NOTE: The decoders
mailing list is no longer active. The list archives are made available for historical reasons.
============================================================================== Robb Kambic Unidata Program Center Software Engineer III Univ. Corp for Atmospheric Research rkambic@xxxxxxxxxxxxxxxx WWW: http://www.unidata.ucar.edu/ ============================================================================== ---------- Forwarded message ---------- Date: Tue, 17 Sep 2002 12:08:57 -0400 From: "[ISO-8859-1] Christian Pagé" <page.christian@xxxxxxx> To: Greg Thompson <gthompsn@xxxxxxxxxxxx> Subject: Re: 20020822: ldm-5.2 pqsurf problems on IRIX Hi Greg, Yes I would be very interested in the code.We do have MySQL running on our LDM/Apache machine, so I could use right away the code. I am not that familiar with MySQL, but how do you setup the location of the database (since it can be quite huge). Is there a way to build an archive from raw wmo metar data using this code too?
Thanks a lot, Christian Pagé On Tuesday, September 17, 2002, at 11:15 AM, Greg Thompson wrote:
Christian, there were long-standing problems with pqsurf like stability and ability to ingest all METARs (our raw files always had ALL METARs while our gdbm files always missed some METARs like Canadian SPECIs). We had horrible stability issues when we changed Linux systems from Debian to Red Hat (pqsurf used to stay alive for weeks at a time then die unexpectedly under Debian but then it began dying every day or two under Red Hat). Finally, I couldn't take it any longer and re-wrote the "metar2nc" Unidata decoder to ingest/decode METARs and populate a MySQL database. I spent nearly 150 hours re-writing the decoder to handle ALL METARs not just the U.S. ones. This DB is far, far better than either raw text files or NetCDF files. You can see the database in action on two websites I maintain: http://www.rap.ucar.edu/weather/surface/ http://adds.aviationweather.gov/projects/adds/metars/ If you're interested in this technique, I can share the code with you. Though it is currently MySQL specific, it would not take much effort to adapt it to either Oracle, mSQL, PostgresSQL, or other db systems. -- +--------------------------------------------------------------+ | Greg Thompson http://www.rap.ucar.edu/~gthompsn/ | | Research Applications Program | | (303) 497-2805 National Center for Atmospheric Research | | (fax) -8401 P.O. Box 3000 Boulder, CO 80307-3000 | +--------------------------------------------------------------+On Thu, 22 Aug 2002, [ISO-8859-1] Christian Pag? wrote:Hi, Your comment made me find the problem, it was the gdbm library : it was not the correct version. I am now using the correct version which is 1.7.3 and everything works fine again! Maybe the gdbm should be bundled with ldm?Christian,I'm glad you found the problem. Actually we are trying to depreciate thepqsurf code because it breaks when every there's a major change in thebulletins. I believe the folks at RAP are working on a solution. I'll cc Celia at RAP, maybe she can point us to a contact person working on thesolution. Thanks, RObb...Thanks Robb! :-) Christian Pag? UQAM On Jeudi, ao?t 22, 2002, at 04:15 , Robb Kambic wrote:On Thu, 22 Aug 2002, Unidata Support wrote:------- Forwarded MessageTo: support@xxxxxxxxxxxxxxxx From: =?ISO-8859-1?Q?Christian_Pag=E9?= <page.christian@xxxxxxx> Subject: ldm-5.2 problems Organization: Universite du Quebec a Montreal Keywords: 200208221254.g7MCs8K03800 LDM 5.2 pqsurfHi, I still have the problem of running ldm-5.2 (see previous emails on this problem). My problem is that pqsurf exits (and ldm-5.2 is brought down) when I have a DBFILE in my pqsurf.conf. Is this a bug in ldm-5.2? It was running fine with ldm-5.1.4.Christian,I a little confused here, does ldm-5.1.4 run fine with the DBFILE inpqsurf.conf? Maybe you need to rebuild GDBM ? Also if the program dump core, could you send the dbx where again?I check the release notes for 5.2, there wasn't any mods to the pqactcode. Robb...Thanks again, Christian Pag? page@xxxxxxxxxxx UQAM http://meteocentre.com/ http://meteoalerte.com/ Agent de recherche et de planif. +1 514 987 3000 ext. 2376 D?partement des Sci. de la Terre et de l'Atmosph?re UQAM ------- End of Forwarded Message==================================================================== === ===== Robb Kambic Unidata Program Center Software Engineer III Univ. Corp for Atmospheric Research rkambic@xxxxxxxxxxxxxxxx WWW: http://www.unidata.ucar.edu/ ==================================================================== === =====Christian Pag? page@xxxxxxxxxxx http://meteocentre.com/ http://meteoalerte.com/ Agent de recherche et de planification +1 514 987 3000 ext. 2376 D?partement des Sciences de la Terre et de l'Atmosph?re UQAM
Christian Pagé page@xxxxxxxxxxx http://meteocentre.com/ http://meteoalerte.com/ Etudiant au Doctorat en Sciences de l'environnement UQAM +1 514 987 3000 ext. 2376
decoders
archives: