Why is ATH.exe using 50% of my CPU?

The new 11.0.2 didn't solve anything either, still the same problem.

Bumping to second zappe's notion

I had ATH.exe using (or attempting to use) over 100% of my CPU earlier. If I wouldn't have walked into the room and caught it then it might have actually damaged my computer. Now I'm definitely turning off wifi sync on all devices to see if it fixes it.

Fix. This. Apple.

Or remove whatever (From whatever has been gathered so far it appears to be the WIFI sync feature) is causing it until you can make it stable.

I don't believe Sync over WiFi is the issue. I've got two Windows 7 Professional PCs running (workstation & laptop). The laptop never has the problem (although I usually don't have iTunes open on the laptop, I do have it setup to Sync with an iPhone 5 and an iPad 3 over WiFi). The workstation is usually running iTunes all day. I've had up to 100% CP usuage issues with AppleMobileeviceService.exe on a regular basis. iTunes had a recent update and now ATH.exe also is causing CPU usage to climb to 100%. I've done the Apple full uninstall and reinstall (time consuming, and ultimately doesn't solve the issue). Now ATH.exe is partnering with AppleMobileeviceService.exe to drive CPU usage to 100%.

I've regularly stopped the AppleMobileeviceService.exe process without any noticable problems. In the past, that has stopped the CPU usage problem until a reboot. Now, ATH.exe cause the problem without AppleMobileeviceService.exe running.

Calling Apple support has NOT resolved this issue since it appeared over a year ago.

I don't understand why the problem occurs on the Workstation that is NOT using WiFi to sync and does NOT appear on the laptop that IS using WiFi to sync.

I also don't understand why Apple hasn't resolved this issue yet.

It's possible there are two different bugs involved here, then, since for virtually everyone else the high CPU usage only happens when WiFi sync is on.

But FWIW, I'm not convinced even the WiFi symc issue is Apple's fault. Earlier in the thread I mentioned having similar issues with Polkast. I also have CPU issues with Splashtop, although only while it is actually in use. What these apps have in common is that they allow remote devices to access resources on the computer. I would be willing to bet at this point the issue is actually in a DLL or some such that is shared by all three of these services, and it is entirely possible it's a Microsoft issue.

One of the things that has convnced me this is not coincidence is that when I do leave WiFi sync on for a while, and get myself into a situation where ATH starts hogging the CPU, then I kill ATH, I find that Polkast *immediately* takes over as a CPU hog. Which suggests to me there is really some resource they share, and as soon as one process finally relinquishes it, the other takes it over. That's not the only possible explanation, of course, and it is true that Polkast does have specific "iTunes integration" built in so it could be that Polkast is really waiting on iTunes itself, not on some shared resource.

But I think the shared resource - such as a DLL - is seeming more and more to be to be the most likely culprit.

The common resources is possible, but more likely than a shared DLL is that the two apple EXEs are the resources being used. Except for the fact that my Workstation has NO WiFi capability, it's on my LAN via ethernet cable. While my Laptop (no issues with the two Apple EXEs and excessive CPU usuage) is on my LAN via WiFi.

This is an Apple issue, otherwise there would also be high CPU usage by other EXEs shown in Task Manager.

Alright so if anyone is still having this issue lets do try some things.

Does this occur in new user? yes or no.

Does it occur if you were to follow support.apple.com/kb/ht1275

Let me know with the results

Now I'm not so sure where the ATH.exe problem is coming from. ATH.exe starts up when i start iFunbox.

Neil Cook wrote:

The common resources is possible, but more likely than a shared DLL is that the two apple EXEs are the resources being used.

I did a hang dump analysis of ATH and AppleMobileDeviceService. The call stack of both indicated that the offending routine lives in Common Files\Apple Application Support\CoreFoundation.dll. I sent all this information along with the dumps to Apple's bug reporting portal, for all the good that's done us.

Is there a straightforward way to explain how to do this hang dump analysis? I'd be happy to do the same for Polkast next time it does its CPU-hogging thing.

