advertisement
Forums

The Forum is sponsored by 
 

AAPL stock: $156.07 ( -2.66 )

*Cached every 60 seconds. For live updating, Click Here

You are currently viewing the Tips and Deals forum
Wow - never, ever had this happen before (OS ? crash)
Posted by: volcs0
Date: December 31, 2007 02:31AM
I was working in 10.5 - doing a lot of complex tasks at once - R/BioConductor, Excel with huge spreadsheets, small window with a quicktime movie playing, Textmate, iTunes, some other stuff.

All of a sudden, the screen goes white and a few seconds later, I get my desktop back, but with every application neatly closed - just like a restart. How very weird. I'm sure I lost some stuff, but I think most things were saved.

I don't think I lost anything, but I can say that I've never had anything like that happen before.
Options:  Reply • Quote
Re: Wow - never, ever had this happen before (OS ? crash)
Posted by: iaJim
Date: December 31, 2007 03:24AM
Do you have plenty of memory? Strange.
Options:  Reply • Quote
Re: Wow - never, ever had this happen before (OS ? crash)
Posted by: GGD
Date: December 31, 2007 03:25AM
The console log might give some clues as to what happened.
Options:  Reply • Quote
Re: Wow - never, ever had this happen before (OS ? crash)
Posted by: Panopticon
Date: December 31, 2007 07:40AM
Just a long shot, but check your Expose & Spaces Preferences.
Options:  Reply • Quote
Re: Wow - never, ever had this happen before (OS ? crash)
Posted by: ztirffritz
Date: December 31, 2007 08:12AM
Quote
Panopticon
Just a long shot, but check your Expose & Spaces Preferences.

I never would have thought of that one, but yeah. If you're not used to working with multiple desktops that would be confusing. It really screws with windows users when I show them my Ubuntu desktops at work.



**************************************
MacResource User Map: [www.zeemaps.com]#
Options:  Reply • Quote
Re: Wow - never, ever had this happen before (OS ? crash)
Posted by: volcs0
Date: December 31, 2007 10:19AM
4 gb RAM in this MBP C2D.

It is not an Expose/Spaces problem.

Here is the Console log. As you can see, every application just quit. And when it was done (only took a few seconds), all of my startup apps started up again. It was really weird. Like someone flipped a switch and quit every application. Like I said - I've never seen anything like this, all the way back to my Apple ][ days.

