Unable to query due to memory issues.

e.g. https://api.eventdata.crossref.org/v1/events?rows=1 gives:

{
"status: "failed",
"message": "An internal error occurred. If you see this message repeatedly, please contact us."
}

Internally, Elasticsearch is producing:

java.util.concurrent.ExecutionException: CircuitBreakingException[[fielddata] Data too large, data for [_id] would be [6621867416/6.1gb], which is larger than the limit of [6400612761/5.9gb]]"}

This is probably due to sorting on the _id field. There are now ~ 100 million Events. To pick a number out of the air, if each Event in the index took 64 bytes, that would be 6.4 GB, which seems reasonable.

Current config with Query API is version 0.2.14 uses default 8 shards per index. Increasing this and re-indexing may help.

Reading:

• https://www.elastic.co/guide/en/elasticsearch/guide/current/_limiting_memory_usage.html#circuit-breaker
• http://evertrue.github.io/blog/2014/11/16/3-performance-tuning-tips-for-elasticsearch/