ElasticSearch timeouts after upgrade to 6.5.1

Oh yes, I forgot that Elasticsearch sets that by default. You could try opening the heap dump locally and seeing if it looks similar to this one - lots of instances of OptOutQueryCache taking up all the space.

However, 6.5.3 is now released, so rather than wading into a heap dump I think it'd be better to try and upgrade and see if it is the problem that I guessed it might be that is now resolved.