-------

12/31/07 2:18:13 AM Finder[117] HIToolbox: received notification of WindowServer event port death.
12/31/07 2:18:13 AM R[10350] HIToolbox: received notification of WindowServer event port death.
12/31/07 2:18:13 AM Chicken of the VNC[10138] HIToolbox: received notification of WindowServer event port death.
12/31/07 2:18:14 AM org.x.X11[7608] 2007-12-31 02:18:13.943 X11[7609:10b] HIToolbox: received notification of WindowServer event port death.
12/31/07 2:18:14 AM org.x.X11[7608] 2007-12-31 02:18:14.361 X11[7609:10b] port matched the WindowServer port created in BindCGSToRunLoop
12/31/07 2:18:13 AM Preview[2299] HIToolbox: received notification of WindowServer event port death.
12/31/07 2:18:14 AM com.apple.launchd[78] ([0x0-0x202202].com.apple.print.PrinterProxy[6387]) Exited: Terminated
12/31/07 2:18:14 AM com.apple.launchd[78] ([0x0-0x163163].com.apple.Terminal[3926]) Exited: Terminated
12/31/07 2:18:14 AM com.apple.launchd[78] ([0x0-0x35a35a].com.geekspiff.chickenofthevnc[10138]) Exited: Terminated
12/31/07 2:18:14 AM com.apple.launchd[78] ([0x0-0x85085].com.apple.TextEdit[2373]) Exited: Terminated
12/31/07 2:18:14 AM com.apple.launchd[78] ([0x0-0x30b30b].com.macromates.textmate[9647]) Exited: Terminated
12/31/07 2:18:14 AM com.apple.launchd[78] ([0x0-0x25025].com.apple.AppleSpell[1849]) Exited: Terminated
12/31/07 2:18:14 AM com.apple.launchd[78] (com.apple.UserEventAgent-Aqua[101]) Exited: Terminated
12/31/07 2:18:14 AM com.apple.launchd[78] (com.apple.pboard[102]) Exited: Terminated
12/31/07 2:18:14 AM com.apple.launchd[78] ([0x0-0x38e38e].org.mozilla.firefox[10436]) Exited: Terminated
12/31/07 2:18:14 AM com.apple.launchd[78] ([0x0-0x11011].com.apple.iTunesHelper[127]) Exited: Terminated
12/31/07 2:18:14 AM com.apple.launchd[78] ([0x0-0x10010].com.proteron.liteswitch[126]) Exited: Terminated
12/31/07 2:18:14 AM com.apple.launchd[78] ([0x0-0xf00f].com.Growl.GrowlHelperApp[125]) Exited: Terminated
12/31/07 2:18:14 AM com.apple.launchd[78] ([0x0-0x2f72f7]."com.Adobe.ESD.AdobeUpdaterApplication"[9534]) Exited: Terminated
12/31/07 2:18:14 AM com.apple.launchd[78] ([0x0-0xc00c].com.apple.finder[117]) Exited: Terminated
12/31/07 2:18:14 AM com.apple.launchd[78] ([0x0-0x39c39c].com.apple.quicktimeplayer[10515]) Exited: Terminated
12/31/07 2:18:14 AM com.apple.launchd[78] (com.apple.helpdata[4387]) Exited: Terminated
12/31/07 2:18:14 AM com.apple.launchd[78] (com.apple.Spotlight[100]) Exited: Terminated
12/31/07 2:18:15 AM com.apple.launchd[78] ([0x0-0xb00b].com.apple.systemuiserver[116]) Exited: Terminated
12/31/07 2:18:15 AM com.apple.launchd[78] ([0x0-0x14014].com.apple.dock[131]) Exited: Terminated
12/31/07 2:18:15 AM com.apple.launchd[78] ([0x0-0x301301].com.microsoft.DatabaseDaemon[9602]) Exited: Terminated
12/31/07 2:18:15 AM com.apple.launchd[78] ([0x0-0x73073].com.apple.Preview[2299]) Exited: Terminated
12/31/07 2:18:15 AM com.apple.launchd[78] ([0x0-0x384384].org.R-project.R[10350]) Exited: Terminated
12/31/07 2:18:15 AM org.x.X11[7608] meta mod is 4
12/31/07 2:18:15 AM com.apple.launchd[78] ([0x0-0x12012].com.bytetaxi.foldershare[128]) Exited: Terminated
12/31/07 2:18:15 AM com.apple.launchd[78] (at.obdev.LittleSnitchUIAgent[96]) Exited: Terminated
12/31/07 2:18:15 AM com.apple.launchd[78] ([0x0-0x300300].com.microsoft.Excel[9600]) Exited: Terminated
12/31/07 2:18:15 AM com.apple.launchd[1] (com.apple.WindowServer[63]) Exited abnormally: Segmentation fault
12/31/07 2:18:16 AM com.apple.launchd[78] (org.x.X11[7608]) Stray process with PGID equal to this dead job: PID 7609 PPID 1 Xquartz
12/31/07 2:18:13 AM firefox-bin[10436] HIToolbox: received notification of WindowServer event port death.
12/31/07 2:18:13 AM Terminal[3926] HIToolbox: received notification of WindowServer event port death.
12/31/07 2:18:13 AM TextEdit[2373] HIToolbox: received notification of WindowServer event port death.
12/31/07 2:18:13 AM Finder[117] port matched the WindowServer port created in BindCGSToRunLoop
12/31/07 2:18:13 AM Chicken of the VNC[10138] port matched the WindowServer port created in BindCGSToRunLoop
12/31/07 2:18:13 AM Preview[2299] port matched the WindowServer port created in BindCGSToRunLoop
12/31/07 2:18:13 AM R[10350] port matched the WindowServer port created in BindCGSToRunLoop
12/31/07 2:18:13 AM firefox-bin[10436] port matched the WindowServer port created in BindCGSToRunLoop
12/31/07 2:18:13 AM Terminal[3926] port matched the WindowServer port created in BindCGSToRunLoop
12/31/07 2:18:13 AM TextEdit[2373] port matched the WindowServer port created in BindCGSToRunLoop
12/31/07 2:18:13 AM SystemUIServer[116] Error (268435459) creating CGSWindow
12/31/07 2:18:13 AM SystemUIServer[116] Error (268435459) creating CGSWindow
12/31/07 2:18:13 AM SystemUIServer[116] Error (268435459) creating CGSWindow
12/31/07 2:18:13 AM PrinterProxy[6387] HIToolbox: received notification of WindowServer event port death.
12/31/07 2:18:13 AM PrinterProxy[6387] port matched the WindowServer port created in BindCGSToRunLoop
12/31/07 2:18:14 AM SystemUIServer[116] HIToolbox: received notification of WindowServer event port death.
12/31/07 2:18:14 AM SystemUIServer[116] port matched the WindowServer port created in BindCGSToRunLoop
12/31/07 2:18:14 AM TextMate[9647] HIToolbox: received notification of WindowServer event port death.
12/31/07 2:18:14 AM TextMate[9647] port matched the WindowServer port created in BindCGSToRunLoop
12/31/07 2:18:14 AM GrowlHelperApp[125] HIToolbox: received notification of WindowServer event port death.
12/31/07 2:18:14 AM GrowlHelperApp[125] port matched the WindowServer port created in BindCGSToRunLoop
12/31/07 2:18:14 AM Spotlight[100] HIToolbox: received notification of WindowServer event port death.
12/31/07 2:18:14 AM Spotlight[100] port matched the WindowServer port created in BindCGSToRunLoop
12/31/07 2:18:14 AM Dock[131] HIToolbox: received notification of WindowServer event port death.
12/31/07 2:18:14 AM Dock[131] port matched the WindowServer port created in BindCGSToRunLoop
12/31/07 2:18:14 AM Little Snitch UIAgent[96] HIToolbox: received notification of WindowServer event port death.
12/31/07 2:18:14 AM Little Snitch UIAgent[96] port matched the WindowServer port created in BindCGSToRunLoop
12/31/07 2:18:14 AM X11[7609] HIToolbox: received notification of WindowServer event port death.
12/31/07 2:18:14 AM X11[7609] port matched the WindowServer port created in BindCGSToRunLoop
12/31/07 2:18:14 AM QuickTime Player[10515] HIToolbox: received notification of WindowServer event port death.
12/31/07 2:18:14 AM QuickTime Player[10515] port matched the WindowServer port created in BindCGSToRunLoop
12/31/07 2:18:14 AM LiteSwitch X[126] HIToolbox: received notification of WindowServer event port death.
12/31/07 2:18:14 AM LiteSwitch X[126] port matched the WindowServer port created in BindCGSToRunLoop
12/31/07 2:18:24 AM com.apple.launchd[1] (com.apple.UserEventAgent-LoginWindow[10601]) Exited: Terminated
Options:  Reply • Quote
Re: Wow - never, ever had this happen before (OS ? crash)
Posted by: TheCaber
Date: December 31, 2007 11:09AM
The WindowServer process died. The console log entries are from all the processes depending on the the WindowServer (child processes, or those communicating with the WS) receiving an event notification that the WS has died. Most of those processes in turn proceed to kill themselves (all the launchd messages in the middle part of the excerpt), since they have no way to reestablish access to a window server to do their display updates.

Look at the crashreporter log (/var/log/crashreporter.log) and the windowserver log (/var/log/windowserver.log). You can view them with the Applications->Utilities->Console.app log viewer, or with TextEdit, or (in a Terminal) with vi .



=TC
Options:  Reply • Quote
Re: Wow - never, ever had this happen before (OS ? crash)
Posted by: volcs0
Date: December 31, 2007 11:58AM
I don't see a crashreporter.log in /var/log

Here is the windowserver.log.

I'm suspicious of R/BioConductor (using X11) or my VNC client, both of which were running...

Thanks again for the help.


Dec 31 02:16:59 [63] kCGErrorIllegalArgument: CGXGetWindowWorkspace: invalid window id: 4bd0d
Dec 31 02:18:17 [10572] Server is starting up
Dec 31 02:18:19 [10572] CGXMappedDisplayStart: Unit 0: no display alias property
Dec 31 02:18:19 [10572] CGXMappedDisplayStart: Unit 1: no display alias property
Dec 31 02:18:19 [10572] CGXPerformInitialDisplayConfiguration
Dec 31 02:18:19 [10572] Display 0x2b38484d: MappedDisplay Unit 1; Vendor 0x10ac Model 0xe008 S/N 861424204; online enabled (0,0)[1680 x 1050], base addr 0x1800000
Dec 31 02:18:19 [10572] Display 0x42719c0: MappedDisplay Unit 0; Vendor 0x610 Model 0x9c67 S/N 0; online enabled built-in (-1440,0)[1440 x 900], base addr 0x1800000
Dec 31 02:18:21 [10572] GLCompositor: GL renderer id 0x00022604, GL mask 0x00000003, accelerator 0x00002827, unit 0, caps 0x00000003, vram 256 MB
Dec 31 02:18:21 [10572] GLCompositor: GL renderer id 0x00022604, GL mask 0x00000003, texture units 4, texture max 8192, viewport max {8192, 8192} extensions 0x00000007
Dec 31 02:18:33 [10572] kCGErrorFailure: CGXDisableUpdate: UI updates were forcibly disabled by application "Finder" for over 1 second. Server has re-enabled them.
Dec 31 02:18:33 [10572] kCGErrorFailure: Set a breakpoint at CGErrorBreakpoint() to catch errors as they are returned
Dec 31 02:24:40 [10572] kCGErrorIllegalArgument: CGXGetWindowWorkspace: invalid window id: 1fb
Dec 31 02:25:14 [10572] kCGErrorIllegalArgument: CGXGetWindowWorkspace: invalid window id: 230
Dec 31 02:25:16 [10572] kCGErrorIllegalArgument: CGXGetWindowWorkspace: invalid window id: 24c
Dec 31 02:34:16 [10572] kCGErrorIllegalArgument: Failure: UCKeyTranslate returns -50
Dec 31 02:35:11 [10572] kCGErrorIllegalArgument: Failure: UCKeyTranslate returns -50
Dec 31 02:35:56 [10572] kCGErrorIllegalArgument: CGXGetWindowWorkspace: invalid window id: 524
Options:  Reply • Quote
Re: Wow - never, ever had this happen before (OS ? crash)
Posted by: volcs0
Date: December 31, 2007 12:04PM
I found the crashreporter.log

Here is a truncated version (without all the binary image stuff).

Process: WindowServer [63]
Path: /System/Library/Frameworks/ApplicationServices.framework/Frameworks/CoreGraphics.framework/Resources/WindowServer
Identifier: WindowServer
Version: ??? (???)
Code Type: X86 (Native)
Parent Process: launchd [1]

Date/Time: 2007-12-31 02:18:10.797 -0600
OS Version: Mac OS X 10.5.1 (9B18)
Report Version: 6

Exception Type: EXC_BAD_ACCESS (SIGSEGV)
Exception Codes: KERN_PROTECTION_FAILURE at 0x00000000bebb7460
Crashed Thread: 0

Thread 0 Crashed:
0 ...ple.CoreServices.CarbonCore 0x938afbbd UCKeyTranslate + 363
1 com.apple.CoreGraphics 0x93d799ab CGSUniCodeForKeyAndFlags + 152
2 com.apple.CoreGraphics 0x93cf45ce CGXCheckForHotKey + 499
3 com.apple.CoreGraphics 0x93cc14c9 sPostContinuation + 1760
4 com.apple.CoreGraphics 0x93cc06c1 CGXFilterEvent + 71
5 com.apple.CoreGraphics 0x93cc09ce CGXProcessHIDEvent + 157
6 com.apple.CoreGraphics 0x93cc0929 sHIDContinuation + 537
7 com.apple.CoreGraphics 0x93cc06c1 CGXFilterEvent + 71
8 com.apple.CoreGraphics 0x93cc03fb CGXGetNextEvent + 148
9 com.apple.CoreGraphics 0x93cc032c CGXRunOneEventPass + 48
10 com.apple.CoreGraphics 0x93cc25b9 eventNotificationHandler + 106
11 com.apple.CoreGraphics 0x93cc2532 CGXPostPortData + 158
12 com.apple.CoreGraphics 0x93cc2328 CGXRunOneServerPass + 556
13 com.apple.CoreGraphics 0x93cbebfb CGXRunOneServicesPass + 377
14 com.apple.CoreGraphics 0x93cc983a CGXServerLoop + 161
15 com.apple.CoreGraphics 0x93c59019 CGXGetRootAdminCredentials + 0
16 WindowServer 0x00001ff4 main + 24
17 WindowServer 0x00001fbe start + 54

