SAFETY-52-refused-hold
385
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 Claude
|
090a8ecc35 |
SAFETY-42 round 2: pin the two thresholds at a config that is not the shipped one
checks / unit (pull_request) Successful in 6m7s
THE BLOCKER. At the `delinquent()` call site, `self.tuning.saturation_w` -> `250.0` (-19 B) was a survivor of all 23 suites. It is not an equivalent mutant: two copies of inverter A, B wedged, `saturation_w: 1800` (inside its own schema int(100,2000)) and a load of 1600 W put B's shortfall strictly between the literal and the configured value - head names nobody, the mutant names B. At the shipped 500 both name it, which is exactly why every fixture in part 4 was blind: their gaps are all above both thresholds. It matters beyond a mutation score. `delivery.py`'s MIN_W_IS_SATURATION_W note and the `delivery_hold_s` block in config.yaml both promise an operator, in shipped prose, that there is no fourth copy of a threshold to keep in step and no way for the two to drift - raise `saturation_w` and this check moves with you. A literal at the call site removes exactly that promise, silently, and the promise is why the design was chosen over a `delivery_w` key. ASKING THE SAME QUESTION OF EVERY OTHER NUMBER THIS CHECK READS FOUND A SECOND SURVIVOR OF ALL 23 SUITES: `bms_floor=inv.bms_soc_floor` -> `10.0` (-13 B), invisible because config.yaml's own example states 10 % and every fixture in this file used it. Two arms bracket it: a pack at 15 % against its own 25 % cut-out is cold and must stay unnamed (the literal names an empty pack), and a pack at 8 % against its own 5 % cut-out is wedged and must be named (the literal explains the fault away). One arm on each side leaves no constant at all. `delivery_hold_s` and `is_live()`'s window were ALREADY pinned - the fixtures had to move them to run - and `LINK_STALE_S` has no option behind it, so "pinned across a changed config" is not available for it and is not claimed. DECISIONS section 9 enumerates all of it with the deltas and which check kills which. Verified in the CI image, tree mounted :ro, `python3 -B`, `__pycache__` purged and counted before every run (TEST-15): R5 saturation_w -> 250.0 -19 B killed by part 4b(a) R6 bms_soc_floor -> 10.0 -13 B killed by part 4b(b), both arms R7 arm(self.delivery_cfg) -> DeliveryConfig() -1 B killed (already) R8 is_live(now) -> is_live(now, 3600.0) +8 B killed (already) R9 pack cut-out -> conjoined site floor +105 B killed twice head: 23/23 test files PASS, no bytecode written during the run. No shipped behaviour changes: the only edits to app/ are comments. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
glenn schrooyenandClaude Opus 5
|
c7eb8c6b97 |
SAFETY-42: close two mutation gaps, and correct what makes N=1 inert
The battery is 12 mutations, one freshly copied tree each, purged, `:ro`,
`python3 -B`, byte deltas measured. First round: 9 red, 3 survivors. All three
are now accounted for and the ledger is in DECISIONS §8.
M5 - `share <= 0` relaxed to `share == 0`, +0 B - WAS A REAL GAP AND THE TWO
OBVIOUS CHECKS DID NOT CLOSE IT. `short_by(-2000, 0) is None` and
`short_by(-2000, -100) is None` both stayed GREEN under the mutant: for a charge
share that is UNDER-delivered `share - measured` is negative and the `gap > 0`
test answers None anyway, so neither input can separate the two spellings. The
one that can is a unit moving MORE than its charge share - -2000 W told,
-3000 W measured, `share - measured` = +1000 W - which the mutant reports as a
shortfall. It is not one; it is something else driving that pack, which is
SAFETY-30's and SAFETY-31's question. That input is now the check, and M5 is RED.
MUTATION.md's third failure mode in miniature: the assertion ran, the sentence
was true, and the data could not reach the failure it named.
M11 - `inv.is_live(now)[0]` replaced by a literal `True`, -15 B - was a WIRING
gap the pure checks structurally could not see. `account(live=False, ...)` was
graded as a FUNCTION; nothing graded that `cycle()` asks the question, because no
fixture had a unit whose writes were being rejected. Such a unit still has its
last ACCEPTED command standing on the wire (`write_setpoint` leaves `last_write`
alone on a rejected write) and still measures nothing, so it looks exactly like a
wedged one to every number this check reads - and reporting it would be the
second, louder alarm about a state the allocator has already handled and logged.
`test_delivery.py` part 6 now builds that unit, with `rejected_since` back-dated
past LINK_STALE_S because the fixture runs in two seconds and the timer is
test_inverter.py's to grade. M11 is RED.
M3 - `len(asked) < 2` relaxed to `< 1`, +0 B - IS STILL A SURVIVOR, and it is an
EQUIVALENT MUTANT. With one asked unit `unaccounted` is a subset of a one-element
set, so it is either empty (`not unaccounted` returns {}) or equal to `asked`
(`len(unaccounted) == len(asked)` returns {}); with zero asked units both
spellings return {} as well. No input separates them, so no check can.
⚠️ AND FINDING IT CORRECTED THE MODULE'S OWN PROSE, WHICH IS THE MORE USEFUL
HALF. `delinquent()`'s docstring claimed `len(asked) < 2` was what makes N=1
inert. It is not - the PROPER-SUBSET rule is, and that line is a fast path and a
signpost. A false attribution of the mechanism in the file's own prose is exactly
what this repository keeps finding, and a mutation that survives is how it
surfaced. The line stays, with the equivalence written down beside it the way
circulation.circulating_w() records its own max(0.0, ...).
Also recorded, rather than left implicit: `account()`'s `parked` branch is graded
as a function but its WIRING is not mutated on its own; part 6's reversed-unit
fixture exercises it end to end, which is weaker, and the ledger says so.
Baseline 23/23 green before and after. app/main.py is byte-identical to the
previous commit - both closures are a test and a comment.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU
|
||
|
glenn schrooyenandClaude Opus 5
|
706e625d99 |
SAFETY-42: a unit that stops delivering, and the discriminator the card asked for does not exist
A unit standing at a non-zero share and producing nothing is seen by NOTHING. SAFETY-30's per-unit sign gate needs OPPOSITION on both sides of 250 W and a unit at 0 W is not opposing, it is ABSENT. The site sum still agrees with the meter in direction. Circulating power is exactly zero when one unit does nothing. The only trace left is `reason == "saturated-freeze"`, which clamp.py:93-96 documents as the alarm an operator learns to ignore. It is HW-01's shape, which ran 6 h 25 m with all three layers spectators. ⚠️ THE CARD'S OWN DISCRIMINATOR WAS BUILT FIRST AND FALSIFIED, AND THAT IS THE MAIN CONTENT. SAFETY-42 asks for "saturation persisting past N while |grid error| > X, both derived". Run on the two-master rig it fails in BOTH directions, and no (N, X) fixes either: FALSE POSITIVE `plantsim.Plant.limit()` returns 0 W for a discharge command into a pack at its own floor, and a wedged unit produces 0 W too. Load 2000 W, equal weights, 2000 W per-unit cap, 120 cycles: THE TWO ARMS ARE IDENTICAL ON ALL 120 in grid, site target, every share and every measured watt. Both saturated-freeze on 116 of them, both hold |grid error| over 300 W on all 120 and settle at 1200 W, both read circulating 0.0 W. The predicate's inputs do not differ, so no threshold on them can differ. And "a cold pack at its floor must not alarm" is the card's own AC 2. FALSE NEGATIVE the standing grid error is what a HEALTHY SIBLING ERASES. Load 800 W, same config, B wedged: over the last 30 of 200 cycles the loop is frozen on 30 of 30 while the worst |grid error| is 30 W, with B standing at +770 W and delivering 0 W. Any X above 30 W is silent for the whole run on the exact fault the card describes. WHAT REPLACES IT. Per unit, for a discharge share, `share - measured`; readable because of a reduction that only exists at N>=2: a unit is named while it is short by more than `saturation_w` with NO ACCOUNT for it, and the units in that state are a PROPER SUBSET of the units asked to discharge. THE ACCOUNT keeps the cold pack out and is the only thing that can: the two cases are identical in the power domain, so the discriminator has to be a non-power fact, and there is exactly one - the pack's own SoC against its own `bms_soc_floor`, asked through `peakguard.discharge_permitted()` rather than transcribed. ⚠️ `bms_soc_floor` ALONE, not `pack_floor_pct()`'s conjunction with `peak_soc_floor`: that floor bounds a money-saving discharge only and using it would explain away a wedged unit anywhere between the two. THE PROPER SUBSET keeps a whole-site cause out - a derate this add-on cannot see hits every unit; one box that stopped listening does not - and it is what makes this INERT AT N=1, by algebra rather than by a roster guard. IT REPORTS. IT DOES NOT ACT, and the strongest argument is for acting. Dropping the wedged unit from `live` genuinely recovers a site the shipped path does not: at load 1500 W, kept in the split the site settles at 800 W with a 700 W error and NEVER comes inside 100 W in 150 cycles (the freeze pins the target, so the sibling's SHARE never grows); dropped, it delivers 1500 W from cycle 14. What refuses it is that the only re-allocation this add-on can express is all-or-nothing: a unit derated to 50 % is short by more than saturation_w with nothing to explain it, so this detector names it too, and at load 3000 W dropping it leaves the site delivering 2000 W against 2400 W kept - 400 W WORSE. A partial re-allocation is what would make acting safe and the allocator cannot express it. That is a card. control.py is NOT MODIFIED. The freeze shapes the loop; this reports on it. NUMBERS, MEASURED. The watts side is `Tuning.saturation_w`, read at runtime, so there is no fourth CFG-03 copy and it cannot drift from the number the control law itself calls "not following". `delivery_hold_s: 185` comes off 2640 closed-loop two-inverter runs in which nothing is faulty (6 gains x 5 slews over the whole schema range, across load reversals, a flapping load and ASYMMETRIC SPLITS - a family SAFETY-30's campaign did not have and the one that dominates here) plus 1890 runs at a roster of ONE as the control: threshold reversals flapping weighted N=1 control * 500 W 20.0 s 0.0 s 30.0 s 0.0 s The N=1 column is 0.0 s at EVERY threshold in all 1890 runs, and the flapping column - the family that defeated SAFETY-30's site window - is zero too, because a common stimulus makes both units short at once and the subset stops being proper. ⚠️ The curve is NOT monotone (55.0 s at 250 W against 40.0 s at 100 W): lowering the threshold can turn a proper subset into the whole roster and BREAK the run. ⚠️ THE MARGIN IS MEASURED, NOT FELT, BECAUSE INVERTER B DOES NOT PHYSICALLY EXIST. `plantsim.INVERTER_B` is A COPIED, `measured=False`, so every two-inverter figure above is two copies of A. Sweeping a different second unit: the worst legitimate episode is insensitive to B's RAMP (30.0 s at half A's) and scales with its LAG - 45.0 s at twice A's measured 10 s, 90.0 s at four times it. 185 s is 6.2x the measured worst, 2.06x the four-times-slower case, 6.2x LINK_STALE_S and 4.9x inside the 900 s quarter. Deliberately not 180, which is SAFETY-02's site hold. SAFETY-34. `unit/<name>/delivery_tripped` in the store, one notification id per unit, re-raised on the first cycle of every process, cleared by the Start press. No flat mirror: a rollback lands on a version that drives one actuator, where this check is inert. ⚠️ SCHEMA CHANGE - `delivery_hold_s` joins options: and schema:, so this needs a version bump before it ships. Additive and defaulted, so an existing options blob stays valid. STATED GAPS: charge shares are not graded (no configured charge ceiling exists, so a constant-voltage taper cannot be explained and arming it would fire on every maintenance charge); every unit wedged at once is not caught (the subset is not proper and no instrument here separates that from a plant that cannot). Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
 admin
|
0581648c73 |
Merge pull request 'SAFETY-13: the output freeze gets an exit, and only where it needs one' (#58) from SAFETY-13 into release/1.0
checks / unit (pull_request) Successful in 5m30s
|
||
|
glenn schrooyenandClaude Opus 5
|
f156fadac0 |
SAFETY-13: the bound came from the wrong half of a two-sided measurement
checks / unit (pull_request) Successful in 5m34s
NOTES.md 5.10 measured the 1.78 kW step BOTH ways - "Up: ~18 s from step to deadband. Down: ~23 s" (NOTES.md:1550). SAFETY-39 takes 18 correctly: its bound is a REFUSAL and the looser number refuses less. This branch reused PLANT_SETTLE_S for the opposite job - how long the plant may still be the EXPLANATION for a divergence - where the conservative half is the longer one. ceil(23/2) = 12, not 9, and being short there does not refuse anything, it accuses. The false positive is real and is measured here. One inverter, one compute(), one plant, no second writer anywhere in the harness: the command falls through zero while the pack is still coasting the old way, unasked_power() calls our own power "unasked", the counter reaches 9 while it is still coasting, and the law emits "freeze-released". At the plant's own measured 18 s/23 s shape, over 272 load-step pairs, the 9-cycle bound fires in 62 and the 12-cycle bound in 2. PLANT_SETTLE_DOWN_S = 23.0 is added beside PLANT_SETTLE_S = 18.0, because the plant's settling time is a PAIR and each bound has to say which half it means and why. SETTLE_CYCLES = ceil(PLANT_SETTLE_DOWN_S / LOOP_TICK_S) = 12. It is no longer the reciprocal of SLEW_RAIL_FRAC; that was a coincidence of both reading the same half, and test_control.py now asserts the inequality. _SAT_COUNT_CAP moves with it, 10 -> 13. config.yaml accepts saturation_cycles as int(1,10), so the whole accepted range is now strictly BELOW the bound and the degenerate configuration this branch used to document no longer exists - pinned against config.yaml's own schema line. Three shipped statements were falsified by that measurement and are rewritten, not renumbered - at 12 the residual is smaller, not zero: unasked_power()'s docstring said anything outside the interval "was put there by something other than this loop". It was our own power: eight cycles after a +1990 W command the command is 0 and the pack still reads +522.3 W. control.py said "SO THE BOUND BELONGS TO (b) AND ONLY TO (b)". It fired in case (a). DOCS.md told an operator "if you see freeze-released, look for the second writer" - on a single-inverter install, hunting something that does not exist. Replaced by a section that gives the innocent causes first (a large load or PV swing; a plant slower than the one measured, 8/320 scenarios at 23 s rising to 268/320 at 50 s) and the second writer only when the state STANDS. Round 1's own results are kept and re-graded, not weakened. The release/1.0 sweep had to widen to reach the new bound (3780 cases): 3267 bit-identical, 432 releases, 27 command changes - all still charge starts from prev_w == 0 - and 81 differing in the saturation counter alone, which nothing downstream can read. The card's case releases on the twelfth saturated cycle instead of the ninth and settles at the same -1190 W against the same frozen integrator. Also corrected, because they were claims and did not reproduce: the coverage table's "output freeze" row was 11, and the mutation form its own note describes gives 6 on this tree, 6 on the first SAFETY-13 commit and 2 on release/1.0; the detector row was left at 11 and is 15; the frozen = False footnote was left at 10 and is 13. And section 6 reported 107 625 as "released states" when it is the number of states SWEPT - 67 500 of them release, and both are counted now. Nothing here reached the inverter. 22/22 test files pass in goodwe-addon-ci, tree rsynced to a plain path, mounted :ro, python3 -B, __pycache__ purged and printed first (0 dirs / 0 .pyc; the worktree carries none). Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
adminandClaude Opus 5
|
6c4565659b |
SAFETY-13: the output freeze gets an exit, and only where it needs one
checks / unit (pull_request) Successful in 5m30s
While the loop is frozen the command's magnitude may fall but never rise. At prev_w == 0 there is no magnitude left to fall, so a charge cannot START. That refusal is deliberate and it is kept: the inverter reporting power we did not command means another writer is on the actuator, which is the §1.5 non-goal. What it did not have was an END - nothing cleared it but the other writer leaving. Reproduced on release/1.0 @ |
||
|
admin
|
c8dcb38a95 |
