Grafana Interview FAQs: Quick Answers (2027)

36 Grafana interview FAQs with short answers: data sources, variables, PromQL panels, alerting, silences, provisioning YAML, auth and roles.

Sep 13, 2025 - 15:34
Nov 10, 2026 - 14:44
 0  59
Grafana Interview FAQs: Quick Answers (2027)

The Grafana questions asked most often in interviews cover what Grafana is, data sources, dashboards and panels, variables, alerting (rules, contact points, notification policies, silences and mute timings), provisioning as code, authentication and permissions, and plugins. Short, accurate answers to these are enough for most first-round and screening interviews.

This is a quick-revision FAQ for the night before an interview: short answers, little theory. For longer troubleshooting cases, use our Grafana scenario-based interview questions, the main Grafana interview page on this site. Facts were checked against Grafana's documentation in November 2026, when the "What's new" index listed Grafana 13.3 as the newest release.

Key takeaways

  • Grafana queries and visualises data that stays in your data sources, such as Prometheus, Loki, Elasticsearch, CloudWatch and SQL databases.
  • Default sign-in is http://localhost:3000 with admin and admin, and Grafana asks you to change the password at first login.
  • Alerting has rules, contact points and notification policies. Silences pause notifications once; mute timings pause them on a schedule. Neither stops rule evaluation.
  • Provisioning loads data sources and dashboards from YAML files in provisioning/datasources and provisioning/dashboards.
  • Variables make one dashboard serve many services, clusters or environments.

Quick reference

TopicOne-line answer
Default port3000
Main config file (deb/rpm)/etc/grafana/grafana.ini
Provisioning file versionapiVersion: 1
Where alerts are routedNotification policy tree, rooted at the default policy
Where notifications goContact points (email, Slack, PagerDuty, webhook and others)
One-off pauseSilence
Recurring pauseMute timing

Basics (questions 1 to 8)

1. What is Grafana?

An open source tool for querying, visualising and alerting on data from many sources. It does not store most of your metrics itself; it reads them from data sources such as Prometheus or Loki and draws dashboards from the results.

2. Why do DevOps teams use Grafana?

One place to see metrics, logs and traces from different systems, shareable dashboards, alerting on the same queries, and a large plugin ecosystem. It is often paired with Prometheus for Kubernetes monitoring.

3. What is a data source?

A configured connection to a backend that Grafana queries, with its URL, authentication and options. Each panel query runs against a data source, and one dashboard can mix several.

4. What is the difference between a dashboard and a panel?

A dashboard is a page of panels arranged in rows. A panel is one visualisation (time series, stat, gauge, table, logs and so on) with its own queries and display settings.

5. What are dashboard variables?

Drop-downs at the top of a dashboard whose values are inserted into queries, such as $cluster or $namespace. Query variables load their options from the data source, so new namespaces appear without editing the dashboard.

6. What are annotations?

Markers on time series panels for events, such as deployments or incidents. Teams often add an annotation from the CI pipeline at each release so a spike can be matched to a change.

7. How do you sign in the first time?

Open http://localhost:3000 (or your configured address) and use admin and admin. Grafana prompts you to change the password, and its docs strongly recommend doing it.

8. What is a Grafana organisation?

An isolated group of users, dashboards and data sources inside one Grafana instance. Many teams use one organisation with folders and permissions instead, because sharing across organisations is harder.

Dashboards and queries (questions 9 to 15)

9. Write a PromQL query for request rate per service.

sum by (service) (rate(http_requests_total[5m]))

Use rate on counters, never the raw counter value. Our Prometheus FAQs cover PromQL in more depth.

10. How do you use a variable in a query?

sum by (pod) (rate(container_cpu_usage_seconds_total{namespace="$namespace"}[5m]))

Enable "multi-value" or "include All" on the variable and use a regex matcher (=~) when several values can be selected.

11. What are transformations?

Steps that reshape query results inside Grafana before display, such as joining two queries, renaming fields, filtering or calculating a new field. They help when the data source cannot do it.

12. A panel shows "No data". What do you check?

The time range, the variable values, the query in the query inspector, the data source health check, and whether the metric name or labels changed. Most cases are a wrong label or an empty variable.

13. Why is a dashboard slow?

Too many panels, wide time ranges at a small step, high-cardinality queries, or many queries repeated by repeating panels. Reduce panels, use recording rules in Prometheus for heavy expressions, and set sensible minimum intervals.

14. What are library panels?

Panels saved once and reused across dashboards. Editing the library panel updates every dashboard that uses it.

15. How do you share a dashboard?

Share a link with the current time range and variables, export the JSON model, or take a snapshot. For version control, export JSON and keep it in Git with provisioning.

Alerting (questions 16 to 21)

16. Explain Grafana Alerting's parts.

Alert rules hold queries, expressions and a condition. Each rule can create several alert instances, one per series. Notification policies route instances by labels to contact points, which define where and how messages go. Silences and mute timings pause notifications.

17. Silence or mute timing: what is the difference?

A silence pauses notifications once, for example during a maintenance window. A mute timing pauses them on a recurring schedule, such as weekends. Both leave rule evaluation running.

18. Why route by labels?

Labels such as team or severity let one policy tree send database alerts to the database team and critical alerts to paging, without duplicating rules.

19. Which alert states are sent as notifications?

Grafana's docs say only alert instances in a firing or resolved state are sent in notifications.

20. How do you avoid alert fatigue?

Alert on user-facing symptoms, require the condition to hold for a period before firing, group related alerts, route by team, and delete alerts nobody acts on.

