Lens, dashboards and saved objects

1. Lens #

The visualization editor.

Principle #

Lens suggests the right visualization when you drag a field: a time histogram for @timestamp, top values for a text field, a metric for a number. You can then change the type (bars, areas, table, gauge, heat map, pie chart) without starting over.

Good practice: The Lens formula lets you compute ratios (error rate, cache hit rate) without precomputing in Elasticsearch.

2. Dashboards #

Interactive and shareable.

Good layouts #

• At the top: key indicators (Metric); in the middle: trends; at the bottom: detail tables.
• Add controls (lists of services, environments) to filter the whole dashboard.
• Use drilldowns to go from a panel to Discover or to another filtered dashboard.
• Organize dashboards by space (Spaces) and by tags.

3. Saved objects as code #

Export, import, versioning.

API #

# Exporter tous les dashboards et leurs dépendances
curl -s -X POST localhost:5601/api/saved_objects/_export -u elastic:$PASS \
  -H 'kbn-xsrf: true' -H 'Content-Type: application/json' \
  -d '{"type": ["dashboard"], "includeReferencesDeep": true}' > dashboards.ndjson

# Importer dans un autre Kibana (ou un autre espace)
curl -s -X POST 'localhost:5601/s/equipe-web/api/saved_objects/_import?overwrite=true' -u elastic:$PASS \
  -H 'kbn-xsrf: true' -F file=@dashboards.ndjson

Versioning the .ndjson file in Git and importing it from CI lets you promote a dashboard from staging to production and restore it after a mistake.

Good to know: The /s/<space> prefix targets a Kibana space; without a prefix, it is the default space.