals in nicely bound cases (I had one I was in my 20's at the time) things moved much slower and a vendor would never consider acting like this because they served primarily business customers (and institutions). Just the memory on my first Unix system was like $4000 a MB. It came with a 70mb hard disk.

To me the change most likely came around 1981 when the IBM PC came out and was sold to consumers who basically believed that it was them that was stupid and not something wrong with the machine. Manufacturers then learned that they could shove things into the pipeline and fix them later and that they actually got a fair amount of slack in the marketplace from the "stupid" end users. People who knew their shit using these machines were totally outnumbered by computer illiterate people. In other cases the ubiquitous "tech guy" made money because things didn't work so who was he to complain? That's what kept Microsoft in power. The tech guy ecosystem.

I'm curious if anyone else agrees at all with my take. Strictly my opinion having been around computers so many years.

It really depends on the number of customers.

If you have 1 million $10 customers and 10,000 have a problem the cost of losing their business isn't amazingly huge, especially as most of them probably won't switch vendors.

When you have ten $1,000,000 customers you might listen more if one has a problem.

Microsoft are much better than Apple at this, heck just look the latest fixed on connect.

http://connect.microsoft.com/VisualStudio/feedback/details/8...

I've had issues which I've posted in newsgroups and forums, that have been turned into bug tickets.

Some teams appear better at this than others mind.

Yet you can still do this and hang Visual Studio 2012, and maybe 2013:

* On a new machine (or maybe a new profile, I haven't checked), enable PowerShell's AllSigned option

* Open Visual Studio and open an MVC4 project, bonus points if it's close to the template in terms of installed NuGet packages

* Watch as Visual Studio hangs indefinitely.

This has been reported a number of times to Microsoft, but I don't think it was fixed in Update 4.

What's happening is PowerShell, running inside Visual Studio is prompting for something to do with a signed PowerShell Script, but Visual Studio hasn't loaded enough of the UI to let you accept the warning/prompt.

As a .NET developer, I can confirm this is the case with VS 2013.

I haven't been able to find one word from them about how they broke everyone's VPNs with Mavericks. It's fine though, I just consider it our just punishment for ever allowing Macs into a business environment.

Can you say a bit more about that? I have some 10.8.x machines that I was going to upgrade to 10.9.x in order to fix some VPN problems that I am having with that. What's up with Mavericks and VPN? (Links, commentary etc..)

I think he refers to the server VPN support from Mavericks server. It is broken for like half a year at least before Apple fixed it.

Mavericks (and server 3.0) was released on October 22nd. December 19th the update was released for their vpn issue. Less than two months does not equal "like half a year".

The issue was vpn behind nat wasn't working as expected, copying raccoon over from 10.8 server fixed the issue (not ideal but it took all of 5 mins).

Not only that, strangely people seem willing to pay premium prices to be handled like that.

With the prices Mac devices cost, I would expect zero percent failure rates with high attention to quality.

The so called Mac premium pricing (which isn't really true if you compare like for like) is for the superior hardware. Better trackpads, cases that don't flex, better keyboards, better overall design. The issue here is likely a software bug for an OS that was only recently released.

You would think someone on HN would know the difference.

Worst cables ever... so far for the 'superior' hardware.

Prices are at their fairest the moment they release new hardware. The later you buy their hardware the bigger the difference with their competitors (They only lower the price of older hardware when releasing a new model and not that much either).

But superior hardware, no, that is a bridge too far. Decent hardware, yes, but not all hardware from Apple is equal.

There's more hardware in a laptop than a CPU, RAM and a Hard Drive.

> The so called Mac premium pricing (which isn't really true if you compare like for like) is for the superior hardware.

If you wish I can also link to failure reports of the said superior hardware, like the recent GPU failures or the iPhone superior quality antenas.

When you buy a product you are paying for the whole package, not parts of it.

So if Apple intends to play the role of Ferrarri, Bang and Olufsen, Montblanc of the computer world, then their products should act accordingly.

