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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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 Hsie, We have some bad news for you that we noticed when trying to switch from LDM-6.11.7 to a newly built 6.12.14: Evidently, all of the decoding actions in your GEMPAK pattern-action file, ~ldm/etc/pqact.conf_gempak, and some of the decoding actions in your McIDAS pattern action files, ~ldm/etc/pqact.conf_mcidasA and ~ldm/etc/pqact.conf_mcidasB, have not been working probably since the upgrade of your LDM from v6.10.1 to 6.11.x. As far as I can tell, this switch was likely to have been made back on April 5, 2013. We determined this by looking more carefully at the error messages in the LDM log file(s), ~ldm/logs/ldmd.log*. A characteristic log message that indicates that a GEMPAK decoder is not being run looks like: Jun 10 19:20:44 rainbow pqact[12623] ERROR: [filel.c:305] Deleting failed PIPE entry: pid=12670, cmd="decoders/bin/dcuair -b 24 -m 16 -d logs/dcuair.log -e GEMTBL=/home/gempak/NAWIPS/gempak/tables -s snstns.tbl data/gempak/upperair/YYYYMMDD_upa.gem" Jun 10 19:20:44 rainbow pqact[12623] ERROR: pipe_prodput: trying again: 111 20160610192044.984 IDS|DDPLUS 184738101 UFUS45 KPSR 101920 /pABV1Y9 Jun 10 19:20:44 rainbow pqact[12672] ERROR: No such file or directory Jun 10 19:20:44 rainbow pqact[12672] ERROR: [filel.c:1423] Couldn't execute decoder "decoders/bin/dcuair" Jun 10 19:20:44 rainbow pqact[12672] NOTE: Exiting Note that this log message was generated by LDM-6.11.7. What is the problem? The problem is that the majority of the actions in the GEMPAK pattern-action file, and lots of the actions in the McIDAS pattern-action files implicitly assume that the current working directory when the action is run is the HOME directory of the LDM. Starting with LDM v6.11.x, the current working directory for 'pqact' (the LDM utility that executes the pattern-action file actions) is defined in the LDM registry, ~ldm/etc/registry.xml, by the tag <datadir-path></datadir-path>. The current setting for this tag is /home/ldm/data. Let's examine one (the first one, actually) of the many actions in the GEMPAK pattern-action file, ~ldm/etc/pqact.conf_gempak: # NAM #211 80km grid CONUS HDS (/mNAM|/mNMM).*#211 PIPE decoders/bin/dcgrib2 -d logs/dcgrib2_NAM211.log -e GEMTBL=/home/gempak/NAWIPS/gempak/tables The reference to the GEMPAK decoder to be executed is a relative path: decoders/bin/dcgrib2 If the current working directory for 'pqact' was /home/ldm, then this relative path reference would be OK. However, since the current working directory for 'pqact' is /home/ldm/data, then the decoder will not be found. The exact same comment can be made for the log file to be written: -d logs/dcgrib2_NAM211.log This says to write the log messages for this action in /home/ldm/data/logs/dcgrib2_NAM211.log. The problem is that there is no /home/ldm/data/logs directory. What can be done? We could change the <datadir-path></datadir-path> tag(s) in the LDM registry from /home/ldm/data to /home/ldm and then restart the LDM, and then the GEMPAK and McIDAS decoded data files will once again be created and written. The potentially huge downside to this change would be that the processing load on rainbow would undoubtedly get much, much higher, and this might well negatively impact your ability to process (FILE) the CONDUIT data that you have been looking at recently. Aside: The other thing we noticed is that the OS on rainbow is now very, very old: RedHat Enterprise 5.11. How upgrading to a current release of RHEL would affect rainbow's operation we can not say. What to do? So, the question to you is if you want us to modify the LDM setup on rainbow to alter the configuration so that GEMPAK and McIDAS decoding would once again work ** realizing that the increased CPU use might actually negatively affect your ability to FILE the CONDUIT data **? As I write this, rainbow is running LDM-6.11.7... 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: GLG-935498 Department: Support IDD Priority: Normal Status: Closed =================== NOTE: All email exchanges with Unidata User Support are recorded in the Unidata inquiry tracking system and then made publicly available through the web. If you do not want to have your interactions made available in this way, you must let us know in each email you send to us.