mongodb_exporter VS ebpf_exporter

For our third project, we have percona/mongodb_exporter. Carlos Salguero is the maintainer for this project. He says it is a project easy to get started, and there is not much-complicated logic behind this. It is about running some MongoDB internal commands to get statistics like diagnostic data or replica status, passing JSON to produce metrics from these commands. You use a complete Makefile to start sandbox instances to test almost everything; you don’t have a virtual machine or different MongoDB instances. The issues are in GitHub and Jira. The primary programming language is Go.

I am running MongoDB-exporter and pushing data to Prometheus by default mongotop metric is disabled in MongoDB-exporter which I have enabled using --collect.topmetrics flag but enabling this flag is giving fewer metrics and I don't need those metrics when I run mongotop -- json on mongo node I am getting a couple of metrics that are pretty useful to me (read, write, total time and stuff per dB, collection, etc ..) is there any flag in mongo-exporter that lets me get the get mongotop -- json equivalent metrics in Prometheus (https://github.com/percona/mongodb_exporter) this the exporter

There's also the ebpf_exporter which has modules to let you get per-device IO latency data.

For example, we alert when the database starts to get slow, because there's real user impact to slow database queries. We track disk IO latency, but it's for debugging after we get alerted by database alerts. Because database problems could be so many different things. Query plans changing, CPU saturation, etc. Slow reads/writes is just one cause of database problems.

There also a prometheus exporter for the data if you have prom set up. https://github.com/cloudflare/ebpf_exporter

If you have Linux, there is the ebpf_exporter. We use it to monitor real-time disk IO latency, with basically no overhead to the system.

My current go-to to start with storage monitoring is the ebpf_exporter. It allows you to wrap the underlying kernel functions with timing wrappers, which allows you to gather detailed data on every IO that happens. Gather that up into a histogram, and ship it to a monitoring system for visualization.