Alloy traces otlp configuration issue

[avramenkovladyslav]

In alloy configuration we have this code:

otelcol.receiver.otlp "grpc" {
  grpc {}

  output {
    traces  = [otelcol.processor.tail_sampling.policies.input]
  }
}

otelcol.processor.tail_sampling "policies" {
    decision_wait = "30s"
    policy {
        name = "latency_policy"
        type = "latency"
        latency {
            threshold_ms = 150
        }
    }

    policy {
        name = "probabilistic_policy"
        type = "probabilistic"
        probabilistic {
            sampling_percentage = 20.0
        }
    }

    output {
        traces = [otelcol.exporter.otlp.tempo.input]
    }
}

otelcol.exporter.otlp "tempo" {
  client {
    endpoint = "http://tempo-distributor.monitoring.svc.cluster.local:4317"
    tls {
      insecure_skip_verify = true
      insecure = true
    }
  }
}

But when traces are delivered to Tempo, we have a strange behaviour querying them. Sometimes they are have status <root span not yet received>

We tried configurations with otelcol.processor.batch and another value of decision_wait for otelcol.processor.tail_sampling, but had no success.
Is there anything else about traces that we can add for stability?