![]() ![]() This tutorial focuses on installing the node exporter completely, but there’s also a quick installation for other tools. II – Installing the Tools NeededĪs a reminder, for our architecture, we are going to need Prometheus, the Node Exporter and Grafana. Now that you have an idea of what a monitoring architecture looks like, let’s install the different tools needed. Note : as part of the configuration, Prometheus can actually monitor itself. The node exporter will run on the port 9100 and Prometheus will run on the port 9090 In our case, the whole stack will be run within the same instance, so there will be no needs to configure any firewall rules. (as a reminder, Grafana has a list of hundreds of dashboards that you import within the UI) ![]() Instead, we are going to use a very powerful feature of Grafana which is the dashboard import. We are not going to build every single panel by ourselves. A dashboard solution, in this case Grafana, displaying metrics gathered from Prometheus.A Node Exporter run as a systemd service that will periodically (every 1 second) gather all the metrics of your system.A time series database, in this case Prometheus made to store the different metrics retrieved by the node exporter.In your architecture, you will have the following components: Now that we know what the existing tools for linux monitoring are, let’s have a look at what we are going to use today : the node exporter and Prometheus.Īs stated before, Prometheus scrapes targets and the node exporter is just one of them. The Linux Performance Monitoring Tools by SolarWinds is a paid tool, but if you want a solution ready to use very quickly, they are a great opportunity for your Linux monitoring. Some examples of those tools are the very established SolarWinds Linux Performance Monitoring Tool (that provides very complete dashboards for your Linux system) or Zabbix with the Metric Collection product. With a complete monitoring system, you can have the security rules handled directly in your dashboarding system (for example Grafana) and you don’t need to provide direct access to your instance to whoever wants to troubleshoot outages.ĭesktop solutions provide a more consistent and probably a more practical solution to system monitoring. Those solutions are great but they have major downsides.īesides being very easy to use, they are often formatted in different ways, making it hard to export them in a consistent way.Īlso, to run those commands, you sometimes need elevated privileges on the system which is not always the case. Some examples of command line tools may be top or htop for the CPU usage, df or du for the disks or even tcpdump for an easy network traffic analysis. They are very known to every system administrator and are often very useful to perform some simple troubleshooting on your instance. There are plenty of command line tools for you to monitor your system. ![]() a – Existing solutionsĪs a system administrator, there are multiple ways for you to monitor your Linux infrastructure. I – Linux Monitoring Basicsīefore building our entire monitoring architecture, let’s have a look at what are the current existing solutions and what problems we are trying to solve with the Node Exporter. Play with prebuilt Grafana dashboards to build 100+ panels in one clickĪt the end of this tutorial, you will be able to build your own monitoring infrastructure and add many more exporters to to it. ![]()
0 Comments
Leave a Reply. |