Merge pull request 'SAFETY-37: a stale reading is None, and arrival cannot be what says so' (#57) from SAFETY-37 into release/1.0
|
||
|
glenn schrooyenandClaude Opus 5
|
8e241a3243 |
SAFETY-37: correct a wall-clock tick and a false size claim
checks / unit (pull_request) Successful in 5m32s
Two blocking findings from review, both in prose, both corrected at every site. 1. "trips on tick 7" was not what the fixture prints. It prints 11, five runs out of five, and the developer's own report says 11. ⚠️ But the deeper problem is that the number should never have been quoted: the tick is counted on the WALL CLOCK inside cycle() (hold_s 2 s, ticks 0.20 s apart), so it is load-dependent - 7 is what a box slow enough to stretch each tick to ~0.29 s would print - and nothing pins it. The DECISIONS transcript now says 11 because it is a transcript of a run; the CHANGELOG stops quoting a tick at all and states the checkable thing instead: it trips without the gate and never with it, which part 3 grades. 2. ⚠️ SURVIVOR 2's SIZE CLAIM WAS FALSE FOR ONE OF THE TWO CONSUMERS THIS CARD EXISTS TO PROTECT. It read "one contributing at most 250 W to a loop whose reporting floor is 700 W" at three sites. That bounds the COMMAND. Once the register is frozen its REPORTED value is unrelated to the command - which is exactly HW-01's shape: commanded ~0 W, register stuck on the last large number. Measured in review: a unit frozen at -1200 W beside a healthy +900 W sibling yields circulating_w = 1800 W against the 700 W floor, 2.6x OVER, with no gate armed. So the scoping is TRUE for the per-unit sign check and FALSE for circulation. Corrected to say exactly that, and the circulation consumer's exposure below MOVE_W is now recorded as a KNOWN GAP rather than a covered one. It survived because part 9's S2 check asserts only `not stale` and part 4 pairs the frozen unit with a RESTING sibling (circulating_w([A_FROZEN_W, 0.0]) == 0.0) - neither grades the size sentence. No behaviour change. Full suite: 22 passed, 0 failed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
glenn schrooyenandClaude Opus 5
|
6864109a01 |
SAFETY-37: a stale reading is None, and arrival cannot be what says so
checks / unit (pull_request) Successful in 5m34s
The architect pass made SAFETY-37, TEL-06 and TEL-08 one systemic problem - "a stale reading holds its last value and reads healthy" - and this builds the single policy. Both consumers the card names are gated by one change, and neither consumer needed a line: `SignWatchdog.evaluate` already CLEARS its window on a None measurement and `circulating_w` already returns None for a roster with a missing term. Neither could ever receive one. The gate is the piece that was missing BETWEEN the shipped read and the shipped checks. ⚠️ ONE CRITERION IN THE CARD COULD NOT BE IMPLEMENTED AS WRITTEN, AND SAYING SO IS THE MAIN CONTENT. The card says "only arrival separates them". At this layer arrival cannot: `read_inputs()` polls Home Assistant's state cache about once a second and HA answers 200 with a number whether or not the publisher behind the entity is alive, so both members of the pair arrive identically for ever. test_freshness.py part 1a BUILDS the arrival-age gate the card asks for and grades it failing - the age is 0.0 s at every instant of both runs, a threshold that can never fire, which is TEL-06's own defect one layer up. The half of the rule that survives is honoured absolutely: nothing here reads a Home Assistant timestamp, and the check asserts the module contains no `last_updated`. So the criterion became: a reading is stale when it has not moved through a change it was OBLIGED to follow. STALE <=> unchanged >= 35 s AND |standing| >= 250 W AND |standing - standing when the reading last changed| >= 250 W Latched until a different value arrives. `Inverter.batt` becomes None - never a held number - which is a missing input, which §11.1 rule 1 answers with 0 W. WHY A CONJUNCTION AND NOT A THRESHOLD, MEASURED 2026-09-06. Replaying the committed 30-day capture of this house's own 1 W power entity (485,841 samples out of HA's recorder, where a row exists only on a CHANGE, so the gap between rows IS the unchanged duration every consumer saw): the longest legitimate unchanged run is 263.2 s, held at -899.0 W. A false-positive-free threshold would have to exceed that, and 263 s is 3.5x too slow for a consumer that trips at 75 s. There is no admissible threshold. Every one of those long runs is under an UNMOVING load, which is what the travel term keys on. THE BOUND, derived at both ends and graded against the shipped config.yaml. Floor 17.0 s = 6.0 plant settle (NOTES §5.9, measured 2026-08-22) + 10.0 ESPHome publish (firmware/goodwe-master.yaml:181) + 1.0 read period (run_control's asyncio.sleep(1)). Ceiling 75.0 s = min(circulation_hold_s 75, unit_sign_disagree_s 85). Band [34.0, 37.5]; 35 is a CHOICE inside a DERIVED band and the check says so - the first draft claimed the band held one integer and went red on its own first run. MOVE_W is the LARGEST admissible value, unit_sign_disagree_w, because below it the consumer cannot speak at all. SoC is STATED, NOT WIRED, and BOUNDS carries it with wired=False and the reason. Its obligation term is energy through the pack, not command travel - a rested pack holds a percent for hours under a 0 W command - and K needs a usable capacity nobody has measured. That is TEL-08's; the record says so. ⚠️ BEHAVIOUR CHANGES AT THE SHIPPED ROSTER OF ONE, and it is named rather than hidden: a FROZEN register now stops the site where it used to read as a healthy measurement for ever. That is HW-01, which was a one-inverter site. A healthy install is unchanged, asserted over sixty reads of a dithering register under a reversing command. A false verdict fails toward 0 W behind the existing stale_input_s hold and clears itself on the register's next move. ⚠️ BOTH UNITS IN EVERY TWO-INVERTER FIXTURE ARE INVERTER A with different numbers passed in; inverter B does not physically exist (~2026-09-12). MUTATION: 16 mutations against a purged, :ro, python3 -B tree, fingerprinted before and after every revert. One survivor, and it is the finding - dropping the magnitude term left every check green because no fixture separated "the command travelled 250 W" from "the command ended up above 250 W". Closed, and recorded as MUTATION.md habit 17. A second process defect is habit 18: a mutation run that timed out between the edit and the revert left the module mutated on disk and nothing announced it. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
admin
|
a4290d39ac |
Merge pull request 'SAFETY-39: slew_w is a ratio to the rail, and gain is the constant that is still wrong' (#56) from SAFETY-39 into release/1.0
|
||
|
glenn schrooyenandClaude Opus 5
|
834e7ca44b |
SAFETY-39: correct the gain figures, which were not the measured ones
checks / unit (pull_request) Successful in 5m15s
Review found validate.py:288-289 carrying numbers that appear nowhere else in the PR's evidence and could not be reproduced: over-absorption "1233 W at slew 1000 and 1344 W at slew 250 ... 1495 W at 0.6, 961 W at 0.3, ~0 W at 0.2". A grep for 1233/1344/1495 found them ONLY in that comment - the DECISIONS record, the PR body, test_control.py's audit table and slew_sim's own output all carried the correct ones. The measured values, from re-running tools/slew_sim.py on the real plant model at the stated conditions (max_w 5000, 2 s tick, 51 export steps 1.0-4.2 kW): slew column 958 (slew 1000) and 1081 (slew 250); gain column 987 (0.6), 876 (0.3), 435 (0.2). ⚠️ "~0 W at 0.2" WAS WRONG IN THE UNSAFE DIRECTION. It invited a future reader to conclude that dropping gain to 0.2 removes over-absorption. It does not: 435 W remains, which is 44 % of the figure at 0.6, not zero. The corrected text says so at the line rather than leaving a round number to be trusted. The conclusion the paragraph draws is unchanged and still holds on the measured figures: over-absorption is flat in slew_w and falls in gain, so `gain` is the constant that would have to move, and SAFETY-39 deliberately does not move it. No behaviour change; comment only. Full suite: 21 passed, 0 failed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
adminandClaude Opus 5
|
5d7e275df4 |
SAFETY-39: slew_w is a ratio to the rail, and gain is what is still wrong
