Hey Russell, thanks a million for that hint. I tried registering all those jars as well but it didn't work. What, from my point of view, is the most annoying thing here is the fact that the data storage doesn't complain about the missing jars but just works. So the following parts of the script fail with misleading messages. I think some warning could be handy here.
Thanks for your other hint as well I think we will test pig 0.10 and switch over for the productive cluster after its bundled with a cloudera stable. We will see. Best Markus Am Freitag, den 15.06.2012, 12:11 -0700 schrieb Russell Jurney: > Oh, you maybe also need to load other jars? I load avro this way. > > REGISTER /me/pig/build/ivy/lib/Pig/avro-1.5.3.jar > > REGISTER /me/pig/build/ivy/lib/Pig/json-simple-1.1.jar > > REGISTER /me/pig/contrib/piggybank/java/piggybank.jar > > > Russell Jurney http://datasyndrome.com > > On Jun 15, 2012, at 1:50 AM, Markus Resch <markus.re...@adtech.de> wrote: > > Hey all, > > we're currently testing to switch over from CDH3 to CDH4. > When I try to read my Avro input data I get en Schema unknown Error: > > bash-3.2$ pig > 12/06/15 08:48:08 WARN pig.Main: Cannot write to log > file: /usr/lib/pig/pig_1339750088923.log > 2012-06-15 08:48:09,415 [main] INFO > org.apache.pig.backend.hadoop.executionengine.HExecutionEngine - > Connecting to hadoop file system at: file:/// > 2012-06-15 08:48:09,421 [main] WARN > org.apache.hadoop.conf.Configuration - fs.default.name is deprecated. > Instead, use fs.defaultFS > 2012-06-15 08:48:09,747 [main] WARN > org.apache.hadoop.conf.Configuration - fs.default.name is deprecated. > Instead, use fs.defaultFS > grunt> REGISTER /usr/lib/pig/contrib/piggybank/java/piggybank.jar; > grunt> dataImport = LOAD '/our/path/to/our/data/data.avro' USING > org.apache.pig.piggybank.storage.avro.AvroStorage (); > grunt> describe > dataImport; > > Schema for dataImport unknown. > > Is this somehow a known issue or do I make something wrong? > > Thanks > Markus
