Add topic about missing fields

docs/en/observability/handle-no-results-found-message.asciidoc

@@ -0,0 +1,37 @@
+[[handle-no-results-found-message]]
+= Understanding "no results found" message
+
+To correctly render visualizations in the {observability} UI,
+all metrics used by the UI must be present in the collected data.
+For a description of these metrics, refer to <<metrics-reference>>.
+
+There are several reasons why metrics might be missing from the collected data:
+
+**The visualization requires a metric that's not relevant to your monitored hosts**
+
+For example, if you're only observing Windows hosts, the 'load' metric is not collected because 'load' is not a Windows concept.
+In this situation, you can ignore the "no results found" message.
+
+**You may not be collecting all the required metrics**
+
+This could be for any of these reasons:
+
+* The integration that collects the missing metrics is not installed.
+For example, to collect metrics from your host system, you can use the {integrations-docs}/system[System integration].
+To fix the problem, install the integration and configure it to send the missing metrics.
++
+TIP: Follow one of our quickstarts under **Observability** → **Add data** → **Collect and analyze logs** to make sure the correct integrations are installed and all required metrics are collected.
+
+* You are not using the Elastic Distribution of the OpenTelemetry Collector, which automatically maps data to the Elastic Common Schema (ECS) fields expected by the visualization.
++
+TIP: Follow our OpenTelemetry quickstart under **Observability** → **Add data** → **Monitor infrastructure** to make sure OpenTelemetry data is correctly mapped to ECS-compliant fields.
+
+//TODO: Make quickstarts an active link after the docs are merged.
+
+* You have explicitly chosen not to send these metrics.
+You may choose to limit the metrics sent to Elastic to save on space and improve cluster performance.
+For example, the System integration has options to choose which metrics you want to send.
+You can {fleet-guide}/edit-or-delete-integration-policy.html[edit the integration policy] to begin collecting the missing metrics. For example:
++
+[role="screenshot"]
+image::images/turn-on-system-metrics.png[Screenshot showing system cpu and diskio metrics selected for collection]

docs/en/observability/index.asciidoc

@@ -78,6 +78,10 @@ include::monitor-infra/inspect-metric-anomalies.asciidoc[leveloffset=+2]
 
 include::monitor-infra/configure-metrics-sources.asciidoc[leveloffset=+2]
 
+include::troubleshooting-infra.asciidoc[leveloffset=+2]
+
+include::handle-no-results-found-message.asciidoc[leveloffset=+3]
+
 include::monitor-infra/metrics-reference.asciidoc[leveloffset=+2]
 
 // AWS monitoring

docs/en/observability/troubleshooting-infra.asciidoc

@@ -0,0 +1,21 @@
+[[troubleshooting-infrastructure-monitoring]]
+= Troubleshooting
+
+Learn how to troubleshoot common issues on your own or ask for help.
+
+* <<handle-no-results-found-message>>
+
+[discrete]
+[[troubleshooting-infra-support]]
+=== Elastic Support
+
+We offer a support experience unlike any other.
+Our team of professionals 'speak human and code' and love making your day.
+https://www.elastic.co/subscriptions[Learn more about subscriptions].
+
+[discrete]
+[[troubleshooting-infra-forum]]
+=== Discussion forum
+
+For other questions and feature requests,
+visit our https://discuss.elastic.co/c/observability[discussion forum].