Lion update 10.7.2 - Dock eating memory and CPU

[vandaveer]

I updated to OSX 10.7.2 yesterday. Since then, if I launch two VM at the same time, and ONLY then, the Dock will start eating CPU and memory. The memory usage will keep climbing until it takes every last available amount of memory. The system will come to a complete crawl. Killing Dock causes it to relaunch of course and then it starts all over. It does give me a temporary reprieve however until it can reach max available memory. I'm running on a MBP with 8 GB of RAM, a hybrid SSD main HD, and my VM's off of a secondary HD in the optibay. I'm wondering if anyone else is seeing this issue or has a solution?

[admin]

Please post the logs here or if you don't want them in the public forum then put them on something like dropbox and PM me a link. Thanks for your help

[jcon]

I booted my VM around 10:00 this morning and it started eating CPU and memory around 2:00.

[Arlo515]

Interesting-...I launched around 9am, and it started to spike around noon. I wonder if its a time factor?

[benad84]

Exact same problem here (10.7.2, 4.0.2). I'll PM khogan the log file.

[jcon]

Same problem today, happened much quicker too, happened about an hour after starting up today.  I can usually tell when it starts spiking because keystrokes start to  repeat themselves in the VM.

[jcon]

Here are my logs around the time when it happened.  Could this have anything to do with Time Machine starting?  I think mine kicked in right around the time the problem started today. 

BTW, if you pause the VM, kill the dock and start the VM back up, it seems to work for a while.

10/25/11 10:49:13.449 AM com.apple.usbmuxd: _SendAttachNotification (thread 0x7fff79fd5960): sending attach for device 04****._apple-mobdev._tcp.local.: _GetAddrInfoReplyReceivedCallback matched.

10/25/11 10:49:13.804 AM usbmuxd: _AMDeviceConnectByAddressAndPort (thread 0x100781000): IPv4

10/25/11 10:49:13.965 AM iTunes: _AMDDeviceAttachedCallbackv3 (thread 0x10f5ed000): Device 'AMDevice 0x7fdc0377fe90 {UDID = ****, device ID = 264, FullServiceName = ****_apple-mobdev._tcp.local.}' attached.

10/25/11 10:51:35.227 AM com.apple.mtmd: attempting to thin because of low free space on Macintosh SSD (/) by removing 2011-10-25 13:50:35 +0000

10/25/11 10:51:35.000 AM kernel: nspace-handler-set-snapshot-time: 1319554297

10/25/11 10:52:18.366 AM GeekTool Helper: Geeklet 545F6D91-82AB-4D7F-BBE6-172ACC48C89D. Task already running (date +%l:%M)

10/25/11 10:52:19.345 AM GeekTool Helper: Geeklet 545F6D91-82AB-4D7F-BBE6-172ACC48C89D. Task already running (date +%l:%M)

10/25/11 11:05:12.286 AM mds: (Error) FMW: WE ARE DROPPING FMW EVENTS!

10/25/11 11:05:12.313 AM coreservicesd: FMOD WATCH EVENTS DROPPED!

10/25/11 11:05:12.430 AM 1PasswordAgent: reloadAllObjects

10/25/11 11:05:12.514 AM coreservicesd: FMOD WATCH EVENTS DROPPED!

10/25/11 11:05:12.514 AM mds: (Error) FMW: WE ARE DROPPING FMW EVENTS!

10/25/11 11:05:12.516 AM coreservicesd: FMOD WATCH EVENTS DROPPED!

10/25/11 11:05:12.570 AM mds: (Error) FMW: WE ARE DROPPING FMW EVENTS!

10/25/11 11:05:12.739 AM 1PasswordAgent: Database (AGHtmlDatabase:file://localhost/Volumes/Space/Dropbox/1Password.agilekeychain/) load time [Cache]: 0.251+0.016 (339 objects)

10/25/11 11:05:13.000 AM kernel: add_fsevent: event queue is full! dropping events (num dropped events: 270153; num events outstanding: 2583).

10/25/11 11:05:13.000 AM kernel: add_fsevent: kfse_list head 0xffffff80b1792c10 ; num_pending_rename 1512

10/25/11 11:05:13.000 AM kernel: add_fsevent: zalloc sez: 0

10/25/11 11:05:13.000 AM kernel: add_fsevent: event_zone info: 4096 0x0

10/25/11 11:05:13.000 AM kernel: add_fsevent: watcher 0xffffff8099509008: num dropped 0 rd 5892 wr 5892 q_size 8192 flags 0xc

10/25/11 11:05:13.000 AM kernel: add_fsevent: watcher 0xffffff8099529008: num dropped 0 rd  444 wr  444 q_size 1024 flags 0x1

10/25/11 11:05:13.000 AM kernel: add_fsevent: watcher 0xffffff8099731008: num dropped 0 rd 5732 wr 5732 q_size 8192 flags 0x4

10/25/11 11:05:13.000 AM kernel: add_fsevent: watcher 0xffffff809a285008: num dropped 48351 rd  160 wr  159 q_size 4096 flags 0x9

10/25/11 11:05:13.188 AM mds: (Error) FMW: WE ARE DROPPING FMW EVENTS!

