R help - Aug 2004 - Rcmdr X11 protocol error message

I am using R 1.9.1 with R Commander GUI under Windows at work,  and under Linux
Mandrake 10 OR at home.   I am having no problems running R Commander under
windows.  Under Linux, though, the opening and sometimes closing of windows from
R Commander produces identical and repetitive error messages (in dialogue boxes
and mirrored in the console):
>"Warning: X11 protocol error: BadWindow (invalid Window
parameter)"  X 10-20.
These error messages occur most frequently when graphics windows are opening,
but can also occur when dialogue boxes pop up, and sometimes even when results
are written to the output window.  However, when the error message is
"OK'd", the plot, or whatever, is displayed as it should be.  It
is not entirely consistent, and will sometimes stop happening for parts of a
session.  Exactly the same commands run directly from the R console do not
produce the same problem.  I searched the mailing list archives, and found the
following exact description of what I have been experiencing, from November
2003:

> Message-id: <200311221950.36723.korponai at georgikon.hu>
> 
> Hi all,
> I've got this using Rcmdr package only.
> R-cmdr> scatterplot.matrix(~PC1+PC2+PC3+PC4 | Species,
> reg.line=FALSE, smooth=FALSE, span=0.5, diagonal= 'density',
> by.groups=TRUE, data=iris)
> Warning: X11 protocol error: BadWindow (invalid Window parameter)
> Warning: X11 protocol error: BadWindow (invalid Window parameter)
> Warning: X11 protocol error: BadWindow (invalid Window parameter)
> Warning: X11 protocol error: BadWindow (invalid Window parameter)
> Warning: X11 protocol error: BadWindow (invalid Window parameter)
> Warning: X11 protocol error: BadWindow (invalid Window parameter)
> Warning: X11 protocol error: BadWindow (invalid Window parameter)
> Warning: X11 protocol error: BadWindow (invalid Window parameter)
> Warning: X11 protocol error: BadWindow (invalid Window parameter)
> Warning: X11 protocol error: BadWindow (invalid Window parameter)
> Warning: X11 protocol error: BadWindow (invalid Window parameter)
> Warning: X11 protocol error: BadWindow (invalid Window parameter)
> Warning: X11 protocol error: BadWindow (invalid Window parameter)
> 
> What's this? The plot is shown.
 
This is exactly the error message I get, with similar 10-20 line repetition. 
And the plots are generated anyway.  Unfortunately, I could not find any reply
to this email.
 
My system is running R 1.9.1 on Linux Mandrake 10 OR (installed from source, not
RPM), with XFree86 4.3, P4 2.4GH, 1GB Ram.  The problem occurs in several
different window managers/desktop environments.  I'm wondering if it is
something to do with the interaction of tcl/tk with the particular X11 config,
but I've reached the limit of my skills in diagnosing.  If anyone can
suggest anything, I would appreciate it.


Regards,

Michael Bibo

Research Officer,
Community Health Services,
West Moreton Health Service District,
Queensland Health.

michael_bibo at health.qld.gov.au 




***********************************************************************************
This email, including any attachments sent with it, is confi...{{dropped}}

"Michael Bibo" <Michael_Bibo at health.qld.gov.au> writes:
> I am using R 1.9.1 with R Commander GUI under Windows at work, and
> under Linux Mandrake 10 OR at home. I am having no problems running
> R Commander under windows. Under Linux, though, the opening and
> sometimes closing of windows from R Commander produces identical and
> repetitive error messages (in dialogue boxes and mirrored in the
> console):
> 
> >"Warning: X11 protocol error: BadWindow (invalid Window
parameter)"  X 10-20.
> 
> These error messages occur most frequently when graphics windows are
> opening, but can also occur when dialogue boxes pop up, and
> sometimes even when results are written to the output window.
> However, when the error message is "OK'd", the plot, or
whatever, is
> displayed as it should be. It is not entirely consistent, and will
> sometimes stop happening for parts of a session. Exactly the same
> commands run directly from the R console do not produce the same
> problem. I searched the mailing list archives, and found the
> following exact description of what I have been experiencing, from
> November 2003:
It's a bit hard to debug without actually installing Rcmdr (and all
its dependents), but this sort of message usually means either that a
window was invalid to begin with (a bit unlikely) or that it
disappeared while events were still being generated for it.

-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907