Description:

When querying traces from "Observe > Traces" tab (without using any filter) and selecting higher time range (and even using high traces count) then the output traces are received only for last few minutes.

Technically, it is expected to see that because Tempo reads trace from backend block and once it finds the traces which matches the value defined in limit traces, then it stops retrieving the traces. Itself 

Documentation:

 [-] https://docs.redhat.com/en/documentation/openshift_container_platform/4.18/html/cluster_observability_operator/observability-ui-plugins#coo-distributed-tracing-ui-plugin-using_distributed-tracing-ui-plugin

Expectation:

Add Important Note in above said documentation to explain the behaviour about the trace queries and using filters for specific service, Trace Duration, Span Duration, Status and Namespace.