[R-SIG-Mac] Totally frozen Gui and Forced Quit does not resolve.

[Bryan Hanson]

Excellent Joe.  If you have further problems, there have also been discussions here in the past two months about the recent files list and an R.app plist which can be removed manually, forcing the app to write a fresh copy.  Bryan

> On Dec 29, 2016, at 11:10 PM, Joseph Kunkel <jkunkel1 at une.edu> wrote:
> 
> Bryan,  Following the directions in section 10.11 I removed every .RData and .Rhistory in my entire tree with no satisfaction.
> 
> Finally, after removing every .Rapp.history ... R failed again but I was able to exit R ungracefully using option 1. 
> 
> On rebooting, R-gui reopened with all the 3 prior R-scripts also opened but the R-console was not frozen and after closing the R-scripts I q() R and on reopening I had a normal R-gui and no R-scripts opened.
> 
> I had known about the .Rhistory and .Rdata files but they were not the problem.  The .Rapp.history was the culprit but trying to be logical and just removing it from the most obvious directories did not work.  I had to remove every single one.  
> 
> Perhaps a clue?  I was unable to remove two of the  ‘.Rapp.history' files using their file paths in a rm statement.  Looking closely at the path there was a gramatically-poor directory in both failed removes.
> One failure had ‘/text file/‘ which had an illegal space in the directory name and the other failure had another illegal character in a path component /1&2/.   After I removed those two  ‘.Rapp.history’ files I was able to break the loop.
> 
> Thanks for the URL with directions that included the solution!
> 
> In addition I will redouble my conscious use of good unix path names!  OS X allows navigating through paths that may be easier to read but are traps for unix command syntax.
> 
> Joe
> 
> 
>> On Dec 29, 2016, at 7:56 PM, Bryan Hanson <hanson at depauw.edu> wrote:
>> 
>> Take a look at section 10.3 and 10.11 in https://cran.r-project.org/bin/macosx/RMacOSX-FAQ.html#Miscellaneous-questions
>> 
>> I also remove .Rhistory routinely (in addition to the files listed in 10.11) when having trouble.
>> 
>> If you need assistance finding these files throughout your system using terminal, let us know.
>> 
>> Another alternative is to get R devel at http://r.research.att.com/ That version of the GUI appears to have some fixes.
>> 
>> Bryan
>> 
>>> On Dec 29, 2016, at 7:46 PM, Joseph Kunkel <jkunkel1 at une.edu> wrote:
>>> 
>>> Dear R-Sig-Mac,
>>> 
>>> I have not received any response to my request for help in unlocking my R-gui from its infinite loop.
>>> 
>>> To restate it simply.
>>> 
>>> 1) I was running the newest version of R as of Dec 19th including the latest R-gui.
>>> 2) I needed to force-quit R with three R-scripts open.
>>> 3) On rebooting the R-gui the three scripts are reopened and the R-console is in its active mode without me being able to get its attention or close it without another force quit.
>>> This is an infinite loop that I have not been able to break.
>>> 
>>> I can run R from a terminal window without the conveniences of the gui but as I have proceded from Dec 19 I have discovered that some of the conveniences of the R-gui are truly valuable and I would like to get back to a state where I can use the R-gui.
>>> 
>>> Is there anyone back from the Holidays yet who can tell me the minimum I need to do to break the infinite loop?
>>> 
>>> Re installing R from CRAN did not work.  There is some log that I need to erase to stop the system from recreating the problem each time I reboot R.
>>> 
>>> Joe
>>> 
>>>> On Dec 19, 2016, at 9:29 AM, Joseph Kunkel <jkunkel1 at une.edu> wrote:
>>>> 
>>>> Using the R Gui a ’new' unsolvable situation has arisen.  I have been using R almost daily since 2002 acording to my records.  This is the first time I have not been able to rectify an R crash. 
>>>> 
>>>> I am running R version 3.3.2 (2016-10-31) — “Sincere Pumpkin Patch”  on Mac Os Sierra Ver 10.12.1 MacBook Pro early 2013 2.7 GHh Intel Core i7, 16 GB 1600 MHz DDR3
>>>> 
>>>> With three R-scripts in their editor windows and one R-script actually invoked I needed to force a quit using the Activity Monitor.
>>>> 
>>>> Then, when the R Gui was re-invoked all the previous Gui windows opened (as with previous experiences),  but this time R was still in process with activity monitor showing R as non-responsive and rainbow wheel spinning whenever a Gui window is hovered.
>>>> 
>>>> 'Activity monitor' or ‘Dock' force quit does not resolve the above issue.  Each time R Gui is reinvoked all windows are restored to the R (Non Responding) state.  Activity monitor shows 2/3 of a core CPU devoted to R with ~125 Mb of memory devoted to R.
>>>> 
>>>> Previously the R Gui R-script windows could be closed and when R was closed and reopened ‘normal’ R-Gui opening could be invoked.
>>>> 
>>>> *** I have deleted R.app from the Applications Folder and reinstalled R version 3.3.2 (2016-10-31) — “Sincere Pumpkin Patch” but seemingly the residual logs extant at the last ‘force quit’ still exist and invoking the R-gui re-establishes the R (Non Responding) state.
>>>> 
>>>> Terminal R and RStudio still work but I would rather use the R Gui, which had been working well for me in the latest versions.
>>>> 
>>>> Can I identify 'the residual logs’ of the R-gui and erase or reset them to a 'fresh start’ state?
>>>> 
>>>> I do not know where to go from this impasse other than using one of my other lower grade computers or wiping my computer and reinstalling the Mac Os.
>>>> 
>>>> Is there a link to directions for a more serious cleaning of R files and logs before I reinstall R?  I would guess that the R link in the R.app Contents directory might invoke more deeply embedded R files that need to be scrubbed/reset?
>>>> 
>>>> Joe Kunkel
>>>> 
>>>> -·.  .· ·.  .><((((º>·.  .· ·.  .><((((º>·.  .· ·.  .><((((º> .··.· >=-       =º}}}}}><
>>>> Joseph G. Kunkel, Research Professor
>>>> 112A Marine Science Center
>>>> University of New England
>>>> Biddeford ME 04005
>>>> http://www.bio.umass.edu/biology/kunkel/
>>>> 
>>> 
>>> _______________________________________________
>>> R-SIG-Mac mailing list
>>> R-SIG-Mac at r-project.org
>>> https://stat.ethz.ch/mailman/listinfo/r-sig-mac
>> 
>