Simple - a few thousand (if it is that many) people out of several million people having a problem is not always evidence of a real problem but rather proof that with a large enough install base some people will always have problems and it might well be for no reason at all beyond shear chance.

If there are 10 million installations of Mavericks (and I've got no idea how many there really are) and 5,000 people reporting a particular problem, that's only .05% of users - which is well within the expected proportion of users you'd expect to have problems with any upgrades. Note, for comparison the recent PS4 release problems and Sony's statement about the 0.4% of users experiencing problems - "This is within our expectations for a new product introduction"

Except you have no idea how many people are affected, you only know how many A) noticed, B) traced the culprit back to kernel_task, and C) reported it. That's likely a very small percentage of the afflicted population.

I'm also not sure how you explain a CPU pegged by a kernel process on a machine which should be doing almost nothing as "shear chance". That's a bug.

That could be because Apple's forums are not the venue for support. For developer support, there's rdar. For customer support, there's AppleCare. They state this pretty clearly, and there are perennial reminders when the iOS betas come out and the forums get flooded with non-devs who brick their iPhones.

Apple gets my repeat business (read: magic) by solving my hardware and software problems reliably and quick when I present them through the appropriate channels, go well over a decade now. For one particular repair, I was given a loaner machine, and another resulted in being updated to newer hardware - common stories amongst people who use their AppleCare.

rdar is very hit and miss - I've got bugs in there that are several years old.

Also unlike other companies all the bugs are hidden so you can't actually see whether other people have the issue, how they've solved it etc.

Apple is probably one of the most unfriendly companies to deal will as they seem obsessed with secrecy even when none is needed.

Is this fairly standard for Apple?

I recently picked up an old MBP and the logic board die on me. Turns out the faulty GPU unit was a known issue, but they never did a recall. Instead they extended the warranty portion for this part by four years.

This is the first real issue aside from iPhone battery woes a few years back.

The mmore I look through forums and such, the more I find Apple not shipping products that meet the market perception of a superiority.

That's nothing, look at this issue with the late 2011 mbps. 171 pages and apple "geniuses" say its not a known issue...

https://discussions.apple.com/thread/4766577

>Any other company that would DARE completely ignore the cries for help of its consumers would be publicly shamed and their reputation would suffer.

Completely ignore? Like several public statements, recall programs, warranty extensions and other such stuff they have done for problems in the past?

They have the bug in their radar bug tracking system, they close duplicate reports (which means they know about it specifically), and they are obviously working on solving it. What can they say before a fix is ready? "We're working on it"? Of course they are.

Perhaps you missed the fact that they top the user satisfaction ratings almost every year.

Try dealing with some other companies...

This, along with my wifi connection dying every hour or two (requiring a restart) makes me wish I had never upgraded to Mavericks.

Link to wifi discussion: https://discussions.apple.com/thread/5535320

Thanks for the link. I have the same Wi-Fi flakiness. :(

I've had this problem with all my Macs at home, although it would never occur with other Wi-Fi access points. This leads me to believe it has something to do with the wireless access point itself.

I was never able to fix the problem, but it went away after I bought an AirPort Time Capsule, which now serves as my wireless router. It has an extremely limited feature set compared to a real router, but I haven't had connection problems once since implementing it.

Do you use a bluetooth keyboard? My wifi flakiness seems to be much less of a problem when my lid is open and I'm typing on the builtin keyboard than when the lid is closed and I'm using the Apple bluetooth keyboard.

It's so intermittent that it's very hard to diagnose, but that's one thing I've noticed.

No, I don't have a bluetooth keyboard. On the Macbooks I use the built-in. On the iMac I use a wired keyboard.

The wifi problem drives me absolutely nuts. Most of the time I'm somewhere I can use a wired connection but the lack of a solution is incredibly frustrating.

In my opinion, Mavericks is the worst OS X version so far. It's caused a handful of crashes/hangs to my Macbook Pro, which I only encountered with about once or twice a year with Snow Leopard, Lion and Mountain Lion. I'm not sure it is caused by the OS itself or the compatibility of the apps I installed, but it's still bad.

