Resource catalogue
Every entity your telemetry describes, in one searchable list.
What a resource is #
A resource is anything your telemetry describes that is not a service, a host, a log or a span: a container, a pod, a deployment, a cluster, a serverless function, or a datastore your services talk to.
Resource types #
- Host — a machine, identified by its host name.
- Container — identified by its container id.
- Pod — identified by its pod UID.
- Deployment — identified by cluster, namespace and name.
- Cluster — a Kubernetes cluster.
- Function — a serverless function.
- Service — a service acting as a dependency.
- Database, cache and queue — datastores reached from a service.
Resource fields #
- Signals — which of traces, logs and metrics reported this resource.
- Services — the services reporting it, up to 20.
- Owner — the owning team, when every reporter agrees on it.
- Environment — where it runs.
- First and last seen — when it appeared and was last observed in the window.
- Attributes — cloud, container, Kubernetes, host and function metadata.
Owner #
Owner is shown when every service reporting the resource names the same owner. When reporters disagree, the field is left empty rather than picking one.
Datastores #
Databases, caches and queues are derived from the calling side of your spans. Send a db.* span with the server address and peer kind and the datastore appears here with the services that depend on it.
Filtering #
Narrow the catalogue by:
- type
- env
- owner
- service
The catalogue is the fastest way to answer 'what else is on this host'. Filter by owner to find everything a team is responsible for across environments.
Onward #
A catalogue row is a starting point: hosts open in Hosts, containers in Containers, pods in Kubernetes. For the traffic between them, Network flows.
Window #
The catalogue covers the selected window, up to 24 hours. A resource not seen inside the window is not listed.