Methodica Technologies · Jan to Apr 2025

Test-results monitoring in Grafana

A Grafana-style dashboard for automated test runs, rebuilt in the browser from my exported InfluxDB sample and checked against the 13 dashboard requirements I documented.

GrafanaInfluxDBMySQLDashboard JSONAlertingRequirements
SituationAutomated test runs were producing data nobody could read at a glance.
TaskGive the team one live view of executions, errors and system utilization.
ActionDocumented 13 dashboard requirements and built Grafana views on InfluxDB test-run data.
ResultExecution health per test bench reads from one screen; this page rebuilds it from my export.

Live demo

Recreated from the exported demo data (a small sample of 20 runs), not a live connection to InfluxDB. Every panel re-queries that sample in the browser when a variable changes.

Panels:
Test runs per day by verdict click a bar to drill down
PassedDoneError
Run duration (s) with alert rule
Run durationAbove thresholdError run annotationNote
Errors by exception class click a row to filter by test
Requirements coverage

    Results

    What the 20 exported runs show when grouped by test case.

    Verdicts by test case

    PassedDoneError

    Average duration by test case

    Completed runs (Done or Passed)Error runs

    How it works

    My work as a Data Analyst at Methodica Technologies, January to April 2025.

    1. Collect test results. Each automated test run is written to the InfluxDB measurement M_BASIC with tags for project, test bench, variant and verdict, and fields for duration and exceptions.
    2. Gather requirements. Documented 13 dashboard requirements with the team, from selective metrics and clear labels to alerts, drill-downs, access control and scheduled reports.
    3. Build panels. Chose panel types by data: stat panels for totals, bar charts for runs over time, tables for exception messages, gauges for utilization.
    4. Add variables and alerts. Template variables filter every panel at once; threshold rules flag slow or failing runs.
    5. Export and reuse. Saved dashboards as JSON, such as a System Utilization dashboard with a MySQL source, 70% and 90% thresholds and a 10 s refresh.

    Data

    First 10 of 20 rows. Times are converted from nanosecond epoch to UTC. The f_int_* flag columns and f_verdict (identical to T_VERDICT) are omitted from the preview.

    Exported dashboard JSON: System Utilization Dashboard