Skip to content

help difficulties on

4 messages · Ben Bolker, David Winsemius, Simon Urbanek +1 more

Original
Ben Bolker
# 
Dear R-sig-Mac'ers:

  I'm posting on behalf of a student who is having trouble with
accessing the help files in R.app (I think) on MacOS X.4 (I think
X.4.11?) .  The basic symptom is that in a clean R session she gets
results like this:
Error in file(out, "wt") : cannot open the connection

  This is generic (i.e. happens with any help request).  After a bit
more poking around we triggered

2010-02-09 15:20:21.010 R[16728] *** Exception handlers were not
properly removed. Some code has jumped or returned out of an
NS_DURING...NS_HANDLER region without using the NS_VOIDRETURN or
NS_VALUERETURN macros.
R(16728,0xa000ed88) malloc: *** error for object 0x3b610038:
Non-aligned pointer being freed
R(16728,0xa000ed88) malloc: *** set a breakpoint in szone_error to
debug
R(16728,0xa000ed88) malloc: ***  Deallocation of a pointer not
malloced: 0x41d60; This could be a double free(), or free() called
with the middle of an allocated block; Try setting environment
variable MallocHelp to see tools to help debug

  We have both tried searching/googling for this particular problem,
without success -- found a fair amount about issues with help.start()
when running firewalls, but nothing that seemed to match our situation.

  * Does this ring any bells for anyone? Have we missed something
obvious (I hope)?  What are the next steps in the
diagnostic/troubleshooting pathway?

  * Is there a searchable version of the R-sig-mac archives?
Rsitesearch doesn't seem to offer it, and the R-sig-mac archives seem to
be available only month-by-month (& not searchable), as far as I can tell.

  thanks,
    Ben Bolker
R version 2.10.1 (2009-12-14)
powerpc-apple-darwin8.11.1

locale:
[1] en_US.UTF-8/en_US.UTF-8/C/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods
[7] base

loaded via a namespace (and not attached):
[1] tools_2.10.1
David Winsemius
#
On Feb 10, 2010, at 4:44 PM, Ben Bolker wrote:

            

      
      
      
    

        
You can dummy up a search function for the r-sig-mac archives by using  
the Google advanced search with the domain name restricted to:

https://stat.ethz.ch/pipermail/r-sig-mac/

http://www.google.com/search?hl=en&as_q=hekp+problems&as_epq=&as_oq=&as_eq=&num=10&lr=&as_filetype=&ft=i&as_sitesearch=https%3A%2F%2Fstat.ethz.ch%2Fpipermail%2Fr-sig-mac%2F&as_qdr=all&as_rights=&as_occt=any&cr=&as_nlo=&as_nhi=&safe=images

I'm pretty sure that Gmane also archives r-sig-mac and I noticed that  
is what the BioConductor page refers people to for their list.

            

      
      
      
    

        

      
      
    

        
David Winsemius, MD
Heritage Laboratories
West Hartford, CT

  
  


  


      

    

    

      
      

        


  

        

      Simon Urbanek
    

    

      
      #
    

  


  

      
On Feb 10, 2010, at 4:44 PM, Ben Bolker wrote:

        

            

      
      
      
    

        
It seems likely a problem in your system setup -- possibly your temporary directory has wrong permission and/or TMP/TEMP settings are wrong. That is a long shot though, please use something like options(error=recover) to dump the call stack and look at the "out" value to see where is the problem on the machine.

            

      
      
      
    

        
That is completely unrelated - do you have a way to reproduce that?

Thanks,
Simon

            

      
      
      
    

        
options(error=recover)

enter the last frame and have a look at the value of out -- that will tell you where on the machine is R trying to create a file that fails. Also the frame stack may be useful in case it's a deeper error.

            

      
      
      
    

        
http://blog.gmane.org/gmane.comp.lang.r.mac

In your case I doubt you'll find much as this seems likely a peculiarity of your machine.

            

      
      
      
    

        

      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Marc Schwartz
    

    

      
      #
    

  


  

      
On Feb 10, 2010, at 3:44 PM, Ben Bolker wrote:

        

            

      
      
      
    

        
Ben

I'll defer to Simon on the debugging part, as he has already engaged, other than to say, if this is a permissions issue, you can run the OSX Disk Utility:

  Applications -> Utilities -> Disk Utility

and run the Repair Disk Permissions process (on the First Aid tab) on the disk to see if that helps.

With respect to generally searching the list archives, I use:

  http://www.rseek.org

which includes the R-SIG-* lists in addition to r-help and r-devel.

HTH,

Marc Schwartz