checks / unit (pull_request) Successful in 5m16s
`slew_w: 1000` was derived on 2026-08-22 against a clamp of 1500 W, where "capping how far the loop can jump ahead of the inverter" was true because the rail itself bounded the excursion. The live install went to max_w 5000 and the constant did not move, so the loop crosses its whole output range in 10 s against a plant NOTES.md §5.10 measured at ~18 s. Measured on the owner's system 2026-09-05 11:43-11:46: a 3.4 kW export drove the setpoint 380 -> 5000 W in five 2 s ticks, the battery to -5154 W past a 5000 VA rating, and the grid from 3.4 kW of EXPORT to 2.8 kW of IMPORT. Nothing was broken; the rail moved. REPRODUCED FIRST, and one half of it is not reproducible at all. On plantsim.Plant and on an independent in-repo model: the setpoint stepping a full 1000 W per tick to the rail (48 of 51 export steps), and the meter crossing from export into import. The -5154 W figure is NOT reproducible in either model, because both clamp the plant at its rating - so the card's AC "battery no longer exceeds max_w in simulation" is unfalsifiable and was rewritten rather than met. THE DELIVERABLE IS THE RELATIONSHIP. SLEW_RAIL_FRAC = LOOP_TICK_S / PLANT_SETTLE_S = 2 s / 18 s = 1/9 slew_w = 0 now means FOLLOW THE RAIL - SLEW_RAIL_FRAC * max_w - exactly as integrator_max_w = 0 already means follow max_w. It is the shipped default: 222 W at max_w 2000, 556 W at 5000, 1067 W at a ~9600 W two-inverter site. validate.py rule 10 refuses an explicit value faster than that, graded against sum(unit_caps(opts)) - the roster sum, i.e. Controller.site_max_w - so a second inverter moves the bound with no operator action. VERIFIED BY BREAKING IT: restoring slew_w 1000 at max_w 5000 is refused, and the refusal names the rail, the crossing time, the measurement and what to do. Equality, one watt of slack, the live interim 400 and slew_w 0 at every rating all still start. GAIN IS THE CONSTANT THAT IS STILL WRONG, and it is NOT changed here. Over 51 export steps the over-absorption is FLAT in slew (958 W at 1000, 1081 W at 250) and monotone in gain (987 W at 0.6, 876 at 0.3, 435 at 0.2): slew_w bounds the OUTPUT, while the integrator moves gain*error per cycle and reaches a 5000 W rail in under three ticks whatever the slew is. Two control-law constants moving in one release on a running house makes the next measurement uninterpretable. ⚠️ N=2: the site rail becomes sum(max_w) ~ 9600 W (inverter B FROM DATASHEET AND SERIAL, not measured, and sim/plantsim.py's INVERTER_B is A copied), and the enforced ratio produces 1067 W there automatically. But a growing rail LOOSENS this bound, and the >= 900 W symptom goes stale at N=2 the same way the constant did - both stated in the record rather than left to be discovered. Nothing reached the inverter. The live install still runs the interim slew_w 400 applied by hand on 2026-09-05, which rule 10 accepts unchanged. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
admin
|
24c3d87d60 |
Merge pull request 'SAFETY-38: pin site_standing with shares that cannot answer for the site' (#53) from SAFETY-38 into release/1.0
|
||
|
glenn schrooyenandClaude Opus 5
|
d5b9317d57 |
SAFETY-38: correct the survivor count, which was stated four ways
checks / unit (pull_request) Successful in 5m11s
Review found the summary lines miscounted, and one of the miscounts filed
a KILL among the survivors. The authoritative tables in §7e and §7f were
right throughout; only the sentences a reader actually relies on were
wrong.
Arithmetic, from §7f's own ledger:
14 measured = 10 surviving
+ 1 already caught (C3 allocator.py:162, red in test_alloc.py)
+ 3 caught by round 4 (C6 clamp.py:253, C7 peakguard.py:493,
B12 arbiter.py:264)
Survivors by file: TWO in allocator.py (C1 :154, C2 :160), three in
arbiter.py (C4a/C4b :344, C5 :392), two in tariff.py (C8, C9), two in
arbitrage.py (B17, B18), one in main.py (B19). Two, not three, in
allocator.py - C3 is the kill.
Before: DECISIONS §6c said "Nine" while listing eleven; the CHANGELOG said
"11 SURVIVORS" while listing eleven. Both counted C3 as a survivor.
⚠️ KEPT VISIBLE RATHER THAN QUIETLY FIXED. This is this card's own
recurring defect in prose rather than in code - a claim wider than the
measurement beneath it - and it is the second false statement this ticket
has put into its own decision record. The correction says so at the line,
because a document whose whole value is that it can be trusted should
record when it could not be.
No app/ change; no test change. Full suite: 21 passed, 0 failed.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU
|
||
|
Claude Opus 5
|
e2efc765af |
SAFETY-38 round 4: the enumeration covered one sink, and it was not the one this codebase uses
checks / unit (pull_request) Successful in 5m12s
The scanner's two censuses between them scored ONE syntactic sink and called it
the family. Every one of census B's 16 members was a sentence that is APPENDED,
ASSIGNED to a `reason`-shaped name, or passed as a LOG ARGUMENT. Not one was
RETURNED - and census A's return bucket only ever held a callee formatting its
own PARAMETERS, so a method formatting its own ATTRIBUTES and handing the string
back was invisible to both. That is the whole describe() / explain() /
reason-builder family, which is where this repository actually keeps its
operator sentences. Worse than the gap: it was in NONE of the five blind spots
the tool printed about itself.
THE SCANNER, WIDENED - and it rediscovered both members without being told:
* CENSUS C, the return sink, resolved through the returning method's CALL
SITES so a returned KEY (store.unit_key, mqtt.unit_state_key - MAINT-05's
card) is not mistaken for a returned SENTENCE. 53 returned pairs, 39
reaching an operator, 16 of those sharing a format spec.
* A sentence handed to a constructor as `reason=` or placed under a "text"
dict key is a MESSAGE. That one line of the sink walker was hiding SIX more
caller-built members, census B 16 -> 22.
* Every census B and C row now prints whether the two interpolations share a
FORMAT SPEC - the difference between a transposition that is silent and one
that merely looks broken.
* The blind-spot list is now TWO lists: what it scans (printed from the
constants, so they cannot drift) and WHAT IT DOES NOT SCAN AT ALL, largest
hole first - `%s`-style lazy logging, MEASURED at 31 calls across 8 modules
and each one listed, not asserted.
CLOSED, both zero-byte survivors of a green 21-file suite:
* clamp.py:253 ClampResult.describe(), requested_w <-> applied_w. No
test_*.py mentioned describe() at all. It is the SS15 action-log row
(main.py:1366) AND the ClampWatchdog notification reason (main.py:2452) -
the alarm SAFETY-40 and SAFETY-41 are about. Under the swap the guard
reports a setpoint clamped UPWARD, to +9999 W by a bound whose rating the
same sentence gives as 2000 W. Pinned on the ORDERED PAIR on the contiguous
literal at BOTH consumers, operands read off ctl.target and
hass.writes[-1] - what the loop asked and what the device was sent - not
off the ClampResult the swap lives in. 4 checks red.
* peakguard.py:493 PeakDecision.explain(), projected_w <-> ceiling_w. "peak
guard idle" was in no test_*.py. The row then reads "projects 5013 W under
a 2300 W ceiling" - 2713 W OVER - on a row whose own words say idle and
whose own flag says ok. Pinned on the decision and on the shipped status
row. 3 checks red.
* test_arbiter.py's scope-violation check, whose CLAIM was wider than its
ASSERTION. Measured this round rather than left at source level: 0 bytes,
21/21 green before, red now. 2 checks.
DISCLOSED, NOT CLOSED - 11 measured zero-byte survivors, each with its file
fingerprint and its 21/21 verdict, in the decision doc SS7e-SS7g. Three in
allocator.py's UnitPlan.explain(), three across arbiter.py's two explain()s,
two in tariff.py, two in arbitrage.py's trade reasons, one in main.py's
SoC-floor notification. An enumerated list with honest gaps beats a fifth round
finding a fifth instance.
CORRECTIONS THIS ROUND OWES:
* The idle sentence does NOT reach main.py:3299/3302 - those are the
floor-limited branch, where explain() returns self.reason before the idle
literal is built. It reaches main.py:3313 and, on the release edge, 1951.
1951 is disclosed unpinned.
* SS4's "Every one of the ten is red" sat under a twelve-row table: it is the
ten site_standing entries.