10/25/11 11:05:13.187 AM coreservicesd: FMOD WATCH EVENTS DROPPED!

10/25/11 11:05:13.193 AM coreservicesd: FMOD WATCH EVENTS DROPPED!

10/25/11 11:05:13.194 AM mds: (Error) FMW: WE ARE DROPPING FMW EVENTS!

10/25/11 11:05:13.212 AM mds: (Error) FMW: WE ARE DROPPING FMW EVENTS!

10/25/11 11:05:13.212 AM coreservicesd: FMOD WATCH EVENTS DROPPED!

10/25/11 11:05:14.598 AM coreservicesd: Received request to reset fmod watch. Latest received id is 241183016615608. Latest sent id is 241195833059060

10/25/11 11:05:14.980 AM mds: (Error) FMW: WE ARE DROPPING FMW EVENTS!

10/25/11 11:05:14.980 AM mds: (Error) FMW: WE ARE DROPPING FMW EVENTS!

10/25/11 11:05:14.980 AM mds: (Error) FMW: WE ARE DROPPING FMW EVENTS!

10/25/11 11:05:14.980 AM mds: (Error) FMW: WE ARE DROPPING FMW EVENTS!

This last set of messages repeats some 4000 times in a short period of time

[jcon]

Just happened again.  It went through several time machine backups without causing the dock issue, but my 3:00 backup seems to correspond to the start of the dock issue.

[HendieDijkman]

It's still happening on my system (Late 2007 MacBook Pro, 2.4Gh Dual Core, 4Gb Memory, Lion 10.7.2). Since my first post on this forum, I have completely reinstalled Fusion, created a brand new Windows 2007 Prof VM. The Dock still goes crazy like everyone else's when I hit a certain magical memory consumption.  I have a feeling this could actually be an Apple problem, with the Dock maybe not being able to handle being swopped out to disk? Fusion still works through all of this, but of course being the biggest memory consumer probably contributes to the symptoms.

[admin]

We are investigating what is causing this issue. One thing I have found that may work is if you disable application sharing this issue will not happen. If anyone wants to test this and post back the results it would be great. I have not been able to reproduce this issue yet so if anyone knows any more info on what I should do to make this happen please let me know.

If someone could test each setting and see if turning off one specific option stops this it would be great. As of right now I just know that turning off all these otions stoped this issue from happening for one customer I spoke with.

1) select Virtual Machine > Settings > Default Applications

2) turn off all 3 options (if someone could test one at a time to see if its a specific option I would really appreciate it)

3) click show all

4) select Applications menu

5) select Never for Show Applications in the menu bar (not sure if this does make a difference, would like to hear some reports)

6) now reboot your Mac and see if the issue still occurs.

[Arlo515]

Thanks for taking a serious look at this issue.

For my part, unfortunately, I alrready have all those settings unchecked as you recommended, so it may just have been luck or some other unseen variable that fixed the other user.

Edit: Is Application sharing something separate from those prefs? All I note is "Shared folders" under Sharing, is that what you mean? Should I be removing all my shared /mirrored folders to test?

[Arlo515]

It does feel like random things trigger it...sadly nothing repeatable or obviously connected.

[AdrielH]

I've had the same problem since the update and following this thread closely.... I'm tied to using Visual Studio (arghh) in VMWare so this has been a productivity killer, and like everyone else here I have been experimenting with all the solutions offered.

I noticed early this morning that the Dock started running away on me after I interrupted the OS X screensaver to get back to work. I set the screensaver delay to "never", quit VMWare, quit the Dock, relaunched VMWare and only experienced one runaway situation the rest of the day - quite an improvement over the typical runaway every hour or two in normal use. (The one runaway that did happen this afernoon came immediately after I hit "save" in TextEdit - the save dialogue never popped up and the Dock immediately started leaking memory and eating up processor cycles. I did the normal fix to get back to work: Quit VMWare, quit the Dock, relaunch VMWare. Not sure how it relates, and this one instance may have just been a fluke.)

Anyway, might want to try disabling screen savers to see if it helps. It does seem like an Apple bug, but something with VMWare is triggering the bug.

[kaz219]

I also never had anything checked under “Default Applications” and still, the problem also happened to me. It only happens after the VM has been running for more than one hour at least.

[jcon]

I also have all of the application sharing options turned off.  Does everyone share their home directory with the mac side?

[benad84]

Interresting. Turning off Shared Folders seems to have fixed the issue for me. I'm guessing that Shared Folders has an issue with the latest Spotlight service (which is partially tied with the Dock process).

[jcon]

I turned time machine OFF and started the VM back up, after a few hours, I had the dock problem again, so it doesn't sound linked to time machine

[benad84]

jcon, have you tried turning off Shared Folders? As you say, I don't think it's due to time machine since it's off for me too, yet turning off Shared Folders fixed the issue for me.

[jcon]

Trying that now, won't know for sure until some time tomorrow afternoon if it worked.

[vmswosh]

Tried the ones with turn off all options of "Default Applications" and "Never Show Applications"... did not work for me. Unfortunately this is very annoying...

[HendieDijkman]

Turning off shared folders has solved the problem for me as well. I have been running for two days now without any runaway Dock issue. Just sayin'