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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: Unidata McIDAS <address@hidden> >Organization: Cornell >Keywords: 200102071732.f17HW4L24435 McIDAS-X batch Jdm, >The default installation for the McIDAS7.7 package >appears to have a corrupt file in it. > >The batch.k file is usually a script, and >the one in the distribution is some binary format. The version of batch.k that is a script is the one in the ldm-mcidas package. It is used in LDM initiated ROUTE PostProcessing as a substitute for the "real" batch.k, the one that gets built in the McIDAS distribution. The reason for having the script at all is so that sites do not have to configure their 'ldm' accounts with any McIDAS stuff. The batch.k script gets put into a directory in the 'ldm's PATH, and then it sets some McIDAS environment variables and eventually run the "real" batch.k, the one in ~mcidas/bin. >This is found in two released versions I >downloaded on Jan 19 and Feb 7 (2001). > >This appears to be causing problems >with composite imagery and Topo composites. I would suspect that the script that should be used for the PostProcessing was not copied into a directory in the ~ldm hierarchy. I recommend that the script gets copied into a directory like ~ldm/decoders and then get edited in place (edited to set the McIDAS environment variables and PATH which has as its first directory the ~mcidas/bin directory where the real batch.k is located). > Is there a corrected version somewhere? There is no problem with the distriubtion.l >I attempted to substitute a 7.6 version batch.k >but I think some of the workarounds I was trying >"hosed" the installation. Soo, I'm reinstalling >from scratch. You didn't have to reinstall. What you need to do is: o get the batch.k script from the ldm-mcidas distribution (in the bin directory in binary distributions; in the src/decode directory in source distributions) o copy that script to a directory like ~ldm/decoders (or ~ldm/util) that is in the PATH for your 'ldm' user o edit that script and set the McIDAS environment variables to match your McIDAS-X installation >Please mail responses to: > address@hidden Will do. >I'm currently in a lab and do not often access mail here. OK. Are you running all of this on a Linux machine? I ask because I am troubleshooting a problem that has emerged while running soundings/hodographs on any machine but accessing data hosted on a remote ADDE server running on Linux. The problem consumed my day yesterday, and I am no closer to a solution than when I started. I am attempting to ascertain whether the problem I am seeing is a 7.0 Linux one; a Linux one in general; or an error in McIDAS that only comes to light on Linux. Since you are known to be well versed in Linux, I seek your advice! >James D. Marco >System Admin, Programmer/Analyst >Cornell University Tom Yoksas >From address@hidden Wed Feb 7 11:17:47 2001 >Subject: Re: 20010207: batch.k problem in McIDAS-X 7.7x Tom, OK. I missed the batch.k difference in the last go-round. I am installing a new McIDAS server for student stuff. The old one is faltering under heavy loads running the Map Room and MAD Lab (Meteorological and Atmospheric Data Lab). And I need to upgrade to RH 7.x this summer, soo, the two will bootstrap each other. Thanks for the fix with the batch.k. Typically, lack of understanding on my part. I just hosed the new install, too. Well, the a new build will take care of it. Soundings/hodographs: Yup. All Linux, a curse and a blessing. I can check the old 7.6 installation from last summer. Actually, you can access the mcidas stuff from there, I think. I think the wrappers are open to your machines. Send along your IP's and I will insure it is. Standard Unidata password for user mcidas...feel free. I will get back to you after some checks in the lab. Thanks! jdm >From address@hidden Wed Feb 7 12:13:04 2001 Tom, OK. There seems to be a problem on the old version of Mcidas 7.6 on RH 6.2, also. I get no data displayed...meteograms are OK. Wrappers are open to you: 128.117.140.xxx (ucar) 128.117.68.xxx (ucar) use the mcidas account at the XXXXXX Feel free to test it here. I am not sure if the configs are set up. I never use this, obviously. I will need to check this on the new stuff, anyway, soo, I'll make it a point to look at that stuff. Thanks again! jdm