* SS6b said "weaker evidence than A1-A3"; A4 was measured, so SS7d's
"A1-A4 and B1" is the right list and SS6b now says it.
* AND THE PROBE THAT GRADES THESE MUTATIONS WAS THE DEFECT, ONCE. Its first
cut numbered the operands BY ORDER OF APPEARANCE, so after a transposition
the other leaf took the same number and all thirteen mutant renders came
back byte-identical - which reads exactly like "the edit is inert". Keying
the value to the NAME fixed it. Habit 4, in the measuring instrument:
MUTATION.md habit 16 is the new entry, and the count there is corrected to
eleven rather than inflated.
14 mutations measured: fresh full-repo copy each, mounted :ro, bytecode purged
and the purge COUNT printed, the edit re-read from disk, length and sha256
printed on both sides, and every one required to change the EMITTED SENTENCE
and not merely the bytes. 12 survived the pre-round suite, 1 was already
caught, 3 are caught now. Full 21-file suite green on the final tree.
NOT ONE BYTE OF app/ MOVES. This card changes coverage only.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU
|
||
|
adminandClaude Opus 5
|
357cb4f07f |
SAFETY-38 round 3: pin the banner the SHIPPED install actually prints
checks / unit (pull_request) Successful in 5m12s
Round 2 pinned the operand order of `SignWatchdog.evaluate()` at main.py:1620 -
the SITE detector. Part 9c asserts that detector is never armed at a roster of
one, so on every install this add-on has, the PER-UNIT banner is the only
sign-disagreement sentence that can ever be printed, and its call site is a
different line. main.py:1560 was left unpinned.
1. THE SAME ZERO-BYTE DEFECT ONE CALL SITE OVER.
`evaluate(stood, against, ...)` -> `evaluate(against, stood, ...)`:
app/main.py 234 542 bytes before and after, delta 0, fingerprint
61aa413f... -> e57e6026..., bytecode purged, tree mounted :ro, line re-read
from disk. On the round-2 tree the FULL 21-FILE SUITE PASSES.
Nothing behavioural could ever have caught it. `watchdog.disagrees()` is
SYMMETRIC in its first two arguments - `commanded_w * against_w < 0` with an
abs() gate on both - so the mutant takes identical actions on every input.
Not "no fixture happens to separate them": no fixture CAN.
What moves is the sentence. Measured by diffing the suite's own printed
banners between the round-3 tree and the same tree carrying only this edit:
9 emitted banner lines invert, in 3 distinct sentences.
shipped commanded -2000 W against a meter reading +2000 W
swapped commanded +2000 W against a meter reading -2000 W
shipped commanded +1000 W against its own measured power -1000 W
swapped commanded -1000 W against its own measured power +1000 W
shipped commanded -2000 W against a meter reading +4000 W (the N=1 arm)
swapped commanded +4000 W against a meter reading -2000 W
Persisted to `sign_reason` and re-read after the restart the latch exists to
survive; also the action log, `_LOG.error` and `status_payload()`'s headline.
CLOSED by the ORDERED PAIR on both per-unit banners, both operands read off
the OBSERVED CYCLE - `inv.last_write` and `inv.batt` off the objects the
plant moved, never off the arguments the detector was handed, which would
compare the swap against itself:
- part 3, the N>=2 trip: "commanded {stood:+.0f} W against its own
measured power {meas:+.0f} W", plus the persisted namespaced copy;
- part 7, the N=1 controller-driven trip: "commanded {stood:+.0f} W
against a meter reading {grid:+.0f} W", plus the namespaced key AND the
flat rollback mirror;
- a PRECONDITION on each that the two operands are different numbers of
opposite sign - without it a transposition produces the same string and
the check is decoration.
Measured after the fix: RED on four named checks in test_watchdog.py, GREEN
in the other twenty files. M11 (the site swap) re-measured on this tree:
still red. Round 1's and round 2's kills are intact.
2. A FALSE STATEMENT ROUND 2 INTRODUCED, AND IT WAS WORSE THAN REPORTED.
Section 4 said "three of these left the whole suite green - M1, M2 and M3 -
and the remaining seven were caught by SAFETY-30's own N>=2 checks". Round
1's vaguer wording was harmless; naming three made it specific and wrong.
Re-measured mutation by mutation on clean release/1.0 @
|
||
|
glenn schrooyenandClaude Opus 5
|
564ba11c11 |
SAFETY-38 round 2: assert the ordered pair, and correct the M3 attribution
Two blockers, one of them mine to fix and one a mis-measurement to undo. No
`app/` change in either: the shipped `sum` is correct.
1. A ZERO-BYTE SURVIVOR AT THE CALL SITE ONE LINE BELOW THE PINNED VALUE.
Round 1 pinned `site_standing` and never pinned WHERE IT GOES. Swapping the
detector's first two arguments at main.py:1620 -
`evaluate(site_standing, self.grid, ...)` -> `evaluate(self.grid, site_standing, ...)`
- is a ZERO-BYTE edit and left the round-1 tree fully green. The cost is the
banner: the site stands at -500 W of charge into a +2500 W import and the
LATCHING, SITE-STOPPING alarm reads "commanded +2500 W against a meter
reading -500 W" - which instrument read what, inverted, on the one line
somebody woken by it reads first (watchdog.py:658-665).
Round 1's check could not reach it. It asserted `"-500 W"` appears SOMEWHERE
in the sentence; under the swap it still does, IN THE METER SLOT. A substring
test grades PRESENCE, never ARRANGEMENT. It now asserts the ORDERED PAIR:
`f"commanded {sum:+.0f} W against a meter reading {grid:+.0f} W"`. Measured:
red in test_watchdog.py on that one named check, green in the other twenty.
2. THE M3 ATTRIBUTION WAS FALSE, AND HOW IT BECAME FALSE IS THE USEFUL PART.
Round 1 recorded `list(standing.values())[0]` as closed by SAFETY-32's
package-wide pin and told future authors NOT TO RE-COVER IT. Re-measured on
release/1.0 @
|
||
|
adminandClaude Opus 5
|
3033420dd7 |
SAFETY-38: record the CI-06 false red, and where each mutation lands
All ten mutations of `site_standing` are red in test_watchdog.py and in nothing else - the other 19 files are green under every one of them, which is the measurement of how unpinned the line was. One exception, recorded rather than dropped: in the batch run M10 (`sorted(standing.values())[-1]`) also reported FAIL test_peakguard.py. Re-run in isolation three times and once more through tools/mutation_run.sh - four clean passes. The batch script had discarded the failure detail so the check cannot be named; what can be said is that it did not reproduce in four attempts and that no claim in this card rests on it. That is CI-06's file behaving the way CI-06 records. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
adminandClaude Opus 5
|
a8bd2b9330 |
SAFETY-38: pin site_standing with shares that cannot answer for the site
`app/main.py:1598`'s `site_standing = float(sum(standing.values()))` is the
first argument of SAFETY-30's latching, site-stopping sign detector AND the
`commanded_w` handed to SAFETY-18's `looks_frozen()`. It was pinned by nothing.
Coverage only: not one byte of `app/` moves - the shipped `sum` is correct.
WHY THE SUITE COULD NOT SEE IT. `_sign_site`, the only N>=2 site fixture in the
repository, runs shares of about -2500 W EACH - every one individually over its
320 W gate - so sum, mean and max(key=abs) all agree on it. Its values happen to
be equal, and a check comparing two things that happen to be equal cannot see
the difference it names. MUTATION.md habit 4, ninth instance, inside the site
half of the card that names the rule in its own text. At N=1 it is arithmetic
rather than a fixture defect: sum == mean == max == the one share, identically.
THE FIXTURE (test_watchdog.py part 9). Two SIMULATED units rated 220 W and
280 W, both obeying exactly, inverted gain, house importing 2000 W: the site
commands -500 W against a +2500 W meter and NO SINGLE SHARE clears the shipped
300 W gate. Four candidate reductions, four distinct answers, one over the gate
- sum -500, mean -250, max(key=abs) -280, [0] -220 - asserted as a precondition.
The asymmetry is load-bearing, and the smaller unit is first so [0] stays
distinct from max(key=abs). Gate shipped; only the hold is shortened.
