The best observability platforms, according to hackers
16 observability platforms recommended in Hacker News comments over the past year, ranked by how many separate people recommend each one.
- Most recommendedGrafana71 people recommend it
- Least regretSigNoz19 recommend, 2 criticize
- Free and lovedPrometheus48 people recommend it
All observability platforms
- GrafanaFree, open-source dashboard and monitoring tool that people use to visualize metrics, logs and alerts; the main caveat is complexity grows quickly as the stack and data volume scale.71recommend
- PrometheusFree open-source monitoring system people use with Grafana to collect and alert on metrics, valued for simplicity. Caveat: a single node doesn't scale well and needs Mimir or Thanos.48recommend
- DatadogRecommended for fast debugging, log search, dashboards and AI-assisted investigation across infrastructure and services. Most common caveat: costs grow quickly with data volume and host count.40recommend
- SentryError monitoring service that alerts on new issues with commit context; people like it for the self-hosted option and integrations. Self-hosting is hard to install and maintain.31recommend
- SigNozOpen-source, OpenTelemetry-native observability tool with strong dashboards and APM charts that people self-host cheaply. Common caveat: alerting is weaker than the rest.19recommend
- SuperlogRecommended as an observability tool that installs with one prompt, adds logs and proposes bug fixes. Common caveat: it may not be mature enough for production systems.15recommend
- HoneycombPraised as a pioneer of observability that contributed to OpenTelemetry and makes debugging useful, with a free tier. Critics cite limited dashboards and confusing pricing.14recommend
- New RelicRecommended for its NRQL query language, dashboards and generous free tier; the most common caveat is that it generates costly data and charges high per-seat licences.10recommend
- OpenObserveRecommended as a simple, self-contained, performant observability stack that is cheap to run and easy to deploy.9recommend
- ClickStackRecommended as an open-source observability stack that feels close to Datadog. One critic finds setup and management painful.8recommend
- JaegerFree, open-source tracing tool people call solid, simple and easy to self-host. One critic finds it heavy to run.7recommend
- ZabbixFree, mature monitoring tool that people trust for servers, VMs and containers, with native high availability and self-checks. One critic said it scaled poorly.7recommend
- BugsinkFree error tracker that people use as a Sentry alternative and recommend to others.7recommend
- TracewayFree, MIT-licensed observability stack that developers say is easy to self-host, with preconfigured SLOs, exception tracking and OpenTelemetry support. Critics question its novelty and longevity.6recommend
- Grafana TempoOpen-source trace database that people recommend for debugging faster than sifting through logs, especially self-hosted with Grafana. One critic finds it heavy to run.5recommend
- Pydantic LogfireDevelopers recommend it for debugging and optimisation, and say logging has become much better. Critics find graph customisation limited and see it as an unwanted add-on.5recommend