kibana-sidecar

A sidecar for Kibana to apply a prepared configuration

Docs

We write documents like the following to the .kibana index (see the config folder for details):

Index Pattern

ID format: index-pattern:{uuid}

Example:

{
  "_index": ".kibana",
  "_type": "doc",
  "_id": "index-pattern:d308c4f0-157e-11e8-8862-f19ddd0b5982",
  "_source": {
    "type": "index-pattern",
    "updated_at": "2018-02-19T15:59:07.941Z",
    "index-pattern": {
      "title": "fluentd-*",
      "timeFieldName": "@timestamp",
      "fields": "[...]"
    }
  }
}

The fields attribute is a string, actually containing escaped JSON code. It is an array of field definitions, like this:

[
  {
    "name": "kubernetes.namespace_name",
    "type": "string",
    "count": 100,
    "scripted": false,
    "searchable": true,
    "aggregatable": false,
    "readFromDocValues": false
  }
]

Config

ID: config:6.1.1

Example:

{
  "_index": ".kibana",
  "_type": "doc",
  "_id": "config:6.1.1",
  "_source": {
    "type": "config",
    "updated_at": "2018-02-19T14:29:37.545Z",
    "config": {
      "buildNum": 16350,
      "defaultIndex": "d308c4f0-157e-11e8-8862-f19ddd0b5982",
      "dateFormat:tz": "UTC",
      "dateFormat": "YYYY-MM-DD HH:mm:ss.SSS",
      "dateFormat:dow": "Monday",
      "discover:sort:defaultOrder": "asc"
    }
  }
}