BOTH CONSUMERS. 9b drives the same discrimination at SAFETY-18's threshold: the
site rail is 0.95 x 500 = 475 W, so the sum rails it while 250, 280 and 220 do
not. Shipped: BLIND_FROZEN, floor 0 W. Under both surviving mutations: not blind
at all, no floor filed - the battery may keep importing through a quarter whose
billed average is unobservable. The open quarter is backfilled with the stuck
reading so the mutant cannot reach the same action by the coverage route and
mask the difference. 9c asserts the N=1 identity on a real one-inverter tree.
SAFETY-32 already closes `list(standing.values())[0]` structurally and this card
does not claim it. Its disclosed blind spot - semantics are invisible to a shape
pin, permanently - is why the mean and max(key=abs) needed a behavioural set.
AND THE SEARCH FOR THE NEXT SURVIVOR FOUND ONE. `float(self.target)` - the
arbiter's site target instead of the sum of what stands on the wire - is -11 B
and survives all 20 files on untouched release/1.0 @
|
||
|
admin
|
9e42cecc24 | Merge pull request 'CI-06: test_peakguard.py answered two of its checks with the wall clock' (#54) from CI-06 into release/1.0 | ||
|
adminandClaude Opus 5
|
9b270a66a5 |
CI-06: test_peakguard.py answered two of its checks with the wall clock
checks / unit (pull_request) Successful in 4m56s
`test_peakguard.py` is flaky on the required `checks / unit (pull_request)`
context. Reproduced, measured and removed - not widened, and nothing skipped:
the file still carries all of SAFETY-07 and SAFETY-25, and gains a check.
TWO MECHANISMS, both the control law's own gain `900 / remaining` applied to a
wall-clock gap this file never meant to have.
1. THE CROSS-TIME PROJECTION (the "one cycle of the intent form..." check).
`_proj3` was a SECOND `PeakTracker.project(b3.p1.averager, datetime.now())`
taken after the real cycle had already returned, and its result was compared
against a floor decided from the FIRST projection - while the comment three
lines above says both forms run "against the same frozen projection".
`excess = 900 x (projected - ceiling) / remaining` moves at `excess /
remaining` W/s, so at a 1500 W floor the 1.5 W tolerance bought exactly
`remaining / 1000` seconds: 333 ms with 880 s left in the quarter and 6 ms
with 10 s left (measured by bisection, probe in the PR). The gap between the
two statements measured 12.7-39.5 ms over 26 instrumented runs, so the check
was decided by WHERE IN THE QUARTER it ran, about `gap_ms / 900` of the time.
FIX: `_recording_decide` keeps the Projection OBJECT the real cycle graded.
Both forms are now decided from it, the clock is out of the comparison, and
the tolerance is EXACT rather than 1.5 W. Measured on a real cycle: the
delivered form re-decided from that object reproduces the controller's own
floor to 0.000000000000 W, and the identity holds to 0.000000000000 W.
It also gains a check - that re-decide now pins `_kw` as main.py's actual
call rather than an approximation of it.
2. THE FIXTURE'S OWN ROOM (the three `wired_over_limit` checks named on the
card: "the real cycle files the peak claim...", "...and the number written to
the inverter is the guard's floor", "the STATUS PAGE row for a floor-limited
guard..."). `range(0, span + 1, 5)` leaves 0-5 s of the open block uncovered
by construction, and `excess` multiplies an uncovered second back up by
`900 / remaining`:
floor ~ want x (1 - uncovered/900) - net_w x uncovered / remaining
Measured on this tree, shipped: at 12 s of remaining a fixture that asked for
a 1500 W floor delivered 613 W (-59 %), and at 2 s the guard did not engage
at all - which fails all three of those checks together. The four-attempt
retry cannot help, because four attempts take ~40 ms and all land in the same
dying quarter.
FIX: one telegram stamped at `now` (which is what a real meter does and what
the docstring already claimed), plus FIXTURE_ROOM_S - the helper waits out
the tail of a quarter it cannot size a fixture in. 24 s is DERIVED from the
1 % accuracy the helper promises, at this file's smallest fixture and the
measured 0.1 s worst-case lag, not picked:
net_w x lag / (remaining x want) <= 0.01 - lag/900
remaining >= 3500 x 0.1 / (1500 x 0.00989) = 23.6 s
Same probe after the fix: -0.3 % at 12 s, -0.5 % at 8 s, still engaged at 2 s.
NO TIMEOUT WAS WIDENED, no check skipped, no xfail. One tolerance was made
STRICTER (1.5 W -> exact) because removing the clock made exactness possible.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU
|
||
|
admin
|
bbc903b2b4 | Merge pull request 'PKG-06: bump 0.6.0 -> 0.7.0, because the schema changed and nobody bumped' (#55) from PKG-06-version-0.7.0 into release/1.0 | ||
|
glenn schrooyenandClaude Opus 5
|
d8e02ad3f8 |
PKG-06: bump 0.6.0 -> 0.7.0, because the schema changed and nobody bumped
checks / unit (pull_request) Successful in 5m2s
Six merges landed since 0.6.0 was deployed and the version never moved. Disk and box both read 0.6.0, so `ha apps update` would have been a NO-OP: Supervisor keys the built image by `version:`, and the update path installs `version_latest`, which cannot differ from `version` without a bump. Found while pre-checking the deploy, not after it silently did nothing. ⚠️ MINOR RATHER THAN PATCH, BECAUSE THE SCHEMA CHANGED. SAFETY-32 added `bms_soc_floor: int(0,50)` as a REQUIRED per-entry key on the `inverters` roster - no trailing `?` - so a config declaring a roster and omitting it on any entry is refused at startup. Verified against the deployed |
||
|
admin
|
4482cd8201 | Merge pull request 'SAFETY-41: raise the clamp alarm's re-assert 30 s -> 150 s' (#52) from SAFETY-41-reassert into release/1.0 | ||
|
glenn schrooyenandClaude Opus 5
|
3527c8a04c |
SAFETY-41: escalate a sustained engagement, so a runaway is never held
checks / unit (pull_request) Successful in 4m59s
Review found the case I asked it to construct, and it is not a corner case. The rate limit alone DELAYS A RUNAWAY: a benign one-cycle clamp that spends the budget, then a continuous §10.3-shaped runaway, held the runaway's first notification for up to 140 s - one REASSERT_S minus a cycle. On the live install, where ordinary railing spends the budget essentially all the time, a runaway arriving during normal operation would be delayed with near-certainty. That is the "benign noise suppresses the bug signal" shape this module's docstring is written against, and the family the first attempt at this card died of. THE ESCALATION. An episode that has not yet spoken because an EARLIER, unrelated episode spent the budget speaks anyway once it has engaged SUSTAINED_CYCLES = 3 times in a row. The discriminator is run length and the two populations genuinely differ: the live flood was runs of length ONE (engage, de-assert, engage), while a runaway engages every cycle. Measured, and it costs nothing it was protecting: live scatter flood (305 cycles, 1-in-6) 11 -> 11 notifications chatter 1-engaged/3-clean over 8000 s 50 -> 50 notifications pinned 900 s quarter (floor is 5) 6 -> 6 alarms runaway delay after a benign episode 140 s -> 10 s ⚠️ MY COUNTERFACTUAL CHECK COULD NOT FAIL FOR THE DEFECT IT NAMED, and review caught that too. It read `ClampWatchdog().note(True, ...) is True` - a FRESH object, which only exercises the `last_alarm is None` branch, the trivially-true one. A fresh watchdog has no earlier episode to be held back BY. It now SPENDS THE BUDGET FIRST: benign clamp, gap past CLEAR_CYCLES, then a continuous runaway, and asserts the delay is well under REASSERT_S. That is the third instance in this file of the family recorded at line 216 and at the 224 s sweep, and it was guarding the most safety-critical property in the change. Also pinned: `last_alarm` SURVIVING the de-assert. Review built the mutation - if the de-assert also refreshed it, chatter collapses from 50 notifications to 1 and the alarm never fires again, with the suite green. That is the "alarm that never fires" class, a different bug from one that fires later. Now caught. Both new checks verified against the mutations they name: removing the escalation -> 1 FAILED; refreshing last_alarm on de-assert -> 1 FAILED. Full suite: 21 passed, 0 failed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
glenn schrooyenandClaude Opus 5
|
8790c4d38f |
SAFETY-41: rate-limit the RISING EDGE, which is where the flood came from
checks / unit (pull_request) Successful in 5m9s
