In the previous article I covered how to set up auto-instrumented tracing for a Node.js app using OpenTelemetry (OTEL). We then sent the spans directly to the open source tracing tool Jaeger. I recommend you give that a read first before walking through this guide because we're going to re-use the instrumentation we set up last time. Today we're going to take things a step further by introducing the OpenTelemetry Collector.

Grafana dashboards are powerful and flexible tools for observing applications and infrastructure, so it’s no surprise we get a lot of questions from the community about how to embed them into their web applications. Over the past few releases, we’ve developed a lot of options for how to do this in Grafana, but there can be confusion about how they work, and when to use each approach.

Prometheus is a very popular open-source monitoring and alerting toolkit originally built in 2012. Its main focus is to provide valid insight into system performance by providing a way for certain variables of that system to be monitored. Prometheus displays the performance of these variables as a graph to allow its users to see their system’s performance at a glance.

You may be ready to make the move to Grafana Cloud, but securely querying private data has been a blocker. If you wanted to query a network-secured data source like a MySQL database or an Elasticsearch cluster that is hosted in an on-premises private network or a Virtual Private Cloud (VPC), you needed to open your network to inbound queries from a range of IP addresses.

Using Grafana Cloud to manage and monitor even your most sensitive data from your AWS services just got easier. If your organization’s workloads are hosted in AWS and you are using a Grafana Cloud instance that’s also hosted in AWS, you can now use AWS PrivateLink to establish a secure connection between your virtual private cloud (VPC) network and Grafana Cloud for all your data.