Hello, thanks for sharing this useful application. I'm wondering if it's possible to add a feature to filter out individual files from being exposed in the path labels?
This would be helpful to prevent inadvertent data leakage, while also helping to limit the cardinality. An alternative could be to add a label to the metrics which would indicate if a path is a file or a directory (eg: node_disk_usage_bytes{type="file", path=...}), and then these can be filtered at scraping time.
Thanks again!