Introduction#
ObserveIQ is a single observability platform. It covers metrics, logs, traces and continuous profiles, and adds the operational surfaces most teams otherwise buy separately: uptime monitoring, alerting, on-call, service levels, status pages, cloud cost reporting and security posture.
These docs cover installing it, connecting your systems, and operating it day to day.
What you can monitor#
| Area | What ObserveIQ gives you | Guide |
|---|---|---|
| Applications | Traces, service map, response times, errors, profiles | APM setup |
| Linux servers | CPU, memory, disk, network, processes, logs | Linux servers |
| Windows servers | CPU, memory, disk, services, event logs | Windows servers |
| Kubernetes | Clusters, workloads, pods, events, logs, cost | Kubernetes |
| Cloud accounts | Metrics, inventory, audit events, spend | Connecting a cloud account |
| Availability | HTTP, SSL, DNS, TCP and ping checks | SLO page |
How data reaches ObserveIQ#
There are two ways, and most deployments use both.
Send data to ObserveIQ. Your systems ship telemetry using open protocols: OpenTelemetry for traces and metrics, Prometheus remote write for metrics, standard exporters for infrastructure. This is the default and is what enables long retention, alerting and correlation across signals.
Let ObserveIQ query your systems. If you already run Prometheus, Loki, Elasticsearch or a SQL database, attach it as a data source and ObserveIQ queries it when a dashboard is opened. Nothing is copied, so you do not pay to store the same data twice.
You do not have to choose up front. A common pattern is to send the small number of measurements that drive alerts and service levels, and query everything else in place.
What you need before you start#
- An ObserveIQ account, and an organisation you are an owner or administrator of.
- An API key, created under Settings → API keys. This authenticates telemetry you send.
- Outbound HTTPS from your systems to the ingest endpoint, or a network path from ObserveIQ to systems it will query.
Nothing here requires inbound access from the internet into your infrastructure, write access to any of your systems, or an agent on your database servers.
Next steps#
Start with the Quickstart to get one host and one application reporting in about ten minutes. Then work through the integration guide for whatever you actually run.