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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: "Anderson, Alan C. " <address@hidden> >Organization: St. Cloud State >Keywords: 200401082149.i08Ln1p2002455 McIDAS ADDE LDM ldm-mcidas Alan, >using mcidas -config to start resulted in a .mcidasrc file created >in the user's home dir, so problem is solved. I will make >a note in our set up book. Sounds good. >As long as I am writing, I will mention one other item. >This problem occurred on the same machine with the missing >.mcidasrc file, but I had solved it so did not mention it earlier. > >When configuring the user account (other than mcidas), the >LSSERVE.BAT failed to complete. Most of it ran but it >failed and gave an error message that was abandoned because >TE XCDDATA (or something related to it) was not found. That is correct, you have to define XCDDATA before running LSSERVE.BAT. I believe that this is in the online documentation. >I recalled that we had to do a TE XCDDATA "path name >when configuring the mcidas account, where path name is >the directory where the XCD data files are to go (I think). Right. >So I ran the same command for the user, TE XCDDATA "path name >and then LSSERVE.BAT was successful. In this case, I used >/var/data/mcidas as the path name since that is the directory >where all of our data files are kept. Execllent. >Can you comment ? Have I created another gremlin? I do have a comment. I strongly recommend that the non-'mcidas' user not have to setup ADDE dataset definitions since this is something that the McIDAS super user 'mcidas' should do for everyone. The recommendation is that 'mcidas' setup ADDE dataset definitions by defining XCDDATA, and running LSSERVE.BAT. The next step is that the ADDE remote server be installed on the machine that is ingesting data and running decoders. The McIDAS user then defines a set of DATALOCs by configuring ~mcidas/data/LOCDATA.BAT and running 'BATCH LOCDATA.BAT" from the ~mcidas/workdata directory. All other users, by virtue of their defining the Unix environment variable MCTABLE_READ to include the ADDE DATALOC definition file created by 'mcidas', ~mcidas/data/ADDESITE.TXT, will then automatically be pointed at the machines setup by 'mcidas' for ADDE access. Those other users, however, can override those definitions by pointing at different machines for their data. They do this by running DATALOC commands from their McIDAS session. Those commands update a file called MCTABLE.TXT which should get put in their ~/mcidas/data directory. MCTABLE.TXT will get put there if they correctly defined the Unix enviroment variable MCTABLE_WRITE (as per instructions). When their McIDAS session goes to access data from an ADDE server, it first reads MCTABLE.TXT to see where it should go. If it doesn't find a DATALOC for the dataset there, it uses ~mcidas/data/ADDESITE.TXT definitions. If it doesn't find what it needs there, it assumes that the data is LOCAL-DATA. It is the last case where dataset definitions created by DSSERVE commands run from LSSERVE.BAT are needed. >Bye Tom See you. Tom