21. Can you customise notification messages?

Yes. Contact points send alert details by default and support templates to change the message, such as adding a runbook link.

Administration and security (questions 22 to 28)

22. What is provisioning?

Loading configuration from files at startup. Data sources go in provisioning/datasources and dashboards in provisioning/dashboards, so Grafana can be rebuilt from Git.

23. Show a provisioned Prometheus data source.

apiVersion: 1
datasources:
  - name: Prometheus
    type: prometheus
    access: proxy
    url: http://prometheus:9090
    isDefault: true
    editable: false

24. Show a dashboard provider.

apiVersion: 1
providers:
  - name: default
    folder: Platform
    type: file
    disableDeletion: true
    allowUiUpdates: false
    options:
      path: /var/lib/grafana/dashboards

allowUiUpdates defaults to false, which stops UI edits from drifting away from Git.

25. How do you store data source passwords safely?

Put them in secureJsonData, which Grafana encrypts, and inject the values from environment variables or a secret store instead of committing them.

26. Which authentication options does Grafana support?

Built-in users plus external providers such as generic OAuth, LDAP and SAML (some options depend on edition). Most companies connect Grafana to their identity provider and map groups to roles.

27. What roles exist?

Viewer, Editor and Admin at the organisation level, with folder and dashboard permissions on top. Give most users Viewer and grant Editor per folder.

28. What are plugins?

Extensions that add data sources, panels or apps. Install only signed plugins you need and keep them updated, since they run inside your Grafana server.

Configuration details interviewers check (questions 29 to 35)

29. Where is Grafana's main configuration file?

For deb and rpm package installs, /etc/grafana/grafana.ini. In containers, most teams set the same options through environment variables instead of editing the file.

30. Which options does a file-based dashboard provider support?

name (required and unique), orgId (default 1), folder or folderUid, type (default file), disableDeletion, updateIntervalSeconds, allowUiUpdates and options.path. options.foldersFromFilesStructure creates Grafana folders that match your directory layout.

31. What happens if someone edits a provisioned dashboard in the UI?

With allowUiUpdates left at its default of false, Grafana will not save the change over the provisioned file. That is the point: the change should go through Git. If you enable UI updates, the file and the database can drift apart.

32. How many alert instances can one rule create?

One per time series or dimension the query returns. A CPU rule over 40 pods can produce 40 instances, each routed by its own labels. This is why labels on the query result matter for routing.

33. What does the default notification policy do?

It is the root of the policy tree. Any alert that matches no child policy is handled by it, so give it a sensible catch-all contact point and never leave it pointing at a channel nobody reads.

34. How would you give a team its own dashboards without seeing others?

Create a folder per team, grant the team's group Editor on its folder and Viewer elsewhere only where needed. Use separate organisations only when data sources themselves must be isolated.

35. How do you upgrade Grafana safely?

Read the "What's new" and breaking-change notes for each version you skip, back up the Grafana database and provisioning files, upgrade a staging copy, open the busiest dashboards and test alert delivery, then upgrade production.

36. How do you reload provisioned dashboards without restarting Grafana?

A file provider rescans its path on the interval set by updateIntervalSeconds, so new or changed JSON files appear on their own. Keep the interval reasonable on large dashboard sets, and make sure each dashboard JSON has a stable uid so links and alert references survive the reload.

Common mistakes in Grafana interviews

  • Saying Grafana stores metrics. It queries data sources.
  • Using a raw counter in a panel instead of rate().
  • Describing alert rules without saying how notifications are routed.
  • Mixing up silences and mute timings.
  • Building dashboards by hand with no plan to keep them in Git.

How to use this page

Read the quick reference, then answer each question aloud in under 30 seconds. If you cannot, open a local Grafana with Prometheus and try it. For more depth, read the beginner set in most asked Grafana questions for freshers, the certification-focused Grafana certification questions, and our guide to dashboards every DevOps team should build. Grafana and Prometheus are taught hands-on in our Prometheus and Grafana course in Pune, in Kharadi and live online.

Frequently asked questions

What is the default Grafana port?

3000. The default address is http://localhost:3000 unless the configuration changes it.

What are the default Grafana login credentials?

Username admin and password admin. Grafana prompts you to change the password after the first sign-in.

Does Grafana store metrics?

Generally no. It queries data sources such as Prometheus or Loki, which store the data, and visualises the results.

What is the difference between a silence and a mute timing?

A silence pauses notifications once; a mute timing pauses them on a recurring schedule. Neither stops alert rule evaluation.

How do I keep Grafana dashboards in Git?

Export dashboard JSON to a repository and load it with a file-based dashboard provider in provisioning/dashboards.

What is a contact point in Grafana?

The destination and message format for notifications, such as email, Slack, PagerDuty or a webhook.

What is the latest Grafana version?

In November 2026, Grafana's What's new index listed Grafana 13.3 as the newest release. Check the docs for later updates.

References

What's Your Reaction?

Like Like 0
Dislike Dislike 0
Love Love 0
Funny Funny 0
Angry Angry 0
Sad Sad 0
Wow Wow 0
Mridul I am a passionate technology enthusiast with a strong focus on DevOps, Cloud Computing, and Cybersecurity. Through my blogs at DevOps Training Institute, I aim to simplify complex concepts and share practical insights for learners and professionals. My goal is to empower readers with knowledge, hands-on tips, and industry best practices to stay ahead in the ever-evolving world of DevOps.