Skip to content

InfluxDB Plugin

Jacob van Walraven edited this page Feb 25, 2020 · 6 revisions

The InfluxDB output plugin stores the generated results into a InfluxDB instance

By default the results will be batched and sent in groups off 200. This can be changed in the plugins configuration to suit each users needs.

If the InfluxDB instance is offline results will be temporarily written to a log file at /tmp/libtrace-bigdata.influxdb. Once back online the application will execute the backlog of results against InfluxDB.

Configuration example:

  influxdb:
    enabled: 1
    # Enable certificate verification
    ssl_verify_peer: 1
    host: http://127.0.0.1
    port: 8086
    database: libtrace
    username: admin
    password: admin
    # Enable to send results to influxdb in batches
    batch_results: 0
    # If batching results, how many to send at a time
    batch_count: 200
    # The retention policy for data stored in influxDB (How long to store
    # data within each influxDB table before purging it).
    # Note: use INF to never purge any data.
    # Available units:
    # ns        nanoseconds
    # u or µ    microseconds
    # ms        milliseconds
    # s         seconds
    # m         minutes
    # h         hours
    # d         days
    # w         weeks
    # format: int_literal duration_unit. e.g. 30d
    retention: inf
    # How many independent copies are stored in the cluster
    replication: 1

Configuration options

enabled

The enabled configuration option is to enable and disable output to kafka.

ssl_verify_peer

The ssl_verify_peer configuration option verifies the SSL certificate is valid.

host

The host configuration option is the address for the InfluxDB REST API.

port

The port configuration option is the port for the InfluxDB REST API.

database

The database configuration option is the database to use.

username

The username configuration option is the username for the database.

password

The password configuration option is the password for the database.

batch_result

Whether to sent results to InfluxDB in batches.

batch_count

If batch results are enabled how many results to send at a time.

retention

How long to keep the results inside influxDB before purging.

replication

How many independent copies are stored in the cluster.