[R-SIG-Mac] Crash when closing R-GUI

[Matthew Cohen]

I think I've found the cause of the crash.  I had set up my plist so  
that, by defauly, R-GUI would not ask me whether I want to save my  
workspace, but rather would assume I don't want to save.  Following  
Simon's e-mail of 7/17, I typed the following into the terminal window:

defaults write org.R.R-project.R save.on.exit -string no

This is apparently what is causing the crash.  Changing it back to  
"ask" makes it stop crashing.  Is it just me?

Matt

On Aug 17, 2006, at 6:51 PM, Rob J Goedman wrote:

> This did happen on my system a couple of weeks ago. Not sure if it  
> is the same issue.
> Does it ask you if it should save the workspace?
>
> Would you mind sending me your plist ( i.e. ~/Library/Preferences/ 
> org.R-project.R.plist)?
> Or you could rename/delete it and see if the crash still happens.
>
> Rob
>
>
> On Aug 17, 2006, at 3:05 PM, mcohen at mail.la.utexas.edu wrote:
>
>> When I try to close R and the R-GUI by clicking either the little  
>> red Aqua
>> button on the left or the "light switch" button on the right, R  
>> crashes, giving
>> this message:
>>
>>
>>  *** caught bus error ***
>> address 0x0, cause 'invalid alignment'
>>
>> Possible actions:
>> 1: abort (with core dump)
>> 2: normal R exit
>> 3: exit R without saving workspace
>> 4: exit R saving workspace
>>
>> Typing q() causes no problems at all.
>>
>> This happens with both the R-2.3.1.dmg on CRAN and the newest  
>> stable builds of R
>> and R-GUI at R.research.att.com.
>>
>> I'm running a iMac G5 with 10.4.7.  I've only noticed this  
>> behavior since R
>> 2.3.1.  That said, I only started occasionally using the R-GUI  
>> again since
>> 2.3.1.  I'm thinking I must be a problem local to my system, given  
>> that no one
>> else has mentioned it and it seems like something someone would  
>> have noticed
>> were it more widespread.  But I can't figure out what could be  
>> wrong with my
>> system, and I'm not even sure how to begin troubleshooting.  Any  
>> suggestions
>> would be greatly appreciated.
>>
>> _______________________________________________
>> R-SIG-Mac mailing list
>> R-SIG-Mac at stat.math.ethz.ch
>> https://stat.ethz.ch/mailman/listinfo/r-sig-mac