Crash Report

My apologies if I’m forgetting something simple. It’s past my bedtime, and I’ve noticed that I always seem to do something stupid if I’m around here late.

You let vic-k type your last chapter, didn’t you? He is the only one who could make that mess.

You forgot to tell me what you were doing that caused the crash - a crash report on its own is like telling me the answer is 42. :slight_smile:

Thanks and all the best,
Keith

Ah. Sorry. :blush:

Well, I’d had Scrivener open and written a little, then left my NaNo project open while I was in other programs like Sea Monkey (a browser), Opera (another browser), and iTunes. Oh, and TextEdit. After some browsing in Opera and typing in TextEdit, I decided I wanted to look up something in one of my projects, clicked over to Scrivener, opened one project from the “Recent Projects” menu, realized it was the wrong one, closed it, and opened the correct project—

Or, at least, I told Scrivener to open the correct project. It crashed. And when I opened it again, my NaNo project (which had been open until the crash) needed to have its search strings updated or whatever the word is. I also notice that the two vertical windows I had open were both empty of the documents I’d had them referring to, so I set that back up. (I’m guessing that’s related to the search strings thing.)

eyes crash askance

vic-k isn’t the only one, obviously. :unamused:

but the answer is 42. I don’t understand the problem.

Thanks. Your description, in conjunction with the startup information in the crash report, indicate a memory bug - the hardest type of bug to track down, unfortunately, as they can appear completely random. There’s a similar bug I’ve been experiencing in my 2.0 code recently but have yet to track down yet. The best thing to do is just to look out for it and if it happens again, see if you can think of any other specific triggers - or if you can get it to happen again. I’ll take a note of it and put it on my list of items to look try to reproduce and track. I hate memory bugs. They are an absolute beast to track simply because they are so hard to reproduce. But usually it will be one specific thing that you have done during the session - the trouble with memory bugs is that it could be something you did hours before.

Is there anything reported on the Console (~/Applications/Utilities/Console.app) around the time of the crash?

Thanks and all the best,
Keith

realizes she never said this before Um… This isn’t the first time I’ve had this happen. :blush: I think it may have something to do with either iTunes or Opera, but it might be the Sea Monkey.

I’ve copied everything from Console for that day through the crash.

studies it

What is Textsoap still doing on my computer? That was a demo. goes to delete it

ACCESS VIOLATION! And I am not talking about Mr K posting. As a matter of fact, nothing Mr K does, while likely to be a violation of some type of access, is not at question here. Which means you have made Mr B a pretty unhappy camper. :frowning:

Does this happen with only one project or is this an equal opportunity crash? If it is only one that would be good. Good in a relative way, but better than this being an EOC.

[size=75]While I am not a programmer by trade, I get to spend a lot of time explaining to random programmers why they need to keep track of how many elements are in an array, reminding them not to increment or decrement a pointer (unless they are a programming god and it is clear are NOT the afore mentioned god-like programmer or I would not be explaining this to them) and for the love of all that is good, to refrain from deallocating memory and then attempting to access it on the very next line.[/size]

Um, I may be wrong, but I think I remember this happening in my “Primpriety” project, too. This time was with my “FRtW” (NaNo 2009) project. So I think it’s more than one.

Can you find a specific sequence of events that always makes this happen?

It is a memory error. The problem there is that a memory error, as I say, will only crash the app long after the error itself occurred. So it’s a matter of looking out for common factors during the running of the program. Meanwhile, I’m running Scriv in a debug mode intended to trace such errors when they occur - but you can’t do that, unfortunately. So if you can just keep posting crash reports and all the things you did during the session until we can track a common factor, that would be very helpful.

Thanks!
All the best,
Keith

So after rereading my last post I want make a statement: My foot note is not directed at KB in any way. It is more an statement that these types of memory errors are not unique to Scriv, and that they are a PITA to chase down. The fact that this is maybe the third time this type of mem error has been reported is a testament to the quality of scriv.

Carradee, you are on a G4, right?

nods Yeah, I’m G4.

I have dates for when some of the other crashes occurred. Lemme see if they’re still in the Console. Nope. Too long ago.

I’ll post the Console report next time it happens.

'nother crash:

Scrivener’s been open much of the day. I’ve been browsing, listening to iTunes, using Excel, etc. Specifically in Scrivener I just finished a section and clicked in the Binder to go to the next section in the outline when the crash happened.

Opening Scrivener back up did not display the most recently opened file, but one I was working on several hours ago (that was having a bit of annotation trouble), at one point, suddenly changing the color for annotations after I’d changed it to “Blueberry”.

Closing that other file and opening the one I was working on most recently gets the “Resynchronize search strings” window, and it doesn’t look like I lost anything except what file I was in.

Comment: I have an iBook, not a PowerBook, nevermind what that summary says.

