ws-index.rb

require 'elastic-workplace-search'
require 'json'

Elastic::WorkplaceSearch.access_token = 'my-access-token'
client = Elastic::WorkplaceSearch::Client.new
Elastic::WorkplaceSearch.endpoint = 'https://my-endpoint.ent-search.us-central1.gcp.cloud.es.io/api/ws/v1'

content_source_key = 'my-source-key'

documents = []
id = nil
i = 0

# https://dumps.wikimedia.org/other/cirrussearch/20200518/enwiki-20200518-cirrussearch-content.json.gz
gzfile = open("enwiki-20200518-cirrussearch-content.json.gz")
data = Zlib::GzipReader.new(gzfile)
data.each_line do |line|

  parsed = JSON.parse(line)

  if parsed['index'] && parsed['index']['_type'] == "page" && parsed['index']['_id']
    id = parsed['index']['_id']
    next
  end

  i += 1
  doc = {}

  if parsed['title'] == nil
    puts "Skipping line #{i} with id #{id} since the TITLE is empty."
    next
  end

  doc['id'] = id
  doc['title'] = parsed['title']
  doc['timestamp'] = parsed['timestamp']
  doc['create_timestamp'] = parsed['create_timestamp']
  doc['incoming_links'] = parsed['incoming_links']
  doc['category'] = parsed['category']
  doc['text'] = parsed['text']
  doc['text_bytes'] = parsed['text_bytes']
  doc['content_model'] = parsed['content_model']
  doc['heading'] = parsed['heading']
  doc['opening_text'] = parsed['opening_text']
  doc['popularity_score'] = parsed['popularity_score']
  doc['url'] = "https://en.wikipedia.org/wiki/#{doc['title'].gsub(/ /, '_')}"
  
  documents << doc

  if i % 100 == 0
    begin
      document_receipts = client.index_documents(content_source_key, documents)
      puts "Uploaded #{i} documents"
    rescue Elastic::WorkplaceSearch::ClientException => e
      puts e
    end
    documents = []
  end
end

Increasing the timeout fixed this:

client = Elastic::WorkplaceSearch::Client.new(overall_timeout: 300)