This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
>From: Edward Hindman <address@hidden> >Organization: CCNY >Keywords: 200203060049.g260nKK25416 McIDAS DSSEVE Ward, I logged onto cirrus a little while ago and tried to track down why the McIDAS DSSERVE command is not working. I had decided that the best course of action was to rebuild your McIDAS distribution, so I tried doing so. The problem I ran into was that the C compiler is not working in much the same way that DSSERVE is not working. Here is the output from a simple attempt to compile a C program: [mcidas@cirrus src]$ cc -c argctime.c cc: Internal error: Segmentation fault (program cpp0) Please submit a full bug report. See <URL:http://www.gnu.org/software/gcc/bugs.html> for instructions. I am assuming that the C compiler was working on cirrus at one point and that you were able to build McIDAS using it. If, on the other hand, this machine had McIDAS built under a previous version of the operating system and then it was upgraded AND nobody has since tried compiling programs, then I would say that something went wrong in the OS installation. So, I am at an impasse: I can't move forward without having a working compilation environment, and that is broken for a reason I can not discern. I think that the best course of action at this moment is for someone to take a hard look at cirrus to make sure that it has not been hacked into. It may be the case that the operating system needs reloading. If done correctly, this will not destroy the McIDAS installation and things should come back up working correctly. As a final comment, earlier today I was able to run various McIDAS IMG* commands (like IMGLIST) after hand creating the RESOLV.SRV file that is the repository of ADDE dataset definitions. It may be the case that a user logging onto cirrus as wxnut may be able to complete the Lecture 16 exercises since s/he won't have to create MYDATA/IMAGES since it already exists. Question: Is the exercise to be run only on cirrus and only by wxnut? If so, you may be OK until someone can get the time to reload the operating system (or find out if there is a bad system shared library). I am sorry to not have better news... Tom >My student Kamal Thapa and I are conducting a study using McIDAS. We have >been able to download historic images from the SSEC and analyze them >using McIDAS following instructions at > >www.sci.ccny.cuny.edu/~hindman/eas308/LECT16.htm > >In fact, you wrote a portion of those instructions last spring earning >yourself the "E-mail visiting professor" title. > >Just today we found the DSSERVE LIST MYDATA and the DSSERVE ADD >MYDATA/IMAGES AREA 1 9999 "All images in AREA format commands do not >work; we cannot append MYDATA to our McIDAS session to analyze the >downloaded images. > >We are running McIDAS at address@hidden >As soon as possible, could you run McIDAS on Cirrus and >see what you learn and advise us how to correct our instructions in Lect16? > >Thank you! > >Prof. Ward Hindman >CCNY >From address@hidden Wed Mar 6 13:11:36 2002 Dear Tom: Thanks for getting to Cirrus so quickly. To answer your question of earlier today, our McIDAS is only installed on Cirrus and is only run by wxnut. We run McIDAS remotely from other PC's through Exceed. You got us going again. We can display and analyze our MYDATA images. I'll have the OS problem you mentioned investigated after we get our research done this month just in case the investigation "kills" McIDAS. Thank you for your continued critical support. Ward Hindman CCNY