- install k6
- tiny Go app with single endpoint
- k6 script testing the endpoint
- inject some failures
- inject some delays
- structured logging:
-
logrus?zap - grafana
- app running in docker
- loki (/w promtail)
- dashboard for reading the logs
-
- prometheus metrics:
- instrument the app
- setup prometheus
- add dashboard
- traces
- otel with correlation propagation (second endpoint?) -> /unstable calls /stable
- tempo
-
dashboard-> slow traces panel
- monitoring correlations
- traces -> logs
- logs -> traces
- metrics -> logs
- logs -> metrics
- metrics -> traces
- traces -> metrics
- correlate k6s /w monitoring
- is it event possible without k6 x Tempo?
- try prometheus remote write with exemplars on failed checks
- prometheus does not allow non-chronological writes
- metrics collected before fail must be sent before the failure metric+exemplar
- while sending the fail, next metrics must be queued/batched and sent immediately after failure is ack-ed
- what if there is a lot of failures? drop it or send one-by-one?
This repository was archived by the owner on Nov 26, 2024. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 0
paluszkiewiczB/k6-grafana-stack
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
attempt of correlating results of k6 load tests with Grafana stack (Grafana, Prometheus, Tempo, Loki)
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published