Verify NVMe endurance signals and projection constraints #5

New Issue

2026-08-31T08:03:03Z

xavierk commented

2026-08-31 08:03:03 +00:00

Parent map: Chart Fenris’s persistent TUI monitoring redesign

Question

Which NVMe SMART signals and endurance baselines can defensibly support Fenris’s usage-adjusted theoretical lifespan, and what limitations must the product disclose? Verify the semantics and precision of percentage used, data units written, power-on hours, rated TBW overrides, implied endurance, hourly write-rate histories, changing habits, and evidence-based confidence measures.

Parent map: [Chart Fenris’s persistent TUI monitoring redesign](https://git.bongbetic.com/xavierk/Fenris/issues/1) ## Question Which NVMe SMART signals and endurance baselines can defensibly support Fenris’s usage-adjusted theoretical lifespan, and what limitations must the product disclose? Verify the semantics and precision of percentage used, data units written, power-on hours, rated TBW overrides, implied endurance, hourly write-rate histories, changing habits, and evidence-based confidence measures.

xavierk added this to the Wayfinder: Fenris persistent TUI monitoring redesign milestone 2026-08-31 08:03:03 +00:00

xavierk added the wayfinder:research label 2026-08-31 08:03:03 +00:00

xavierk added a new dependency 2026-08-31 08:03:19 +00:00

#2 Define the persistent observation store and legacy migration

xavierk self-assigned this 2026-08-31 08:04:04 +00:00

xavierk commented

2026-08-31 08:20:50 +00:00

Resolved in NVMe endurance signals and projection constraints.

Answer: Fenris can defensibly project when a host-write endurance baseline would be consumed if the observed usage habit continues; it cannot predict drive failure. Use Data Units Written for reported host-write deltas, prefer a verified exact-model/capacity rated-TBW override, and use wall-clock observation history for the rate. Keep Percentage Used as a separate coarse vendor signal; only derive explicitly labeled implied endurance for values 1–254, never for 0 or saturated 255. Power On Hours is contextual because NVMe permits it to omit non-operational powered states. Projection confidence should be categorical and expose baseline provenance, history duration, interval coverage, recency, representative cycles, horizon agreement, and habit stability—never an accuracy percentage.

The asset specifies DUW's 512,000-byte upward-rounded units, baseline precedence and formulas, gap/counter-discontinuity handling, changing-habit constraints, confidence states, and required user disclosures. It also rejects the current capacity × 600 fallback and the presentation of DUW / Percentage Used as a known endurance rating.

Questions surfaced for the next Wayfinder session: whether to show rated-budget and vendor-wear projections side by side; mandatory TBW provenance fields; exact warming-up/coverage/change thresholds; representation of powered-off, deliberately disabled, and unexplained intervals; stable controller identity; whether regime changes require acknowledgement; whether scenario horizons are sufficient without statistical intervals; and TUI treatment of unsupported DUW, saturated Percentage Used, and counter discontinuities.

Resolved in [NVMe endurance signals and projection constraints](https://git.bongbetic.com/xavierk/Fenris/src/branch/research/nvme-endurance-signals/docs/research/nvme-endurance-signals.md). **Answer:** Fenris can defensibly project when a host-write endurance baseline would be consumed if the observed usage habit continues; it cannot predict drive failure. Use Data Units Written for reported host-write deltas, prefer a verified exact-model/capacity rated-TBW override, and use wall-clock observation history for the rate. Keep Percentage Used as a separate coarse vendor signal; only derive explicitly labeled implied endurance for values 1–254, never for 0 or saturated 255. Power On Hours is contextual because NVMe permits it to omit non-operational powered states. Projection confidence should be categorical and expose baseline provenance, history duration, interval coverage, recency, representative cycles, horizon agreement, and habit stability—never an accuracy percentage. The asset specifies DUW's 512,000-byte upward-rounded units, baseline precedence and formulas, gap/counter-discontinuity handling, changing-habit constraints, confidence states, and required user disclosures. It also rejects the current `capacity × 600` fallback and the presentation of `DUW / Percentage Used` as a known endurance rating. **Questions surfaced for the next Wayfinder session:** whether to show rated-budget and vendor-wear projections side by side; mandatory TBW provenance fields; exact warming-up/coverage/change thresholds; representation of powered-off, deliberately disabled, and unexplained intervals; stable controller identity; whether regime changes require acknowledgement; whether scenario horizons are sufficient without statistical intervals; and TUI treatment of unsupported DUW, saturated Percentage Used, and counter discontinuities.

xavierk closed this issue

2026-08-31 08:20:51 +00:00

xavierk referenced this issue

2026-08-31 08:22:01 +00:00

Chart Fenris’s persistent TUI monitoring redesign #1

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Blocks

#2 Define the persistent observation store and legacy migration

xavierk/Fenris

Reference: xavierk/Fenris#5