I just had a crash few mins ago after I reinstalled the OS entirely last week. According to the crash log, it has something to do with the "kernel_task". That sucks!

Each major release of OS X, I know that I should wait until the x.1 release, but I get excited and install the x.0 release anyway. And every release, the same thing happens. Tons of bugs, crashes, and frustration and I swear that next release I'll wait until x.1.

I guess it would be a wise move. However, we are in 10.9.1 right now and the problem doesn't seem to be gone. Sighhh...

Man, am I just insanely lucky? I've never had such trouble with x.0 releases, during the upgrade or after.

FWIW, I've used at least 2 Macs concurrently since '09, and one since '06.

I guess.

I saw that most of the time people did complain about the first releases of the OS. I didn't get most of those problem, though, until Mavericks.

I'm with you. Probably most users are, too. People who aren't experiencing these problems have no need to inhabit threads like these.

> Mavericks is the worst OS X version so far

You have clearly never used Cheetah.

Sorry, I must have made it clearly. It's among the OS that I've used. I just have been using Mac for about 4 years.

As someone thats been on the osx train since 10.2. I get somewhat amused by the "OSX N+latest is the worst OSX" comments.

I had the (dis)pleasure of using someone's 10.1 machine, oy wow, surprised that even got released.

As a counterpart, mavericks has been the most stable osx for myself so far. Additionally the battery life improvements gave me an extra hour on battery alone. You can pry mavericks from my cold dead hands at this point. The energy tab is also really useful at finding out what is using up battery needlessly (looking at you chrome/firefox).

I feel jealous with your luck. I've had as many crash as I can imagine since switching to Mac from Windows (before, I don't even remember how many times I got freezing apps or BSOD). And it seems to occur that often since Mavericks so it's my sole assumption.

I agree about the battery life improvements part as a nomadic user. However, it's totally depends on which applications you are using. For me, chrome/firefox, which eats up most of the battery, is used constantly, therefore, cannot be turned off (to save energy, because it doesn't support AppNap, I suppose). I don't see any help looking at the Energy Tab as whatever it's like, I don't have a choice there.

Or OS X Public Beta! Wow, was that thing ever slow on my B&W G3!

So very low! Those were fun times tho, doing speed tests by seeing how quickly windows would open ;)

I think Snow Leopard (10.6) was the most stable OS X release.

Problems I have, on a brand new 15 retina MBPro: 1- after wake from sleep, wifi needs to be turned off and on to find anything

2- no sound from internal or external speaker until restart, although the OS behaves as nothing is wrong, the onscreen indicators and system preferences act as if they are muting and unmuting and changing the volume

3- kernel panics every now and then

4- display jitters (like graphic buffer corruption) a couple of times

5- after wake up from sleep , keyboard does not respond while track pad is, which means I cannot login, needs force shutdown,

"on a brand new 15 retina MBPro"

You have a warranty. Use it.

Complaints here do not solve your problems.

If Apple cannot solve your problems, within the terms if your warranty, they will offer alternatives (eventually) - up to and including a refund.

I understand people want to raise issues to "public awareness" that they feel are something Apple should pay attention to. And in the case of the original submission, VERY SPECIFIC details were figured out - and Bug Reported to Apple! Apple even closed some of the reports as a duplicate. They know about this problem - and it's for a new OS that has only been out for a handful of months - and the issue is for specific hardware builds.

They are working on it.

You, however, have a laundry list of real issues - USE YOUR WARRANTY.

Some of those (minor) problems, in fact most, are the standard kind of "new OS" bugs. None of them sounds like its about hardware faults. So your advice is not really applicable here. He'd better wait for 10.9.2 and such.

My advice is 100% applicable. Especially with issue #5, but in general as well.

You don't seem to understand: He is unhappy with what he purchased because he is experiencing problems that should not exist in the product. A brand new product. And not just a single issue - but several of them!

