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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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.
Hi Martha, re: > My apologies for being late getting back to you...yesterday was very > hectic, and today has its problems, but I was able to determine that No worries. Things have been busy on this end also :-) re: > (1) TRACE=1 qualifier with IMGDISP from non-AIX MCIDAS session does not > generate a file on the AIX box with the agetserv version 2007c This is very odd. Please verify that the read/write permissions on the ~mcidas/mcidas and ~mcidas/mcidas/data directories are set for owner and group read and write. By the way, I am assuming that the users 'mcidas' and 'mcadde' are in the same group in your setup. Also, you can run the same IMGDISP...TRACE=1 test as 'mcidas'. In this case, the file 'trce' should be created in the ~mcidas/workdata directory as long as the dataset is being accessed as LOCAL-DATA. If the access is through the remote ADDE interface, 'trce' should be written to ~mcidas/mcidas/data. > (2) Replacing the version 2007c agetserv with version 2006 (version > 2006a to be exact) Causes everything to start working again. Ah Ha! So, the agetserv executable built in your v2007c installation is bad. Since there was no change to the source code for agetserv (agetserv.fp) between v2007 and v2007, and since my rebuild and tests yesterday on our AIX 5.1 system do not show any problem with agetserv, I am led to believe that something happened to the agetserv executable during the build. > So the problem appears to be with our AIX 5.3 build of agetserv 2007c. > I would guesstimate that it doesn't work more frequently than it does > work, BTW, say, 10% yes and 90 % no. I would assume that if there is a problem, it should either work or not work. > I will leave the older version of agetserv in place, and will probably > do the upgrade you suggest next week - I am off tomorrow and don't like > to leave a non-working MCIDAS distribution in place. Good plan. > If you would like me to try any other tests in the meantime, please let > me know. Here is what I would do now that we know that the agetserv executable is not working correctly: <as 'mcidas'> cd ~mcidas/bin mv agetserv agetserv.v2006 <- save a copy cd ~mcidas/mcidas2007/src rm agetserv agetserv.o arealine.o make libmcidas.a agetserv ln agetserv ~/bin -- rerun the IMGDISP tests that failed before If the newly built version of agetserv does not work, you could rebuild the distribution from scratch. NB: this can be done without losing the working installation you now have: <still as 'mcidas'> cd ~mcidas/mcidas2007/src make clobber <- deletes everything from the ~mcidas/mcidas07/src directory built during the make process; this does not do anything to the executables in ~mcidas/bin!! make all <- or 'make mcx' if you are not building XCD on this box rm ~mcidas/bin/agetserv ln agetserv ~/bin <- assuming that you have saved the v2006 copy of agetserv -- rerun the tests I am betting that something strange happened during your original v2007c build that caused a non-viable copy of agetserv to be created. A rebuild might well result in a working version. > Thanks. No worries. Cheers, Tom **************************************************************************** Unidata User Support UCAR Unidata Program (303) 497-8642 P.O. Box 3000 address@hidden Boulder, CO 80307 ---------------------------------------------------------------------------- Unidata HomePage http://www.unidata.ucar.edu **************************************************************************** Ticket Details =================== Ticket ID: FTB-213912 Department: Support McIDAS Priority: Normal Status: Closed