This ended up being fairly simple; we just fetch per-process IO from procfs, and these values appear to be largely identical to what's reported by the netlink taskstat.
This just reads metrics from /proc/[pid]/io and reports them as part of other process metrics, same as how things like memory usage are reported.
Why is it important?
We want per-process I/O metrics.
Checklist
[x] My code follows the style guidelines of this project
[x] I have commented my code, particularly in hard-to-understand areas
[x] I have added tests that prove my fix is effective or that my feature works
What does this PR do?
Part of https://github.com/elastic/beats/issues/7461
This ended up being fairly simple; we just fetch per-process IO from procfs, and these values appear to be largely identical to what's reported by the netlink taskstat.
This just reads metrics from
/proc/[pid]/io
and reports them as part of other process metrics, same as how things like memory usage are reported.Why is it important?
We want per-process I/O metrics.
Checklist
CHANGELOG.md