This isn't a device that (in his personal experience) worked fine at some point in the past on a different version of software but now is having problems on the latest software update.

This is a device that, brand new from Apple, has problems. This point is especially relevant as you cannot downgrade a Mac from Apple to an OS version that came out prior to the introduction of that specific model of machine. The Late 2013 MacBook Pro Retina devices shipped with Mavericks. They will never run Mountain Lion, you cannot avoid Mavericks on them for the time being.

This is the entire point of a warranty (and lemon laws in many locations): You don't have to "wait" for something you just bought to get better / work properly.

You take it back and show them the problems you're having.

Either they can fix it or they can't. If they can't fix it, he can get his money back or some other alternative offered by Apple that he might agree to (maybe a different model, or a complete replacement).

His money is not trapped in this device, so he doesn't have to put up with this experience - unless he actually prefers to complain and be unhappy.

>You don't seem to understand: He is unhappy with what he purchased because he is experiencing problems that should not exist in the product. A brand new product. And not just a single issue - but several of them!

Which product came with a brand new OS version. And which, like most OS launches has several software issues. Which are not gonna be solved if he returns his device (or only randomly, if they affect just one of a few Apple OEM partners chipsets etc).

>This is the entire point of a warranty (and lemon laws in many locations): You don't have to "wait" for something you just bought to get better / work properly.

That only holds if it's a faulty device. For OS and driver bugs, you very much have to wait. And no matter how long you wait, there will always be some bugs to, err, bug you, in it.

I must strenuously disagree. If these were software problems then online forums and Apple tech sites would be filled with folks tearing down the virtual walls. If you get a box with bizarre stuff going on and a search doesn't turn up a screaming horde with exactly your symptoms, then in all likelihood it's a local problem. That is a silver-lining to the decreasing post-purchase configurability of the Macbook Pro line: it makes it a heck of a lot easier to diagnose potential hardware issues.

Taken all together, these almost certainly point to bad hardware. It's possible that a bad third-party kernel extension might be the culprit as well, so testing in SafeBoot can be useful to ferret out such issues.

Last but not least, I'm personally making heavy use of the same hardware and OS daily with zero issues. I'm pretty sure it's not the Apple-supplied software.

>I must strenuously disagree. If these were software problems then online forums and Apple tech sites would be filled with folks tearing down the virtual walls. If you get a box with bizarre stuff going on and a search doesn't turn up a screaming horde with exactly your symptoms, then in all likelihood it's a local problem.

That a bug is a software bug doesn't mean that it manifests in all identical systems. It depends on various factors (ie. combination of third party peripherals, e.g some Macs have Samsung SSDs, others have other brand), install software, user workflow, etc.

Why does it matter whether it's a hardware or software problem? If it's broken it's broken.

Of course if it's a software problem he shouldn't expect a hardware replacement to fix it, but if he wants a refund he should get it.

>Why does it matter whether it's a hardware or software problem? If it's broken it's broken. Of course if it's a software problem he shouldn't expect a hardware replacement to fix it, but if he wants a refund he should get it.

It matters because software problems will always exist, there is no bug free software or OS, and the important ones get fixed in upcoming upgrades.

What would he do with his refund if it was a software problem? Switch to some magical unicorn non-issue OS or get another identical machine and OS?

I have a brand new 15inch Retina MacBook Pro. Have never had a single one of those issues.

I agree that he should seriously consider using his warranty and getting another machine.

thanks to all people who commented, I don't want my money back , I want a working MacBook, and it is not easy to bring this up to and apple store, since it does not happen all the time, I need to go there and try to reproduce it, and they I expect they say let's try a OS reinstall, then I'll go home and then it is likely that it does not solve some of the problems and .... it is not what I expect from what I consider best laptop (my) money can buy,

I also have a brand new 15" rMBP, and I have these exact same issues. I'll be returning mine for service next week.

