Troubleshooting

Check log files for errors

1. ${tomcat.home}/logs/catalina.out: This is where Tomcat prints messages about Tomcat startup and uncaught exceptions from any webapp. Monitor this occasionally. Delete it when you install a new version of thredds.war.
2. ${tomcat.home}/logs/localhost.{date}.log: This is where log messages for Tomcat are sent.
3. ${tomcat.home}/content/thredds/logs/serverStartup.log: As TDS starts up, it logs messages here. Usually only interesting if TDS is failing to start.
4. ${tomcat.home}/content/thredds/logs/catalogInit.log: Errors and warnings in your catalog configuration files are shown here. Look at this closely.
5. ${tomcat.home}/logs/access.{date}.log: This is where Tomcat access logs are put, if you have enabled them. You need to manage these by removing them, say, once a month.
6. ${tomcat.home}/content/thredds/logs/threddsServlet.log: Each request that the TDS responds to gets logged here. Errors and warnings that are associated with a request are logged here. You need to manage these by removing them, say, once a month.
7. ${tomcat.home}/content/thredds/logs/cache.log: Cache scour messages are kept here. Only interesting if you are having problems with cache directories getting too large.
8. ${tomcat.home}/content/thredds/logs/featureCollectionScan.log: Only used if you use feature collections.
9. ${tomcat.home}/content/thredds/logs/models.log: Only used if you serve GRIB files.

Common errors in the TDS Configuration Catalogs:

Duplicate path(s)

    <datasetRoot path="testdup" location="C:/data/" />
    <datasetScan name="duplicate path" path="testdup" location="/home/workshop/data/" serviceName="dodsServer" />

  Error: already have dataRoot =<testdup>  mapped to directory= <C:/data/> wanted to map to=</home/workshop/data/> in catalog

Duplicate id(s)

    <dataset name="Test Single Dataset 2" ID="testDataset2" serviceName="odap" urlPath="test/testData2.grib2" dataType="Grid" />
    <dataset name="Test Single Dataset 3" ID="testDataset2" serviceName="odap" urlPath="test/testData3.grib3" dataType="Grid" />

  WARNING: Duplicate id on  'THREDDS Catalog Name/Test Single Dataset 3' id= 'testDataset2' 

Data directory doesnt exist

If you refer to a non-existant directory:

    <datasetRoot path="sage" location="C:/data/notexist/" />
    <datasetScan path="sage" location="C:/data/notexist/" />

  Data Root =sage directory= <C:/data/notexist /> does not exist
or:
  Invalid InvDatasetScan <path=testAll; scanLocation=bad/content/testdata>: CrawlableDataset for scanLocation does not exist.
    ... Dropping this datasetScan [testAll].

DatasetScan has missing or invalid service

If you refer to a non-existant service, or omit a service:

  <datasetScan name="bad service name" path="segundo" location="C:/data/" serviceName="badd" />
  <datasetScan name="no service name" path="tertiary" location="C:/data/" />

you will get an error message in ${tomcat.home}/content/thredds/logs/serverStartup.log:

  **Error: DatasetScan (Top Dataset/bad service name ): must have a default service
  **Error: DatasetScan (Top Dataset/no service name): must have a default service

Data not compatible with service

If you use a service on a file which that service cant deal with:

  <dataset name="Image of my Data" urlPath="images/labyrinth.jpg" serviceName="dodsServer" dataType="Grid" />

• You wont get an error in serverStartup.log
• But when you try to access it:
  • Browser: DODServlet ERROR: Cant read C:/data/images/labyrinth.jpg: not a valid NetCDF file.
  • nj22/IDV:: Lexical error at line 1, column 1. Encountered: "" (0), after : ""

Data file does not exist

If you refer to a non-existant file in a dataset:

  <dataset name="My Data" ID="Y" urlPath="images/labyrinth.nc" serviceName="dodsServer" dataType="Grid" />

• You wont get an error in serverStartup.log
• If you try to access it (OPeNDAP)
  • Browser: message = "Cant find images/labyrinth.nc"
  • nj22/IDV:: "Cant find images/labyrinth.nc"
• If you try to access it (HTTP server)
  • Browser: Error 404 The requested resource () is not available.
  • nj22/IDV: Error 404 Not Found

DatasetScan points to an empty directory, or filters out all files in the directory.

Aggregation scan points to an empty directory

No warning - but when you try to access the dataset, client gets:

Error {      code = 500;      
  message = "There are no datasets in the aggregation DatasetCollectionManager{ 
  collectionName='/data/goes/**/.gini' recheck=15.0 min    dir=/data/goes/ filter=WildcardMatchOnPath{wildcard=*.gini$ regexp=.*\.gini$}";  
};

and the threddsServlet.log has:

SEVERE: path= /ncmlTest.html 
   java.lang.IllegalStateException: There are no datasets in the aggregation DatasetCollectionManager{   
   collectionName='/data/goes/**/.gini' recheck=15.0 min dir=/data/goes/ filter=WildcardMatchOnPath{wildcard=*.gini$   regexp=.*\.gini$} 

This document is maintained by John Caron and was last updated July 2011