Here are the Console errors. I think the first one is from when I was having some trouble importing a site as a web archive.

11/14/09 5:27:10 PM Scrivener[165] *** -[NSCFArray objectAtIndex:]: index (-1( or possibly larger)) beyond bounds (3) 11/14/09 10:37:18 PM DEVONthink Pro[2014] Custom ColorPicker class with name .com.freeverselib could not be loaded. 11/14/09 10:37:33 PM mdimport[2017] Paths: id(501) ( "/Developer/Applications/Xcode.app/Contents/Library/Spotlight/SourceCode.mdimporter", "/Library/Spotlight/Microsoft Office.mdimporter", "/System/Library/Spotlight/RichText.mdimporter", "/System/Library/Spotlight/Archives.mdimporter", "/Library/Spotlight/iWork.mdimporter", "/System/Library/Spotlight/Font.mdimporter", "/Users/Misti/Applications/Devonthink Pro/DEVONthink Pro.app/Contents/Library/Spotlight/DEVONthink Spotter.mdimporter", "/System/Library/Spotlight/Audio.mdimporter", "/System/Library/Spotlight/MIDI.mdimporter", "/Library/Spotlight/AppleWorks.mdimporter", "/System/Library/Spotlight/PS.mdimporter", "/System/Library/Spotlight/Application.mdimporter", "/System/Library/Spotlight/QuickTime.mdimporter", "/Applications/NeoOffice.app/Contents/Library/Spotlight/neolight.mdimporter", "/System/Library/Spotlight/Image.mdimporter", "/System/Library/Spotlight/Automator.mdimporter", "/System/Library/Spotlight/QuartzComposer.mdimporter", "/Users/Misti/Applications/Fresh.app/Contents/Library/Spotlight/OpenMetaSpotlight.mdimporter", "/System/Library/Spotlight/PDF.mdimporter", "/System/Library/Spotlight/Bookmarks.mdimporter", "/System/Library/Spotlight/iCal.mdimporter", "/Users/Misti/Applications/ProVoc.app/Contents/Library/Spotlight/ProVoc.mdimporter", "/System/Library/Spotlight/vCard.mdimporter", "/Users/Misti/Applications/Skim.app/Contents/Library/Spotlight/SkimImporter.mdimporter", "/System/Library/Spotlight/iPhoto.mdimporter", "/System/Library/Spotlight/SystemPrefs.mdimporter", "/Developer/Applications/Xcode.app/Contents/Library/Spotlight/uuid.mdimporter", "/System/Library/Spotlight/Chat.mdimporter", "/System/Library/Spotlight/Mail.mdimporter", "/Users/Misti/Applications/Adium.app/Contents/Library/Spotlight/AdiumSpotlightImporter.mdimporter", "/Users/Misti/Applications/Scrivener.app/Contents/Library/Spotlight/ScrivenerMDI.mdimporter" ) 11/14/09 10:57:59 PM com.apple.launchd[86] ([0x0-0x17017].com.literatureandlatte.scrivener[165]) Exited abnormally: Bus error 11/14/09 11:00:42 PM Scrivener[2098] Custom ColorPicker class with name .com.freeverselib could not be loaded. 11/14/09 11:00:42 PM Scrivener[2098] Custom ColorPicker class with name .com.freeverselib could not be loaded.

Did I avoid giving the answer w/o the question, this time?

Wait, you lost work? That shouldn’t be possible. Everything except the one file? Where is this project saved? Is it being synchronised at all?

There is definitely an error thrown on the console - an exception. Could you keep the console open as you work and look to see if you can recreate it? See if you can follow the steps you made in Scrivener leading up to the problem to see if you can generate that “out of bounds” error again?

More importantly, though, ANYTHING you can tell me about how you lost data, any information at all - your set up, how you are saving, everything - would be really appreciated.

Thanks and all the best,
Keith

I read Carradee as meaning that she only lost her place in the project, not any data. That when the project reopened after resynchronising, the caret was no longer where it was at the moment when Scrivener crashed. Am I right, Carradee?

Mark

D’oh! You’re absolutely right, Mark. I read this line:

as “it looks like I lost everything except what file I was in”. I should drink my coffee before replying in future - I answered while checking my mail as I was just waking up during my Sunday lie-in. That’s my excuse, anyway.

With that panic out of the way… I would say that the main issue is the “out of bounds” exception that is getting thrown. This could cause other problems along the way and may be leading indirectly to the crash. So if Carradee can reproduce this, that would be great.

Thanks and all the best,
Keith

Mark’s right. I lost my place, not any data. :slight_smile:

I’ll see if I can do that.

Out of interest, had you emptied the trash in any of these instances? I just saw a very similar bug myself after emptying the trash, although have yet to reproduce it. Argh!
Thanks,
Keith

I know I didn’t empty the trash the day of this latter one, and I’m pretty sure I didn’t the day of the first one, either.