> 5- after wake up from sleep , keyboard does not respond while track pad is, which means I cannot login, needs force shutdown

I have a similar problem sometimes and it appears to be some kind of USB bus / driver issue. If you plug in a USB keyboard, wait for it to be recognized, then unplug it the built in keyboard almost always starts working again. I've seen USB related error messages show up in logs when this happens.

Before you put the laptop to sleep unplug all USB devices and make sure the keyboard and trackpad are functioning correctly.

Holy cow, get warranty service ASAP. As someone with a brand new 15" rMBP, I do not ever have any of these problems. My system is absolutely rock solid in heavy daily use.

This must be somehow hardware related. I went through 5 rMBPs, because I got bad screens 5 times in a row. There is an entire 600 page thread on that as well. That being said, none of the 5 had this problem. Does it do the same thing with a clean install of OSX?

You can clone your current installation to a USB hard drive and install OSX from scratch. Carbon Copy works well, or just use dd. If it does the same with a clean install, take it to Apple, and make them deal with it.

I've had no problems with my MBP for the last 2+ years. Happy user here.

No overheating?

Nope, it gets as hot/noisy as I feel is appropriate for the amount of work it's doing.

My fridge has been working for the last 10 years, Happy user here.

guys there's a fix. i have a 2013 mba. the problem is in applehda. you can look at the hpet shoot up after sleep using `powermetrics`

here's a video to reproduce the issue

http://youtu.be/Q8OqdMq98j4

and here's a dmg that has the fix. basically it replaces applehda with the one from 10.8.5

http://puu.sh/5RBvj.dmg

i've been using it happily for ten days. some people claimed it wouldn't work with haswell, but it does for me. YMMV though

it doesn't work for rmbp 2013 late.

i think the latest mbp's might have a different product id on the intel hda. does your sound disappear with the old driver?

yep. but it solved the kernel 100% problem.

Workaround:

- First. stop playing music (in your browser, itunes or whatever).

- Next wait 5 minutes

- Last, close lid (go to sleep) and open again again.

Works for me.

My biggest problem is that my mac freezes once a day.

So, Apple installed PulseAudio?

Closing the lid (sleeping) is what is causing it for many people (myself included). I just have to leave it open or reboot.

My Mac freezes a few times a week. Just the window manager, because I can still ssh in. I know it isn't the lid, because it is an iMac, but it may be going to sleep that kills it.

Does your computer get hot ?

Because the only cause of Mac's freezing are pretty serious hardware issues.

i posted a real fix below. sadly this fix the one you mentioned doesn't always work for me. also, usually i wouldn't notice that the bug happened until my battery was half drained

Another (much more) common issue which seems to be related to Mavericks is the Finder problem where clicking Finder results a sudden hang of the system a the spinning colour wheel.

I'm fairly new to Apple products (~2 years ago) this is the worst problem so far but I'm surprised at such little feedback from Apple on it.

Not Mavericks-specific, but Finder is notorious for hanging when mounted volumes become unavailable or unreliable. And it's certainly possible that some of the network filesystem changes in Mavericks — in particular, it now uses SMB in certain cases where previous versions would have defaulted to AFP — could lead to this behavior in situations where you didn't see it before.

Solution on a late 2013 MBP for me (when the issue occurs no sound is playing anymore - no system sounds, no videos, no audio): Plugging in earphones and unplugging. Killing the coreaudio process as recommended at a few places online doesn't do the trick, unfortunately.

I've been wrestling this as well, and found this post too. Another problem I frequently wrestle with is my air sleeping on not coming back (the keyboard stays lit though). Only way to get it back is too reboot.

I experienced this right out of the gate after upgrading. I was very curious to see what my battery savings would be and decided to ditch my charger, only to have my battery gone in an hour. It's since subsided– I'm not really sure why, though at one point I did about everything I could think of short of reinstalling. I removed all kernel extensions I had added (KR4MB for example), removed SIMBL, reset PRAM, cycled the battery, etc. Now the only issue I have is about 10s after closing the lid, the fan goes nuts for another 10s before going to sleep.

