Ryan Quey avatar image
Ryan Quey asked Erick Ramirez edited

Why are clustering keys missing when instantiating Dao from InventoryMapper?

I am using the Cassandra Java Driver, v.4.6.1.

While creating an instance of my DAO instance, I hit the following error:

com.datastax.oss.driver.api.core.servererrors.InvalidQueryException: Some clustering keys are missing: primary_genre, feed_url

Those are indeed my clustering keys, however I'm not sure what they are missing from.

The error appears similar to this one, except the difference is that I didn't realize I was even hitting the db. The stack trace shows that this error comes up on the bottom line, where I'm creating a DAO instance:

InventoryMapper inventoryMapper = InventoryMapper.builder(db.session).build(); 
PodcastDao dao = inventoryMapper.podcastDao("podcast_analysis_tool", "podcasts_by_language");

Here is some of my other code, in case it's helpful

InventoryMapper class :

public interface InventoryMapper {
  PodcastDao podcastDao(@DaoKeyspace String keyspace, @DaoTable String table);

Class definition:

public class Podcast {
   private String language;
   private String primaryGenre;
   private String feedUrl;

CQL to define the table:

CREATE TABLE IF NOT EXISTS podcast_analysis_tool.podcasts_by_language (owner TEXT, name TEXT, image_url30 TEXT,image_url60 TEXT,image_url100 TEXT,image_url600 TEXT, api TEXT, api_id TEXT, api_url TEXT, country TEXT,feed_url TEXT, genres LIST<TEXT>, api_genre_ids LIST<TEXT>,primary_genre TEXT,release_date TIMESTAMP,explicit BOOLEAN,episode_count INT,rss_feed TEXT,found_by_queries List<frozen<Map<Text, Text>>>, description TEXT,summary TEXT,description_subtitle TEXT,webmaster TEXT,owner_email TEXT,author TEXT,language
TEXT,website_url TEXT,updated_at TIMESTAMP, PRIMARY KEY ((language), primary_genre, feed_url)) WITH CLUSTERING ORDER BY(primary_genre ASC, feed_url ASC);
java driver
10 |1000

Up to 8 attachments (including images) can be used with a maximum of 1.0 MiB each and 10.0 MiB total.

1 Answer

Ryan Quey avatar image
Ryan Quey answered

Found out the answer on my own:

In my PodcastDao class, the `@Update` method wasn't sending in all the clustering columns. In other words, even when just instantiating a DAO instance for a class, if the definitions aren't correct it will throw an error as if already performing the CQL query (in this case, throwing as if it had done an UPDATE).

10 |1000

Up to 8 attachments (including images) can be used with a maximum of 1.0 MiB each and 10.0 MiB total.