solr-conf.yaml

# configuration for SOLR resources
  
config:
  solr.indexer.url: "http://solr:8983/solr/docs"

  # Solr spout and persistence bolt
  solr.status.url: "http://solr:8983/solr/status"
  solr.status.bucket.field: host
  solr.status.bucket.maxsize: 5
  solr.status.metadata.prefix: metadata
  solr.status.max.results: 10

  # time in secs for which the URLs will be considered for fetching after a ack of fail
  spout.ttl.purgatory: 30
  
  # Min time (in msecs) to allow between 2 successive queries to SOLR
  spout.min.delay.queries: 2000
  
  # Delay since previous query date (in secs) after which the nextFetchDate value will be reset to the current time
  # Setting this to -1 or a large value means that the ES will cache the results but also that less and less results
  # might be returned.
  spout.reset.fetchdate.after: 120
  
  # Solr MetricsConsumer
  solr.metrics.url: "http://solr:8983/solr/metrics"
  # solr.metrics.ttl.field: '__ttl__'
  # solr.metrics.ttl: '1HOUR'

  # For SolrCloud, use this settings instead of solr.indexer.url
  #
  #   solr.indexer.zkhost: "http://localhost:9983/"
  #   solr.indexer.collection: docs
  #
  # the same applies for the spout/persistence bolt and the metricsconsumer

  topology.metrics.consumer.register:
       - class: "com.digitalpebble.stormcrawler.solr.metrics.MetricsConsumer"
         parallelism.hint: 1