Edit: Spring 2011 MacBook Pro.

My personal list of gripes (1st-gen 15" MBPr) is short, but it's sad that they all appeared after the Mavericks upgrade:

1. wifi will fail to reconnect on wake.

2. almost 40% of RAM (about 6 GB of 16) in use right after boot with no apps running. Was about 15% before Mavericks.

3. I got the dreadful Keychain error once, luckily managed to fix it (I use FileVault, so no SafeBoot for me and losing the key would have been fairly disastrous).

All these very clearly appeared after I took the plunge and moved to 10.9.1. Somebody screwed up.

Same problems. Have to plug and unplug headphones. to fix kernel_task

It's related to the audio driver and is triggered most frequently when you close the laptop with headphones in. Simply unplug them, close the laptop, and wait for 5-10 seconds. reopen... you will still see the cpu spike, but it will end within a few seconds of everything awaking... it is annoying, but this seems to resolve it for now.

I gave up on Mavericks after it went on an irreparable rampage of asking me for a "keychain password", then after trying a fix it just wouldn't boot anymore.

It required me to recover using Time Machine (and had a lot of workarounds, because it can't recover booting from the recovery partition in the HD, sigh)

I had the same issue with the keychain and repaired permissions on files to fix the issue.

Yeah, this might have helped.

The solution I tried involved changing some files in /Library/Keychains, but it didn't work

And the worse thing is that after that, a disk scan (using the Recovery boot) showed some errors on the HD, but these were non-recoverable

If you boot into recovery drive and go under Utilities -> Terminal type 'resetpassword' and in that same window there will be a ACL ( OS X Access Control Lists ) reset for an account. If you do that when you reboot, it prompts you to make a new keychain before logging you in.

I've been running into this problem with my 2013 MBA. It happens on wake, and the only that fixes it is a reboot. I'm getting pretty fed up with the issue since it's tanking my battery life and there seems like nothing I can do to fix it permanently.

Unrelated to this particular issue, but lately Adobe Updater has taken to running lsof in a tight loop, waiting for me to exit Acrobat so it can update. Makes for a pretty solid red bar of system CPU use.

On mine, unplugging the external screen for a while fixes everything http://www.youtube.com/watch?v=_kppD1PotFw

That explains why my Mac has been crazy sluggish lately...

I noticed slowdown on my brand new Air with nothing installed (beside a browser, xcode, and VLC). I have to reboot multiple times per day, closing the lid doesn't help, it causes it. Really unacceptable and a lot of people are seeing the same thing.

I'll be regretting not going with a Lenovo until they fix this.

Could also be other issues unless you see this specific set of symptoms. Mine gets sluggish but it seems to have something to do with the 'systemstats' process. When I open up something that queries it (e.g. click on the power icon in the menu bar to make it go to "Collecting power usage information...", or open Activity Monitor), it goes nuts, pegging the CPU and ballooning to >2GB of memory. Haven't figured out what to do about that, besides kill -9'ing it and avoiding doing anything that'll cause it to respawn.

Reply to self, in case anyone comes across this:

Moving the /private/var/db/systemstats directory out of the way and rebooting seems to have solved the problem. Guess something there was corrupt or too large.

One key thing that I've noticed consistently is this happens when I have my headphones plugged in. Unplug them, put it back to sleep and wake it.

I had an issue with mavericks on an old early 2008 MBP - - mavericks nearly cooked the computer on a fresh install.

This bug appeared for me after the update to 10.9.1 (from 10.9) on my 2013 MBA.

yup, also have this bug. It remedies itself after a minute or two.

Maybe you're holding it wrong. #antennagate

Antennagate? You mean that old non-issue blown up by some media out of all proportion, that happened to phones before and since (including competitors phones), and despite which "issue" the iPhone 4 went on to sell further tens of millions of devices with the same design?

No, I think unlike that, this is a real problem.

man, you guys/gals have no sense of humour

Reply on news.ycombinator.com