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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: "Fingerhut, William A" <address@hidden> >Organization: Lyndon State >Keywords: 200207191535.g6JFZa920941 McIDAS-X XCD REDIRECT Bill, >Are your feeling any better today? Hope you are. Yes, I am well on the road to recovery. Good thing 'cause I am out of time before my workshop! >I have been pouring over my xcd problem and going no where. >I can't find anything wrong, but we know that something must >be wrong. > >The machine is ingesting DDS, and the data monitors seem to >be running. A ps -ef | grep ldm show the DM's. This only means that startxcd.k is running and that the copy of DECINFO.DAT in ~mcidas/workdata is setup to tell startxcd.k to run the data monitors that you watn. >I just ran: wxtlist.k APRO=RR5 AST=BTV >and found this text bulletin on my machine. > >There are *.RAP and *.IDX files in the xcd data directory, >but no *.IDT or *.RAT files. This is bad. It indicates a configuration problem. >I looked at ~mcidas/workdata/XCDINST.LOG >and, since BILDTEXT is supposed to create *.RAT (?), noticed >BILDTEXT INIT SAOMETAR.RAP SAOMETAR.RAT 6500 4 C4 36 1 80 SAOMETAR X >32 >BILDTEXT: No datatypes defined for decoder SAOMETAR >BILDTEXT: M0buildidt: GetDecoderTypes return -2 >Input configuration now being built > >pointer file name=SAOMETAR.RAP text file name SAOMETAR.RAT >number of stations allowed in the file 6500 >number of stations currently in the file -5 >number of reports per observation period 4 >number of words needed to define a station 1 >number of observation blocks per station 36 >number of hours between observation block 1 >number of bytes per observation line 80 >number of words in station information blk 1448 >number of words in observation blk 40 >maximum size of TXT file 33554432 >starting lw address of first word of id 1024 >starting lw address of second word of id 1024 >starting lw address of station blocks 8192 >master station id table to use STNDB.CORE >decoder used to define id table SAOMETAR >padding SAOMETAR.RAP to word 9420192 with hex80s >BILDTEXT: done > >Is the "no datatypes found" a clue? Yes, this is definitely a clue. Exactly what it is telling us I am not positive of, however. I would guess that the IDGROUP invocation run from XCD.BAT did not run successfully. IDGROUP creates GROUPS.DAT and COUNTRY.DAT, both in the ~mcidas/workdata directory. The comment in the online help for IDGROUP: IDGROUP is a utility for creating and editing groups of stations used with rapid text accessing applications. indicates that this is probably at the root of the problem somehow. All I can recommend is that you start the XCD configuration steps from the beginning: o stop the LDM o delete GROUPS.DAT and COUNTRY.DAT o make sure your REDIRECTions are correct o define the McIDAS string XCDDATA o delete the *.RAP files in the output directory o make sure that SCHEMA can be found (use DMAP SCHEMA) o run: BATCH XCD.BAT BATCH XCDDEC.BAT o restart the LDM I am willing to jump onto your system and take a quick look if you like. If you would like me to do this, I will need logins as 'ldm' and 'mcidas'. >I have no idea what to try now. Should I reinstall xcd? I don't think a reinstallation is needed. A reconfiguration seems to be in order, however. Tom