No description

Find a file

Oliver Neumann 6d04bf0260 Add background job SLO (post-incident)		2025-11-04 08:00:00 +00:00
tests	Add unit tests	2025-06-02 09:00:00 +00:00
README.md	Document burn rate helper	2025-08-11 10:00:00 +00:00
requirements.txt	Initial commit — basic SLO eval working	2025-05-19 08:30:00 +00:00
slo_tracker.py	Add burn rate calculation helper	2025-07-28 17:45:00 +00:00
slos.yml	Add background job SLO (post-incident)	2025-11-04 08:00:00 +00:00

README.md

slo-tracker

Minimal CLI for tracking SLO burn rates from Prometheus. Outputs a quick summary to stdout — useful for morning standups and incident reviews.

Built for my own use. Tested on the Nexus prod setup.

Usage

python slo_tracker.py --config slos.yml --window 7d

Config format

slos:
  - name: "API availability"
    query: 'sum(rate(http_requests_total{code!~"5.."}[5m])) / sum(rate(http_requests_total[5m]))'
    target: 0.999
    window: 30d

Requirements

pip install prometheus-api-client pyyaml rich

Burn rate

The burn_rate() helper calculates how quickly you're spending your error budget relative to a 30-day window. A burn rate > 1 means you'll exhaust the budget before the window closes.