[Meta][Metricbeat] - Collect additional Elasticsearch node metrics for platform analytics
- Dominant language
- Go
- Stars
- 12.7k
- Forks
- 5k
- Avg merge
- 2d 15m
- Merged PRs (30d)
- 385
Description
Platform analytics collects various cloud platform metrics. It used to rely on found-elastic-monitor (FEM) prior to it being deprecated, which gathered node stats including os.cpu.percent that we used to report on.
We would like `os.cpu.percent` to be collected by metricbeat so we can report it again. We're temporarily going to report on `process.cpu.percent`, which is close but not an exact match of `os.cpu.percent`
Contributor guide
Research direction
Start by tracing Metricbeat's Elasticsearch node-stats metric collection and comparing the existing process.cpu.percent path with the deprecated FEM's os.cpu.percent data. Done means Metricbeat emits os.cpu.percent for Elasticsearch node metrics so platform analytics can report the original value.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- elasticsearch, go
- Domain
- observability-sre
- Issue type
- Feature
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 48/100