Derive hour observations, day aggregates, and monitoring periods #22

New Issue

2026-08-31T18:24:26Z

xavierk commented

2026-08-31 18:24:26 +00:00

Parent

Implement the Fenris persistent TUI monitoring redesign

Canonical contracts: fenris-redesign spec · acceptance criteria — criterion IDs cited below live in the register.

What to build

Repeated collection runs across a fixture timeline accumulate real habit history in the observation store. Each hour classifies by the named constants — powered-off when the hour's power-on-hours delta is below 90 % of its wall-clock span, active at ≥ 256 MiB written, idle when powered on and sampled below that, unknown otherwise — with the four splits summing to exactly 3600 s. UTC-bounded day aggregates derive monotonically from hour rows (no 23- or 25-hour days ever exist). Coverage — the known share of wall-clock seconds inside monitoring periods — is computed and carried on the rows. Raw samples prune opportunistically to 14 days. Monitoring periods behave: a collection run finding no open period opens one at the run moment, never backdated; wall-clock outside periods is excluded from numerator and denominator; unexplained gaps inside a period keep the aggregate counter delta, remain unknown seconds, and reduce coverage. No absent hour is ever interpolated, estimated, or fabricated.

Acceptance criteria

Hour classification uses exactly the named constants and decision order; disabled time is never an hour state (PR-4)
The denominator is wall-clock seconds inside monitoring periods including powered-off and unknown time; disabled periods excluded from numerator and denominator (PR-5)
Hours and days are UTC-bounded; day derivation from hours is monotonic; no 23/25-hour days (ST-4)
Raw samples pruned to 14 days; hour observations and day aggregates retained indefinitely (ST-5)
No absent hour is ever interpolated, estimated, or fabricated; gaps stay unknown seconds and reduce coverage (FL-3)
A run finding no open monitoring period opens one at the run moment, never backdated (FL-8)

Blocked by

Collector tracer bullet: one collection run from fixtures to the observation store

## Parent [Implement the Fenris persistent TUI monitoring redesign](https://git.bongbetic.com/xavierk/Fenris/issues/20) Canonical contracts: [fenris-redesign spec](https://git.bongbetic.com/xavierk/Fenris/src/branch/main/docs/spec/fenris-redesign.md) · [acceptance criteria](https://git.bongbetic.com/xavierk/Fenris/src/branch/main/docs/spec/acceptance-criteria.md) — criterion IDs cited below live in the register. ## What to build Repeated collection runs across a fixture timeline accumulate real habit history in the observation store. Each hour classifies by the named constants — powered-off when the hour's power-on-hours delta is below 90 % of its wall-clock span, active at ≥ 256 MiB written, idle when powered on and sampled below that, unknown otherwise — with the four splits summing to exactly 3600 s. UTC-bounded day aggregates derive monotonically from hour rows (no 23- or 25-hour days ever exist). Coverage — the known share of wall-clock seconds inside monitoring periods — is computed and carried on the rows. Raw samples prune opportunistically to 14 days. Monitoring periods behave: a collection run finding no open period opens one at the run moment, never backdated; wall-clock outside periods is excluded from numerator and denominator; unexplained gaps inside a period keep the aggregate counter delta, remain unknown seconds, and reduce coverage. No absent hour is ever interpolated, estimated, or fabricated. ## Acceptance criteria - [x] Hour classification uses exactly the named constants and decision order; disabled time is never an hour state (PR-4) - [x] The denominator is wall-clock seconds inside monitoring periods including powered-off and unknown time; disabled periods excluded from numerator and denominator (PR-5) - [x] Hours and days are UTC-bounded; day derivation from hours is monotonic; no 23/25-hour days (ST-4) - [x] Raw samples pruned to 14 days; hour observations and day aggregates retained indefinitely (ST-5) - [x] No absent hour is ever interpolated, estimated, or fabricated; gaps stay unknown seconds and reduce coverage (FL-3) - [x] A run finding no open monitoring period opens one at the run moment, never backdated (FL-8) ## Blocked by - [Collector tracer bullet: one collection run from fixtures to the observation store](https://git.bongbetic.com/xavierk/Fenris/issues/21)

xavierk added the ready-for-agent label 2026-08-31 18:25:12 +00:00

xavierk added this to the Build: Fenris persistent TUI monitoring redesign milestone 2026-08-31 18:25:14 +00:00

xavierk added a new dependency 2026-08-31 18:25:39 +00:00

#21 Collector tracer bullet: one collection run from fixtures to the observation store

xavierk added a new dependency 2026-08-31 18:25:41 +00:00

#24 Migrate the legacy observation history

xavierk added a new dependency 2026-08-31 18:25:42 +00:00

#25 Compute the projection core: baseline precedence, provenance, and the confidence rule table

xavierk added a new dependency 2026-08-31 18:25:45 +00:00

#29 Ship the fenris-monitor helper, polkit policy, and systemd units

xavierk referenced this issue

2026-08-31 18:26:50 +00:00