Thread 1:
0 libSystem.B.dylib 0x902418e6 mach_msg_trap + 10
1 libSystem.B.dylib 0x902490dc mach_msg + 72
2 com.apple.CoreGraphics 0x93cad15e eventThread + 3369
3 libSystem.B.dylib 0x90272075 _pthread_start + 321
4 libSystem.B.dylib 0x90271f32 thread_start + 34

Thread 2:
0 libSystem.B.dylib 0x90248ace __semwait_signal + 10
1 libSystem.B.dylib 0x90272ced pthread_cond_wait$UNIX2003 + 73
2 libGLProgrammability.dylib 0x9502df32 glvmDoWork + 162
3 libSystem.B.dylib 0x90272075 _pthread_start + 321
4 libSystem.B.dylib 0x90271f32 thread_start + 34

Thread 3:
0 libSystem.B.dylib 0x90248ace __semwait_signal + 10
1 libSystem.B.dylib 0x90272ced pthread_cond_wait$UNIX2003 + 73
2 com.apple.CoreGraphics 0x940ac41d layer_blit_thread + 229
3 com.apple.CoreGraphics 0x940ac942 layer_blit_async_thread + 95
4 libSystem.B.dylib 0x90272075 _pthread_start + 321
5 libSystem.B.dylib 0x90271f32 thread_start + 34

Thread 0 crashed with X86 Thread State (32-bit):
eax: 0x2ca87460 ebx: 0x93cf43ef ecx: 0x00000030 edx: 0x92130000
edi: 0x2ca87400 esi: 0x00000001 ebp: 0xbfff6cd8 esp: 0xbfff6c80
ss: 0x0000001f efl: 0x00010213 eip: 0x938afbbd cs: 0x00000017
ds: 0x0000001f es: 0x0000001f fs: 0x00000000 gs: 0x00000037
cr2: 0xbebb7460
Options:  Reply • Quote
Re: Wow - never, ever had this happen before (OS ? crash)
Posted by: Article Accelerator
Date: December 31, 2007 12:20PM
First, update to 10.5.1.

Second, consider not running the following nasty/semi-nasty/redundant/non-native software:

- Chicken of the VNC
- Proteron LiteSwitch
- FireFox
- Adobe Updater
- Microsoft Database Daemon
- Folder Share
Options:  Reply • Quote
Re: Wow - never, ever had this happen before (OS ? crash)
Posted by: TheCaber
Date: December 31, 2007 12:34PM
SIGSEGV is a segmentation violation signal, usually caused by an attempt to access memory at an address not available to the process' thread. Common instances are dereferencing a NULL pointer value (0x0), or an address in low memory (in the range 0x0 to 0xFFFC), or an address way outside the physical or virtual memory address range (say, 0xFFFFFFFC).

Had the pointer been an odd address, or pointed to a location not properly aligned for the data type (e.g., 32-bit data access not on a longword boundary), then a SIGBUS (bus alignment error) would have been raised on a processor that didn't allow nonaligned memory accesses.

SIGSEGV usually indicates that a data structure containing pointer data has been overwritten by another thread or process; the crashing thread blithely picked up what it thought was a valid pointer and tried to use it.

Without access to the source code for the WindowServer, it will be difficult to guess where the error occurred.



=TC
Options:  Reply • Quote
Re: Wow - never, ever had this happen before (OS ? crash)
Posted by: volcs0
Date: December 31, 2007 02:11PM
Quote
Article Accelerator
First, update to 10.5.1.

I can't get any more 10.5.1 than 10.5.1, can I? Anyway, those malicious programs have been running for months without problems. And it has not happened again - I was just wondering about this specific crash - never seen anything like it.


TC: Thanks for that thoughtful analysis. I was doing a ton of heavy analysis with R - lots of 40-50mb gene expression files. It was during some of these analyses - with both processors pegged at 100% for a long time that the crash occurred. Anyway, I will keep an eye on it. Thanks for taking the time to think about this with me. I appreciate the feedback.
Options:  Reply • Quote
Sorry, only registered users may post in this forum.

Click here to login

Online Users

Guests: 183
Record Number of Users: 52 on November 20, 2014
Record Number of Guests: 847 on February 04, 2015