Kubesense

Data Explorer

The Data Explorer is an interactive query builder and visualization tool for your observability data. Use it to explore metrics, logs, and traces — then save your work as dashboard panels.

Data Explorer

Overview

The Data Explorer lets you:

  • Build queries against Metrics, Logs, and Traces data sources
  • Choose from multiple visualization types
  • Configure panel appearance and behavior
  • Export panels to new or existing dashboards

The explorer state is synced to the URL, so you can bookmark or share any query configuration.

Query Builder

Data Sources

Select the data source for your query:

  • Metrics — Query time-series metrics using PromQL
  • Logs — Query log data with filters and aggregations
  • Traces — Query distributed trace data with filters and aggregations

Builder Mode vs Code Mode

The query builder offers two modes for Metrics queries:

ModeDescription
Builder (visual)Select a metric, add label filters, and apply transformations step by step
CodeWrite raw PromQL queries directly, with variable substitution support

Metrics Transformations

In Builder mode, you can apply transformations to your metric queries:

  • Rate — Calculate per-second rate of change
  • Aggregations — Sum, average, min, max, count across label dimensions
  • Top/Bottom K — Show only the highest or lowest K series
  • Rollup — Control how data points are aggregated over time intervals
  • Math operations — Apply arithmetic to query results

Logs and Traces Queries

Logs and Traces queries support:

  • Filters — Filter data using advanced query, MFD, or SPL modes
  • Aggregation functionsrow_count, unique_count, avg, sum, max, min, p99, p95, p90, p75, p50
  • Group By — Group results by one or more attributes
  • Sorting — Sort results by value or time, ascending or descending
  • Pagination — Control the number of results returned

Multiple Queries

Click Add Query to add additional queries to the same panel. Each query is labeled alphabetically (A, B, C, etc.).

Formulas

Click Add Formula to create computed values by combining queries with math expressions. For example:

  • Error rate: B / A
  • Combined throughput: (A + B) / C

Visualization Types

Choose the visualization that best fits your data:

TypeBest For
Time SeriesTrends over time — line and area charts
BarCategorical comparisons
StatSingle-value KPIs and summary numbers
PieProportional breakdowns
Top ListRanked lists of values
TableTabular data display
ListScrollable log entry list

Step Control

Configure the Step interval for data aggregation (e.g., Auto, 30 seconds, 1 minute, 5 minutes). This controls the resolution of your query results.

Exporting Panels

Once you have built and configured a panel, you can:

  1. Add to an existing dashboard — Select a dashboard and the panel is appended to it
  2. Create a new dashboard — Creates a new dashboard with this panel as the first entry