<div dir="ltr">i think posting to ncl-talk was not entirely wrong since there is an aspect of ncl (and cdo) in it. thankfully Rick's answer is confirmed as due to memory problem. After passing six files unmerged, the program ran and was killed after 30% execution. Before it was zero execution.<div>Thanks very much all.</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Thu, Aug 22, 2019 at 8:18 PM Dennis Shea <<a href="mailto:shea@ucar.edu">shea@ucar.edu</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div><b>perl</b>/<b>CDO</b>/<b>python</b> : <span style="color:rgb(0,0,255)"><b>No NCL</b></span>. Why is the question posted to <b>ncl-talk</b>?</div><div>--------------------------------------------------------------------------------------------<br></div><div>[1] re: "running icclim (a <b>python library for index calculation</b> from CNRM-CERFACS-France) fails, giving a message killed" <br></div><div><br></div><div><div>Rick's answer is correct.</div></div><div><br></div><div>Maybe posting a python based question to ?stackexchange? or CERFACS would be best?</div><div><br></div><div>[2] Certainly, ncl-talk people are not readily familiar with "several RCMs". What RCM's?<br></div><div><br></div><div>When this type of question is posted to any group, you should include [attach] the output from (say) 'ncdump -h' both a sample source and the '"merged" (concatenated) file. Otherwise, people are just guessing. <br></div><div><br></div><div>merging the files with nco "ncrcat <a href="http://file1.nc" target="_blank">file1.nc</a> ... <a href="http://file6.nc" target="_blank">file6.nc</a> <a href="http://output.nc" target="_blank">output.nc</a>" is worse.</div><div><br></div><div>Maybe try the NCO 'ncrcat' operator. This requires that 'time' be "UNLIMITED"<br></div><div><br></div><div>%> <b>ncrcat </b>-O file[1-6].nc <a href="http://output_ncrcat.nc" target="_blank">output_ncrcat.nc</a></div><div><br></div><div>Good Luck<br></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Thu, Aug 22, 2019 at 9:01 AM Rick Brownrigg via ncl-talk <<a href="mailto:ncl-talk@ucar.edu" target="_blank">ncl-talk@ucar.edu</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div>Hi Zilore,</div><div><br></div><div>I don't know the answer to your specific science question(s), but if icclim is terminating with a message that simply says "killed", then that's a Linux system's very unhelpful way of saying it ran out of memory entirely. Are you indeed running under Linux and are the data you are processing rather large?</div><div><br></div><div>Rick</div><div><br></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Thu, Aug 22, 2019 at 7:42 AM Zilore Mumba via ncl-talk <<a href="mailto:ncl-talk@ucar.edu" target="_blank">ncl-talk@ucar.edu</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr">I have five yearly data from several RCMs, i.e 19510101 to 19551231, ... 20050101 to 20101231. These are tasmax, tasmin and precip. I want to work on climate indices on these files, but only interested in the period 19710101 to 20001231. I have written a perl script which combines six files with the command "cdo <a href="http://file1.nc" target="_blank">file1.nc</a> <a href="http://file2.nc" target="_blank">file2.nc</a> ... <a href="http://file6.nc" target="_blank">file6.nc</a> <a href="http://outputfile.nc" target="_blank">outputfile.nc</a>".<br>The perl scrip seems to work fine.<br>However:<br>1. ncdump -h combined file gives the right number of times in the file, 10958 365x30+8 leap years)<br>2. display in ncl displays the map ok<br>3. A check on values at specific times shows they are the same as in the original file<br>But:<br>running icclim (a python library for index calculation from CNRM-CERFACS-France) fails, giving a message killed<br>running icclim with an original file calculates the index ok.<br><br>My request for assistance is whether anybody has experienced such a problem and how it can be resolved.<br>merging the files with nco "ncrcat <a href="http://file1.nc" target="_blank">file1.nc</a> ... <a href="http://file6.nc" target="_blank">file6.nc</a> <a href="http://output.nc" target="_blank">output.nc</a>" is worse. Plotting in ncl gives "unknown format<br>Assistance will be appreciated.<br></div>
_______________________________________________<br>
ncl-talk mailing list<br>
<a href="mailto:ncl-talk@ucar.edu" target="_blank">ncl-talk@ucar.edu</a><br>
List instructions, subscriber options, unsubscribe:<br>
<a href="http://mailman.ucar.edu/mailman/listinfo/ncl-talk" rel="noreferrer" target="_blank">http://mailman.ucar.edu/mailman/listinfo/ncl-talk</a></blockquote></div>
_______________________________________________<br>
ncl-talk mailing list<br>
<a href="mailto:ncl-talk@ucar.edu" target="_blank">ncl-talk@ucar.edu</a><br>
List instructions, subscriber options, unsubscribe:<br>
<a href="http://mailman.ucar.edu/mailman/listinfo/ncl-talk" rel="noreferrer" target="_blank">http://mailman.ucar.edu/mailman/listinfo/ncl-talk</a></blockquote></div>
</blockquote></div>