If you haven't yet heard the news, Oracle Endeca Information Discovery now has a YouTube channel (http://www.youtube.com/user/OracleEID/)!
We have released a new screencast series for "Getting Started with Endeca Information Discovery". This screencast series will help users get up to speed quickly on the basics of Endeca Information Discovery. By the end of this screencast series, you should be ready to build your own application. You can find the full outline here: http://www.oracle.com/technetwork/middleware/endeca/learnmore/index.html
If you haven't yet done it, install v2.3 and the Quick Start application (http://www.oracle.com/technetwork/middleware/endeca/downloads/index.html). Once you have the Quick Start application up and running, you are ready to get started!
We want to hear from you! This forum thread is intended to gather any feedback you want to share about your experiences with the screencast series.
Here are a few troubleshooting hints based on issues I ran into and mistakes I made when working through the examples.
When loading data, if your graph says you processed 80, 654 records, check that you configured the correct number for Number of Skipped Records property in the Data Reader component and for the Skip Source Rows property in the Metadata. One should be configured as “1” and the other should be configured as “0”. If both are set to “1”, an extra record will be skipped.
When creating the Survey Whitelist component, if you get an invalid data error for the .csv file, make sure you have a carriage return at the end of the last record. The record is incomplete without the carriage return. (In other words, your .csv file should end with a blank line.)
If you get the error “output port 0 is undefined” in the log, you have connected an edge to the wrong output port in the source component. The edge should be connected to the top output port. (Note: This error will be in the log BEFORE the stack trace.)
If you cannot map the data in the joiner, add an edge from the joiner to a target. An output join is required to map data.
In "Getting Started Part 3.3 - Create Attribute Groups", if you receive this error:
ERROR [WatchDog] - Node WEB_SERVICE_CLIENT0 finished with status: ERROR caused by: Cannot add nonexistent properties to groups: ProductName, ProductCategoryName, ProductSubcategoryName
ERROR [WatchDog] - Node WEB_SERVICE_CLIENT0 error details:
org.apache.axis2.AxisFault: Cannot add nonexistent properties to groups: ProductName, ProductCategoryName, ProductSubcategoryName
Then it means you did not properly update the attributes in the Product Join and Category Join during Part 2.X - Exercises for the User. The original screencast for 2.X inadvertently left out that step. But, it has now been corrected. So, if you run into this error, re-watch GS2.X here: http://www.youtube.com/watch?v=ntO9M7ozLZ0&list=PLF23635ACA47F1E6D&index=8&feature=plpp_video
ERROR [SocketReader] - Socket timed out while connected to the dgraph: java.net.SocketTimeoutException: Read timed out
ERROR [SocketReader] - Received abort message from server: Socket timed out while connected to the dgraph: java.net.SocketTimeoutException: Read timed out
ERROR [Consumer-0] - Error while trying to cancel Bulk Ingest: Read timed out
java.net.SocketTimeoutException: Read timed out
at java.net.SocketInputStream.socketRead0(Native Method)
ERROR [WatchDog] - Graph execution finished with error
ERROR [WatchDog] - Node ENDECA_BULK_ADD_OR_REPLACE_RECORDS1 finished with status: ERROR
ERROR [WatchDog] - Node ENDECA_BULK_ADD_OR_REPLACE_RECORDS1 error details:
INFO [WatchDog] - [Clover] Post-execute phase finalization: 0
INFO [WatchDog] - [Clover] phase: 0 post-execute finalization successfully.
INFO [WatchDog] - Execution of phase  finished with error - elapsed time(sec): 191
ERROR [WatchDog] - !!! Phase finished with error - stopping graph run !!!