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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: Gilbert Sebenste <address@hidden> >Organization: NIU >Keywords: 199910062147.PAA18740 McIDAS-X 7.60 Gilbert, >OK. I have a little time to write. > >My disk drive is going ballistic, and I see you're doing the install as I >type this. That's great! If there is anything else you need, let me know. >I willdo what I can. I installed McIDAS-X,-XCD this morning as per the online instructions in: http://www.unidata.ucar.edu/packages/mcidas/mcx/mcidas-x.html This included: o installing f2c, f2c.h, and libf2c.a as 'root' in /usr/local/bin, /usr/local/include, and /usr/local/lib, respectively o making McIDAS-X,-XCD: <login as 'mcidas'> cd mcidas7.6/src make all There were a couple of problems in the build: for some reason, 'ar' failed to put a.o and a2sort.o in libmcidas.a. This caused a number of routines link to fail (you can see which ones by reviewing ~mcidas/mcidas7.6/src/makelog). I suspect something fishy with memory caching in RH 6.0. I corrected the problem by: ar r libmcidas.a a.o a2sort.o make all make install.all o updating the set of REDIRECTions for the user 'mcidas'. The adjustment comes from the set of REDIRECTions defined in ~mcidas/data/LOCAL.NAM which is, in turn, a modified copy of ~mcidas/data/EXAMPLE.NAM: cd ~mcidas/data cp EXAMPLE.NAM LOCAL.NAM <edit LOCAL.NAM and set /data/mcidas as the directory for almost all files> cp DATALOC.BAT LOCDATA.BAT <edit LOCDATA.BAT and set weather.admin.niu.edu as the server to contact for ADDE data transfers> cp DSSERVE.BAT LSSERVE.BAT <I made no changes to LSSERVE.BAT since your install seemed pretty normal (i.e. you are keeping 10 of each kind of image received in the Unidata-Wisconsin datastream) cd ~mcidas/workdata redirect.k REST LOCAL.NAM o specifying the XCD data output directory as /data/mcidas as a McIDAS string: cd ~mcidas/workdata te.k XCDDATA \"/data/mcidas o configuring XCD decoder information: <still as mcidas in ~mcidas/workdata> batch.k XCD.BAT batch.k XCDDEC.BAT batch.k LSSERVE.BAT o setup where 'mcidas' should get its data by ADDE: <still as mcidas in ~mcidas/workdata> batch.k LOCDATA.BAT I also installed the McIDAS remote ADDE server as per the instructions in: http://www.unidata.ucar.edu/packages/mcidas/mcx/adde_mcx.html This included as 'root': o creating a user named 'mcadde'. 'mcadde's HOME directory is the same as the HOME directory for 'mcidas'. 'mcadde' does not allow logins by virtue of its default shell being /bin/false <as 'mcidas'> o updating ~mcidas/.mcenv. This was mostly correct. The only change I made was filling out PATH to include directories that 'mcidas' has set. In order to get XCD working, I followed the instructions in: http://www.unidata.ucar.edu/packages/mcidas/mcx/mcidas-xcd.html This page also included the information necessary to setup McIDAS data scouring. In doing the mods for XCD, as the user 'ldm' I: o added 'exec "xcd_run MONITOR" to ~ldm/etc/ldmd.conf o added XCD entries to ~ldm/etc/pqact.conf o turned on McIDAS GRID file decoding since your machine appeared to have plenty of space o setup McIDAS data file scouring by editing ~mcidas/workdata/mcscour.sh and setting PATH and LD_LIBRARY_PATH to match those in 'mcidas' .cshrc I also: o updated the ldm-mcidas decoders distribution to the current release o copied xcd_run from ~mcidas/bin to ~ldm/decoders. I edited xcd_run and set PATH and LD_LIBRARY_PATH to match your system o copied batch.k from ~ldm/decoders/ldm-mcidas-7.6.1/bin to ~ldm/decoders. I edited xcd_run and set PATH and LD_LIBRARY_PATH to match your system o added ~ldm/decoders to 'ldm's PATH in .cshrc o adjusted the lwtoa3 entry in ~ldm/etc/pqact.conf to tidy it up and remove debug logging (the -x flag) o sourced 'ldm's modified .cshrc file and stopped and restarted the LDM: <login as ldm> <edit .cshrc and add /home/ldm/decoders to path> source .cshrc ldmadmin stop <wait for all LDM processes to exit> ldmadmin start o turned on creation of topographic and GOES-East/West composite images: <login as mcidas> cd workdata route.k REL C route.k REL N I tested the ADDE remote server by doing a number of image loads, surface plots, etc. Everything except the 21Z GOES-West images (and products derived from them) seemed to be working, but I was not exhaustive. The 20Z GOES-West images were fine, so I didn't spend a lot of time worrying about the 21Z ones. All-in-all, the install was done as per the Unidata McIDAS-X online instructions. The only thing that differed was the correction of the problem of 'ar' failing to put a.o and a2sort.o into libmcidas.a. >By the way, just as an FYI to you and everyone else at UNIDATA, what I am >also doing is getting the live Master Nexrad Grid (MNG) product from >Kavouras. It's a 1 KM resolution, 48 color mosaic that simply blows WSI's >2 KM resolution product right out of the water in terms of resolution, >quality control, and the 48 color capability (updated every 5 minutes). To >get this to work, it's rather crazy. Mike Dross at UNCC is sending me the >data from North Carolina as he gets it off the satellite; and then I >re-distribute it to U of I who have also signed on. I have become the >relay of Kavouras' MNG. It's $150/month ($50 less than NOWRAD), and well >worth it. If anyone else is interested, or if you want a test feed, you >can contact me. A one year contract is required, but I have had this >over a month now, and I love it. WXP 5.04 handles the radar display well. So, you are indirectly working for Kavorus? This sounds like a pretty "funny" setup for getting the MNG product to sites. What is the arrangement that you have with Duke Power and Kavorus to do this? Tom >From address@hidden Tue Oct 12 16:46:45 1999 Thank you very much! I just got back from a trip to one of our satellite campuses. I'll have time to read all this tomorrow afternoonin detail. Thank you again! Gilbert