Come join our live training webinar every other Wednesday at 11am PST and hear LogicMonitor experts explain best practices and answer common questions. We understand these are uncertain times, and we are here to help!
LogicMonitor provides a wrapped version of OpenTelemetry that is pre-configured to forward traces from your instrumented applications to the LogicMonitor platform. We recommend installing this OpenTelemetry Collector before instrumenting your applications, because you will need to reference the endpoint of the OpenTelemetry Collector during instrumentation.
As a best practice, we recommend having a (single) Bearer token associated with a dedicated API-only user for forwarding trace data to the LogicMonitor platform. This bearer token needs, at a minimum, the following permissions:
To create a Bearer API Token in LogicMonitor:
1. From Settings > Users & Roles > Users, choose “Manage” to edit an existing user or “Create a new user”.
2. Select the Token Type “Bearer” when adding the API Token.
1. From the Exchange page > OpenTelemetry Collectors page, click “Add”.
2. In the Install an OpenTelemetry Collector wizard, select a Bearer API Token to use for communication between the OpenTelemetry Collector and the LogicMonitor platform. See Prerequisites.
3. Select the platform that the OpenTelemetry Collector will run on: Linux, Docker, or Kubernetes. Each platform has different install options you can choose from. See the following sections for more information.
Once you’ve installed one or more OpenTelemetry Collectors, you will be able to view and manage the installed OpenTelemetry Collectors via the Exchange > OpenTelemetry Collectors page. See Managing OpenTelemetry Collectors.
Note: OpenTelemetry Collectors added for Kubernetes and Docker platforms will not display in the LogicMonitor UI until they are installed because the installation process creates the Collector within LogicMonitor. OpenTelemetry Collectors for Linux will be listed immediately, though it will be listed as unregistered until installation is complete.
If you are installing an OpenTelemetry Collector on a Linux server, copy the cURL command provided and run it on your Linux server to download the installer.
Note: The curl command to download the Collector binary is only valid the next 2 hours.
Once you’ve downloaded the installer, you’ll need to make it executable (chmod +x installer_file) and then run the executable (./installer_file).
chmod +x installer_file
You can install the OpenTelemetry Collector on Linux as a root or non-root user. For root users, lmotel will run as a service. Otherwise, lmotel will run as a process. The installation path for non-root users is:
# installation_path=/home# status check:
$ ps -ef | grep lmotel
If you select to install on Docker, the UI will provide you with a preconfigured Docker run command for running a container with LogicMonitor’s OpenTelemetry Collector Docker image.
The best way to install an OpenTelemetry Collector (or several OpenTelemetry Collectors) in a Kubernetes cluster is to use the LogicMonitor provided Helm Charts. These Helm Charts run the OpenTelemetry Collector as a replicaset. The UI provides preconfigured Helm commands for adding LogicMonitor charts and installing the OpenTelemetry Collector.
If you want to customize the values, for example to change the namespace the Collector is installed in or the number of replicas, you may do so after copying the commands.
After installing one or more OpenTelemetry Collectors, you need to instrument your applications to emit traces and send the traces to these collectors. This instrumentation can be done using OpenTelemetry client libraries for automatic instrumentation (recommended) or using the OpenTelemetry API and SDK if you prefer to manually instrument. See Sending Traces to LogicMonitor.
In This Article