Review falsified the first version of this change, and the correction is the substance of it. Raising REASSERT_S 30 -> 150 removed NONE of the owner's notifications. The re-assert interval governs only the still-engaged path; the rising edge fired unconditionally. At the measured ~5 s cycle the live pattern - 48 engagements in ~25 minutes, about one cycle in six - leaves ~25 s of clean cycles between engagements, past CLEAR_CYCLES x 5 s = 15 s. So the alarm DE-ASSERTED between almost every engagement and each one returned as a fresh rising edge. Measured on the shipped class over that window: 48 episodes, 48 notifications at REASSERT_S = 30, and still 48 at 150. The interval now bounds the NOTIFICATION RATE rather than the repeat rate within one episode: `last_alarm` survives a de-assert, and a new episode speaks only if the interval has elapsed. Same window afterwards: 51 episodes, 51 engagements, 11 notifications. ⚠️ NOTHING IS DROPPED, AND THE TWO PROPERTIES THAT MATTER ARE PINNED: - a PINNED loop still speaks 6 times in the 900 s billing quarter, against a floor of 5 (part 2, unchanged); - a rising edge from a COLD watchdog still fires IMMEDIATELY, so a runaway beginning from quiet is never held back by a budget spent on an earlier, unrelated episode. That check is the counterfactual and it is the half that would go red if the limit ignored elapsed silence. Every engagement is still counted, every episode still increments, and the action-log row is still written on every engaged cycle. The budget is on notifications, never on evidence. The new pin is written in wall-clock seconds and in the shape the field produced, so it does not move when the constant does. Verified by reverting the rising edge to unconditional: 2 checks go RED naming it. Full suite: 21 passed, 0 failed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
glenn schrooyenandClaude Opus 5
|
1d393ced22 |
SAFETY-41: correct the comment, and fix the check it cites
Review found two false statements in the comment and one weak check. All three are fixed here; the constant itself is unchanged at 150 s and was never in question. 1. "That caps this constant at 900/5 = 180 s" was FALSE ABOUT THE CHECK IT CITED. test_clamp.py part 2 sweeps i = 0..899 and the rising edge at i=0 is free, so it measured `1 + floor(899/R)` - one alarm more than a real quarter earns. Measured: 224 s was GREEN across the whole suite. Anyone raising this to 200 s would have had a green board and a requirement quietly unmet - the same defect recorded at test_clamp.py:216, one notch weaker. The check now grades floor(QUARTER_S / REASSERT_S) directly, so it enforces the ceiling the comment claims for it. Verified: 150 PASS, 180 PASS, 181 FAIL, 224 FAIL. 180 is also the true requirement edge - on the worst sliding 900 s window 180 earns 5 alarms and 181 earns 4. 2. The "60 cycles of the §10.3 runaway" note was attributed to this module and truncated. It is main.py:2479 and reads "60 cycles of the §10.3 runaway produce 0 write-path engagements and 2 loop-side ones". Quoted in full, attributed correctly. 3. "48 engagements in 25 minutes" was left implying a pinned rail. It is roughly one cycle in six - real, and far above the 2-in-60 the guard was sized for, but not continuous. Restated, and labelled as a live-log reading rather than a measurement in this repo. Also recorded: the 1500 W tuning derivation is in the OTHER repo (admin/goodwe NOTES.md, "Simulated 2 kW load step (gain 0.6, clamp 1500)"), so this repo cannot corroborate it and a reader should not go looking here; and MIN_PER_QUARTER = 5 is a test-local integer, not a spec figure - the 900 s quarter is system-wide but the 5 is not, so the 180 s ceiling rests on one number a test author chose. Full suite: 20 passed, 0 failed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
glenn schrooyenandClaude Opus 5
|
3516947b5e |
SAFETY-41: raise the clamp alarm's re-assert 30 s -> 150 s
The owner reported the OUTPUT_CLAMP notification as a persistent annoyance on the live install. It re-asserted every 30 s, all day. ⚠️ THIS IS A REGIME CHANGE, NOT A CHANGE OF MIND ABOUT THE ALARM. 30 s was chosen when a loop-side engagement was rare - clamp.py's own note records "60 cycles of the §10.3 runaway produce 2 loop-side engagements". The live install produced 48 IN 25 MINUTES of ordinary operation, because `max_w` is 5000 W where the loop's tuning was derived at a 1500 W clamp (SAFETY-39). The integrator therefore reaches its bound constantly and the alarm narrated it every 30 s. An alarm that speaks that often about the normal case is how an operator learns to ignore the one that matters, which is the exact failure clamp.py's docstring is written against. ⚠️ 150 s IS THE REQUIREMENT'S OWN EDGE WITH ONE STEP OF MARGIN, not a comfort choice. test_clamp.py part 2 pins the cadence in WALL-CLOCK terms rather than against this constant: a loop pinned on the rail must speak at least MIN_PER_QUARTER = 5 times inside the 900 s quarter that bills the capacity tariff. That caps the constant at 900/5 = 180 s. Measured: 180 s -> 5 alarms per quarter (exactly the required 5, no margin) 150 s -> 6 alarms per quarter 300 s -> 3, and the floor check goes RED 180 s would sit on the boundary, so any later change to the firing logic would tip it red for a reason unrelated to that change. The FLOOR is the thing that must not move, and it does not. ⚠️ WHAT THIS DOES NOT DO. It changes how often the alarm repeats. It does NOT make the alarm correct: the message still calls the integrator bound a BUG SIGNAL and still pairs a request from one moment with a `self.target` read at another. That is SAFETY-40. And it does not stop the loop railing in the first place; that is SAFETY-39. This is a stop-gap and says so at the line. Nothing was suppressed: every engagement is still counted, the action-log row is unchanged, and the §10.3 runaway still raises the alarm - an earlier attempt at this that split the alarm so only `clamped` notified was REVERTED because test_clamp.py:398 proved it would have disabled runaway detection entirely (a runaway presents as `i_clamped`, and control.py:78-81 says `clamped` is unreachable on a default install). Full suite: 20 passed, 0 failed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
admin
|
86eaef2517 | Merge pull request 'TEL-07: a second MQTT device for inverter B, and A keeps the one it has' (#51) from TEL-07 into release/1.0 | ||
|
adminandClaude Opus 5
|
26c5ee8adc |
TEL-07: a second MQTT device for inverter B, and A keeps the one it has
checks / unit (pull_request) Successful in 4m58s
MQTT discovery announced ONE device and hung every entity off one unique_id prefix, so at two inverters B's per-unit telemetry had nowhere to go. It has somewhere now, and not one entity that already exists moved. ⚠️ THE ASYMMETRY IS DELIBERATE. The first roster entry's per-unit rows are announced on THE DEVICE IT ALREADY HAS; every later entry gets its own device, via_device the controller. "One device per inverter" is tidier and is the change that renames every entity on the live install - and because HA keeps the id it composed at FIRST discovery, a rename produces a NEW entity beside a DEAD one: the history stays behind and every dashboard, template and automation naming it goes on pointing at a sensor that quietly stopped updating, including the three live watchdog automations. The composed ids come out symmetric anyway (sensor.goodwe_inverter_a_battery_power / ..._inverter_b_...), so the tidy design's only real benefit is bought here without its cost. * four per-unit rows - setpoint, battery power, SoC, SAFETY-02 sign latch - announced only from a roster of two. At N=1 nothing is announced and nothing is published: no second device, no renamed entity, no added topic. * grid and p1_sample_age_s stay SITE-scoped and are announced ONCE, at every roster length 1-6. One P1, one grid connection, and p1_sample_age_s is a watchdog input - a second copy is a second answer to "is the meter fresh?". * inverter.roster_names() is the one answer to what the units are called, read by build_roster() and by the publisher, which must know the roster before a Controller exists because announcing happens on the CONNACK. * DECISIONS/TEL-07-second-mqtt-device.md records the two designs side by side. VERIFIED BY DIFFING THE ANNOUNCED PAYLOADS, not by reading the code. The release/1.0 baseline in test_mqtt.py was generated by running that commit's module, not transcribed. Roster of one: 9 topics, 0 missing, 0 changed, 0 added. Roster of two: 0 missing, 0 changed, 8 added. MEASURED, NOT ASSERTED. Sixteen mutations through tools/mutation_run.sh; twelve are in the file as a running table, each named with the property it breaks. Three of those properties exist because the table found their ABSENCE - dropping the unit from the state topic, moving `grid` into the per-unit set, and `units=()` in main()'s publisher call were all GREEN on the first run, as was a roster fallback naming two nameless entries `inverter-a`. Part 9 enumerates what the file still cannot see, with sizes. ⚠️ INVERTER B DOES NOT PHYSICALLY EXIST (~2026-09-12). Every roster of two here is SIMULATED, from the datasheet and serial in NOTES.md §1.1b. Nothing about B has been measured and no entity described here has been seen on a broker. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
