We're using this at our company. And we wanted to simplify it a bit to not host node-exporter in each pod. Thought we'd might as well publish our changes as a PR. :)
Added YAML manifests to use in-cluster prometheus scraping
Updated script
Writes to /shared_vol/scrape.txt instead
Buffers metrics in memory before writing to file, which resolves #5
Use numfmt instead of hand-made function for parsing data format
We're using this at our company. And we wanted to simplify it a bit to not host node-exporter in each pod. Thought we'd might as well publish our changes as a PR. :)
Added YAML manifests to use in-cluster prometheus scraping
Updated script
/shared_vol/scrape.txtinsteadnumfmtinstead of hand-made function for parsing data formatRemoved old node-exporter-based YAML manifests
Updated README.md