Elasticsearch Exporter
Export Elasticsearch service health to Prometheus.
To run it:
make
./elasticsearch_exporter [flags]
Flags
./elasticsearch_exporter --help
es.uri
: Address (host and port) of the Elasticsearch node we should connect to. This could be a local node (localhost:8500
, for instance), or the address of a remote Elasticsearch server.es.all
: If true, query stats for all nodes in the cluster, rather than just the node we connect to.es.timeout
: Timeout for trying to get stats from Elasticsearch. (ex: 20s)web.listen-address
: Address to listen on for web interface and telemetry.web.telemetry-path
: Path under which to expose metrics.
NOTE: We support pulling stats for all nodes at once, but in production this is unlikely to be the way you actually want to run the system. It is much better to run an exporter on each Elasticsearch node to remove a single point of failure and improve the connection between operation and reporting.
Elasticsearch 2.0
Parts of the node stats struct changed for Elasticsearch 2.0. For the moment we'll attempt to report important values for both.
indices.filter_cache
becomesindices.query_cache
indices.query_cache
becomesindices.request_cache
process.cpu
lostuser
andsys
time, so we're now reportingtotal
- Added
process.cpu.max_file_descriptors