NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
hmmm, thats funny, im not sure why we dont get a stacktrace from this call in DataRootHandler: logCatalogInit.error("readCatalog(): Exception on catalog=" + catalogFullPath + " " + t.getMessage() + "\n log=" + msg, t); heres the entry in log4j.xml: <!-- catalogInit.log is recreated each time Tomcat is started --> <appender name="catalogInit" class="org.apache.log4j.FileAppender"> <param name="File" value="${tds.log.dir}/catalogInit.log"/> <param name="Append" value="false"/> <layout class="org.apache.log4j.PatternLayout"> <param name="ConversionPattern" value="%m%n"/> </layout> </appender> i will see if i can reproduce this. Jon Blower wrote:
Hi John, Where can I get the full stack trace? catalogInit.log only has the summary information. Cheers, Jon -----Original Message-----From: John Caron [mailto:caron@xxxxxxxxxxxxxxxx] Sent: 10 July 2010 16:34To: Jon Blower Cc: thredds@xxxxxxxxxxxxxxxx Subject: java.lang.NullPointerException: Name is null readCatalog(): Exception on catalog=/usr/local/tomcat/content/thredds/idd/atm_mod.xml Name is null log=null catalog java.lang.NullPointerException: Name is null I havent been able to figure out where this is coming from. If anyone sees it again, send me the full stack trace. thanks. Jon Blower wrote:Hi John, Would you be happy to send me (or the list) the relevant fragment of your TDS catalog that works in 4.2? I've seen exactly the same error trying to override the units for a particular variable in an FMRC aggregation and I'd be very interested to understand your workaround. Cheers, Jon -----Original Message----- Date: Fri, 18 Jun 2010 09:21:17 -1000 From: "John Maurer, IV" <jmaurer@xxxxxxxxxx> To: thredds@xxxxxxxxxxxxxxxx Subject: Re: [thredds] TDS 4.2 FMRC catalogInit error: Name is null Message-ID: <fb7ec9b949f72.4c1b3a8d@xxxxxxxxxx> Content-Type: text/plain; charset="us-ascii" Hi All, Here is a brief follow-up on some new progress I've made on this problem: Because of the error from TDS 4.2, I have returned to TDS 4.1 for now but at least found a workaround for the _FillValue mismatch. Because FMRC is returning NaN for _FillValue, I have set this to NaN with NcML in my TDS config file for these datasets. As I stated below, this was not working for me at first because I was making the modificationwithinthe aggregation tag. Placing this modification prior to theaggregationtag instead now gives me the expected results: the metadata on the TDS OPeNDAP page and the OPeNDAP output now both have NaN for _FillValuesothere is no longer a mismatch causing errors for me in visualization applications such as LAS. Cheers, John Maurer Hawaii Ocean Observing System (HiOOS) ----- Original Message ----- From: "John Maurer, IV" <jmaurer@xxxxxxxxxx> Date: Thursday, June 17, 2010 4:10 pm Subject: TDS 4.2 FMRC catalogInit error: Name is null To: thredds@xxxxxxxxxxxxxxxxHi All, I'm working with the latest development release of TDS (4.2) in orderto take advantage of the FMRC refactor. However, I get an error foreachof my FMRC datasets in my TDS catalogInit.log that says something like the following:readCatalog(): Exception oncatalog=/usr/local/tomcat/content/thredds/idd/atm_mod.xml Name is nulllog=null catalog java.lang.NullPointerException: Name is null I do not see where "Name" is supposed to be used in thedocumentation,and I already have lower-case "name" settings in place. Any ideaswhat'sgoing on here? I realize this is not a stable release yet, so my apologies if I'm jumping the gun. Here is the URL to our THREDDS catalog:http://oos.soest.hawaii.edu/thredds/catalog.html Go to "Ocean Model Output" or "Atmosphere Model Output" and you willget an HTTP Status 404 - Not Found that results from this problem.A little background on why I care: Because of the way FMRC used to work, I am getting NaN for _FillValuein my Best Time Series even though the source NetCDF dataset fileshave_FillValue set to 1.0E37 in their metadata. This discrepency iscausingproblems when displaying my data in LAS 7.1.1: the metadata say one thing, but TDS delivers another. I have tried using NcML to get thetwoin sync but without any success. From my understanding in talking with others, this is a known bug in TDS that will be fixed in 4.2. I have converted all my FMRC catalogs from the old datasetScan syntax to the new featureCollection syntax as described here:http://www.unidata.ucar.edu/projects/THREDDS/tech/tds4.2/reference/collections/FeatureCollections.html. I have attached one example.Thanks for your time! Cheers, John MaurerHawaii Ocean Observing System (HiOOS)-------------- next part -------------- An HTML attachment was scrubbed... URL:<http://mailman.unidata.ucar.edu/mailing_lists/archives/thredds/attachmehttp://www.unidata.ucar.edu/mailing_lists/nts/20100618/0ab7599a/attachment.html> End of thredds Digest, Vol 17, Issue 17 *************************************** _______________________________________________ thredds mailing list thredds@xxxxxxxxxxxxxxxx For list information or to unsubscribe, visit:
thredds
archives: