0

There are a lot of questions about this error, but none for this condition.

I am running Elasticsearch 5.4.1 with a java client(1.8) which uses the API to make Elasticsearch calls. It is on a Mac. I have a large number of documents which I have to search on / insert/ and merge on. (~ 8000 documents)

I get this exception

None of the configured nodes are available: [{#transport#-1}{gNzkHZmURzaabE336I-T4w}{localhost}{127.0.0.1:9300}]

The thing is, it works with a smaller number of entries (e.g. 5000 documents). So the connection seems to be going down for some reason? Should I allocate more memory/ use more nodes to it? Is there some weird garbage collection going on?

According to the activity monitor for the Mac, memory pressure is fine.

user2689782
  • 747
  • 14
  • 31
  • Are you doing [bulk](https://www.elastic.co/guide/en/elasticsearch/reference/current/docs-bulk.html) queries ? Maybe you can try to check if increasing the `queue_size` settings for write operations solves the problem. – Pierre-Nicolas Mougel Jun 03 '19 at 09:04
  • I am doing some bulk queries...ok let me take a look. Thanks! – user2689782 Jun 03 '19 at 14:44

0 Answers0