[GRASSLIST:2741] Re: i.rectify and image resolution

[ulf at mail.zmt.uni-bremen.de]

I tried again with 

   watch -n1 "df -k" 

but disk space does not seem to be the problem -- none of the involved
disk partitions reached more than 85% during i.rectify.

I do not want to exclude the temp file hypothesis because during
several tries with i.rectify, I got twice an error message about
temp files. But in all other cases I got the following
"success"-message: 

   ***********************************************
   Rectify [braganca at umehlig] (LOCATION braganca_img)
    into  [brag1986 in umehlig] (LOCATION braganca_image)
   complete
   -----------------------------------------------
    2000 rows, 2000 cols (4000000 cells) completed in 0:30
    8000000.0 cells per minute

Thanks, Ulf

p.s.: I have compiled grass with something like "gcc-3.0 -O2
      -march=pentiumpro" ... Should I re-compile i.rectify without
      optimization?

 > I temporarily had this problem: reason was that the file system 
 > was full. Can you do us a favor and retry, and watch the disk
 > free status?
 > You may use
 > watch df -k
 > for that purpose (if watch is installed on your machine).
 > Noter, the temp files are stored in the LOCATION, not in /tmp!
 > 
 > Unfortunately i.rectify doesn't send an error if disk full occurs.
 > But it silently cleans up after finishing - so the user doesn't 
 > realize the problem.
 > 
 > In my case a cleanup of unused stuff helped. However, please try
 > again before freeing space - this is the chance to track down
 > the bug/problem (either to confirm above idea or to point out that it is a
 > different problem).
 > 
 > Thanks in advance,
 > 
 > Markus
 ---------------------------------------------------------------------
 > That behaviour may be caused by your disk filling up during the 
 > rectification. I.rectify will then produce an incomplete cover that might 
 > cause a segfault when you try to display it (At least it did for me...) As 
 > far as I have understood, a change in this behavoiur is one of the things 
 > to go into the final grass 5.0.
 > 
 > Morten

-- 
=======================================================================
 Ulf Mehlig    <ulf.mehlig at zmt.uni-bremen.de>
               Center for Tropical Marine Ecology/ZMT, Bremen, Germany
-----------------------------------------------------------------------