A straight forward answer could be LSP's on the comptuer which is Layerd service providers Apple has an article that explains itunes performance issues such as -

Symptoms

You may notice the following symptoms on your Windows PC after installing the iCloud Control Panel, iTunes 10.5 or later, or Safari 5.1 or later:

• iTunes is slow to respond.
• Windows is slow to respond.
• iTunes stops responding when backing up or syncing with an iPhone, iPad, or iPod touch.
• Some processes installed by iTunes are using up a lot of resources in the Windows Task Manager.
• Connecting an iOS device causes iTunes to quit unexpectedly.
• Safari will not load any websites.
• iTunes Store looks empty and white as shown below:

If you view Article - http://support.apple.com/kb/TS4123

It will explain how to remove potential LSPs that could be causing the issue - I've had this issue before if you have issues along the way feel free to ask questions

What Apple neglects to tell you is what to do if the only LSA's running on your machine are Apple. There are numerous threads about this, Apple provided the article you mention some time ago although it shows updated in October, 2012, and the problem persists. No one has resolved the issue successfully with this article that I have found. Apple did provide a response and I don't expect them to address it again regardless of whether that fix works for everyone or not. This issue does not affect Apple users, only Windows users so Apple has no interest in fixing it.

Creating the dump file is simple. Just go to the task manager, right click on the hung process, and pick "Create dump". (This works on Windows Vista and 7, not sure about 8). I'm going to assume you're running a 64 bit OS.

Getting the analysis of the resulting dump is the more complicated part. You first need to install the Debugging Tools for Windows.

http://msdn.microsoft.com/en-us/windows/hardware/gg463009.aspx

Run the x64 version of WinDBG. Go to File->Open a Crash Dump and select the dump file you created.

You'll see something like:

wow64cpu!CpupSyscallStub+0x9:

00000000`73152e09 c3 ret

and a prompt waiting for you to input something. The "wow64cpu" is an indicator that you're debugging a 32 bit application and will get nonsensical data unless you switch modes. So enter the following command:

!wow64exts.sw

The next command is kind of a whopper:

~*k

This dumps the call stacks for every thread in the application. You'll see all sorts of stuff. Most of them will have "WaitForSingleObject" or some sort of "Wait" at the top - this is normal and means the thread is gracefully waiting for a signal to do something. The suspicious thread looks like this:

0207f91c 72a07ec3 ntdll_76eb0000!RtlLeaveCriticalSection+0xd

0207f984 74bb7423 mswsock!WSPGetSockOpt+0x97a

0207fc5c 72e33702 ws2_32!getsockopt+0x159

WARNING: Stack unwind information not available. Following frames may be wrong.

0207fc80 72e357a0 CoreFoundation!CFSetCreateMutableCopy+0x2f4

0207fd8c 76ee389a CoreFoundation!CFSocketGetDefaultNameRegistryPortNumber+0x9f8

0207fe9c 76ee9ec5 ntdll_76eb0000!RtlpFreeHeap+0xb7a

0207feb4 00000000 ntdll_76eb0000!_RtlUserThreadStart+0x1b

I see the same "CreateMutableCopy" function in the snapshots I took of ATH and AppleMobileDeviceService while they were eating CPU. It could be executing this in a tight loop - it's hard to say what's going on really, I don't have their sources.

Jahooba wrote:

Now I'm not so sure where the ATH.exe problem is coming from. ATH.exe starts up when i start iFunbox.

That's simply because iFunbox relies on iTunes to establish the WiFi connection with your iDevice. If you have the ATH problem you'll see it whether you use iTunes or iFunbox.

And I agree with yraegel, there are many of us who don't have any other LSP's running. The absolute only one is Apple on my machine. Their KB trying to blame it on third party resources is lame at best. Their inattention to something that has been around as a customer complaint for eons however is worrisome and indicative of their attitude.

Currently I have a bat file I simply run to stop/start the service which cures it but whereas I used to have to run this maybe once every few weeks, with the latest iTunes update it has become an every day affair each morning I log into my desktop (telltale sign is the cooling fan is whirring away). I'm getting to the point I am going to create a Task to run that bat file every few hours.

Well.. I also have this ATH problem... my advice is to close iTunes when not using it, the ATH.exe is killed when you close iTunes. I'm using iTunes only for syncing my iPhone, if you use iTunes as a media player or for other stuff that requires it to run most of the time, then I suggest you find alternatives.

E-D@N wrote:

if you use iTunes as a media player or for other stuff that requires it to run most of the time, then I suggest you find alternatives.

Kind of defeats the purpose of being able to easily use home sharing of your music library and/or even easily sync. If I have to go open iTunes on my desktop each time I want to sync a device, I might as well just plug the thing in using the usb, no?

Sure there are alternatives out there to act as a music server, but in all honesty why should this be even considered to be a necessity? No Apple device is that cheap that one should be having to look for other software to have the device interact using a reasonable amount of resources. I realize Apple's iTunes is free, but their devices sure as heck aren't.

All it really requires is for Apple coders to make sure ATH is killed as a running service once it is no longer being used. Is that so hard or unreasonable?