I have three environments which are set up to use Solr: local, staging and production. The local configuration differs slightly to the staging/production however the other two are identical.

I recently did a schema update across all instances which required me to re-index the data in order to use boost functions correctly. Everything was working as expected until for some reason the production server unexpectedly started throwing out 503 errors and timeouts when using any boost function in the query string (e.g. bf=log(my_field)). The field I'm attempting to apply the function to is defined as an integer and is is non-multivalued.

We had a similar problem on the staging server, however this was resolved by re-indexing the data, which is what has been taking place on production. The strangest thing is that these functions were working with the newly indexed data, but have suddenly stopped.

Any ideas on why this is happening on one server, but not the others with the same dataset?

  • Okay, so it was the cache which was giving the timeout, the server itself is giving a java error: java.lang.StringIndexOutOfBoundsException: String index out of range: 0. Any ideas as to why this could suddenly start happening when using a boost? – RoryGilchrist Mar 13 '15 at 9:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.