admin
|
f72c48dbfd | Merge pull request 'UX-01: translations and task ordering for the add-on Configuration tab' (#50) from UX-01 into release/1.0 | ||
|
Claude Opus 5
|
fef6ef20a4 |
UX-01: translate the add-on config, and order it by task
checks / unit (pull_request) Successful in 5m1s
The add-on had NO translations/ directory, so Home Assistant fell back to rendering the raw option KEYS. An operator opening Configuration read `unit_sign_disagree_w`, `saturation_cycles`, `integrator_max_w` - seventy of them, in file order, with no description and no way to tell a constant derived from 1890 closed-loop test runs from their own inverter's rating. ⚠️ PRESENTATION ONLY. Not one default, not one schema range and not one line of behaviour moved. The 175 non-comment lines of config.yaml are byte-identical to release/1.0's, reordered; the only comment lines that changed are the group banners and five sentences whose "above"/"below" the reorder made false. * translations/en.yaml and translations/nl.yaml: a name and a description for all 70 options and all 7 per-entry keys of the `inverters:` repeatable section. The Dutch is written for a Flemish homeowner - afname/injectie, kwartierpiek, capaciteitstarief - not transliterated from the English. * ⭐ Every derived constant says WHERE IT CAME FROM and whether to touch it. `unit_sign_disagree_s` cites its 1890 closed-loop runs (SAFETY-30); `circulation_w` its 924 runs and the measured 400 W knee (SAFETY-31); `meter_freeze_s` its 2399 s capture (SAFETY-18); `bms_soc_floor` register 0x0560. The two numbers that are the operator's - max_w and peak_cap_w - say "THIS IS YOUR NUMBER". `cycle_cost_threshold_eur`, which is neither, says it is an open product question (OPT-08). * config.yaml reordered into task order in BOTH halves, comments carried with their keys: hardware -> meter -> ratings and limits -> tariff -> maintenance -> safety -> advanced. ⚠️ test_inverter.py:1788's trap is PRESERVED: the nested per-entry `max_w:` still precedes the flat one under `schema:`, and `inverters: []` still ships in the `options:` half. Moving the nested block below the flat one is a 1-byte edit; it was run and goes red in test_inverter.py and test_translations.py. * tools/supervisor_shape_probe.py section F MEASURES the result against ghcr.io/home-assistant/amd64-hassio-supervisor:2026.08.0 - Supervisor's own _read_app_translations(), its own UiOptions, and the frontend's label resolution transcribed from the shipped bundle - and prints every rendered row in both languages. `all cases passed`, exit 0. * ⚠️ ONE MEASURED NEGATIVE RESULT, NOT GLOSSED: the per-entry keys inside the `inverters:` section still render RAW on frontend 20250925.1, because a list-of-objects goes to ha-selector-object and the converter sets no `label` and no `translation_key`. The `fields:` translations are shipped and validated - that is Supervisor's own shape for this - but this card does not claim they reach the screen. The `inverters` label and description do. * test_translations.py: a translation naming a non-existent option, or an option with no translation, fails CI - both directions, both languages, top level and per entry. Mutation-tested BOTH WAYS: eleven in-process mutations each required to NAME the key they broke, plus a negative control; six more run on disk through the purge-and-re-read protocol. Two of the eleven are the ones that cost the WHOLE FILE rather than one label - an unrecognised top-level block, and a `network:` value that is not a string. Supervisor drops a file that fails SCHEMA_APP_TRANSLATIONS with only a log warning, so those silently return the entire tab to raw keys; neither is reachable through the `configuration:` coverage, which is why they were hunted for and closed. * tools/translations_lint.py is the hand parser CI needs (no PyYAML in the add-on image) and the probe grades it, key by key and string by string, against Supervisor's own loader. It is strict and RAISES rather than skipping; nine malformed shapes prove it. Full 20-file suite green (19 existing + test_translations.py). Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
admin
|
3993f9f32b | Merge pull request 'SAFETY-32: bms_soc_floor per pack, rule 8 per unit' (#48) from SAFETY-32 into release/1.0 | ||
|
adminandClaude Opus 5
|
19decf433f |
SAFETY-32 round 5: stop closing sites, build an instrument that finds its own scope
checks / unit (pull_request) Successful in 4m59s
Rounds 2, 3 and 4 each closed exactly the survivor they were handed and each time
the reviewer found the same shape one step over: the reduction, the call site, the
sibling argument, and now the sibling MODULE. Round 4's enumeration stopped at the
`Controller` class boundary; round 5's four blockers were reductions in validate.py,
over a list DERIVED from the roster three calls away from it. Closing four more sites
by hand would have produced a fifth rejection, so this round widens the instrument
instead and lets it enumerate.
⚠️ THE BLOCKERS ARE NOT EQUIVALENT MUTANTS - a safety refusal disappears, verified by
calling the shipped validator both ways on the real tree:
roster max_w 5000 (inverter-a) beside 2000 (inverter-b), sign_disagree_w 3000
clean -> 1 problem, "...max_w (inverter-b) is 2000 W, so the
SAFETY-02 sign-disagreement watchdog can never fire"
min(caps[:1]) +4 B -> 0 problems. The site STARTS.
caps_ix[:1] +4 B -> 0 problems.
the SAME roster with the small cap FIRST (2000, 5000)
clean -> 1 problem, naming inverter-a
min(caps[-1:]) +5 B -> 0 problems.
⚠️ AND THE ONE-ROSTER FIXTURE COULD NOT HAVE SEEN THAT. On [5000, 2000] the `[-1:]`
mutant still refuses, because last IS min there. A fixture of N members cannot separate
two reductions that agree on N members - which is why the behavioural half below is a
SET and not a roster.
THE INSTRUMENT (test_peakguard.py part 12j, widened past `Controller`)
Every `.py` under app/ is read off disk and parsed - NO MODULE IS NAMED IN THE CHECK -
and the roster is found by transitive closure from ONE seed, the `inverters` option key:
through returns, through lists ACCUMULATED in a loop over a roster (which is exactly
what `caps`, `roster` and `problems` are, and what round 4's rule could not express),
through attributes and properties, and THROUGH CALL ARGUMENTS, which is how it walks
into peakguard, arbiter, allocator, circulation, maintenance and store without being
told they exist. A slice of a roster stays a roster so a mutation cannot cut the taint;
an index is one unit and does not.
183 roster reductions across 9 modules, out of 23 files parsed:
main 96, allocator 24, validate 19, arbiter 14, circulation 10, maintenance 10,
peakguard 4, inverter 3, store 3. 14 modules reduce over no roster at all.
⚠️ THE HAND-WRITTEN MODULE LIST WOULD HAVE BEEN WRONG IN BOTH DIRECTIONS. It named
watchdog.py, which reduces over no roster at all (SAFETY-30's sign watchdogs are
strictly per-unit), and omitted maintenance.py and store.py, which both do. A vacuous
claim about one module and no claim at all about two others - which is the argument for
computing the scope, made with data.
The pin rediscovers ALL FOUR blockers without being told where to look, and finds a
FIFTH beside them (validate.py:783, rule 5's own message) that no review round named.
Its counterfactual runs at two depths: every enumerated site's expression mutated with
each of the four cuts this card has been defeated by (183 x 4 = 732 spellings, all
reported), and `[:1]` written into app/validate.py at each blocker position with the
WHOLE analysis re-run over the mutated source.
THE SWEEP, AND THE NUMBER THAT MUST NOT BE QUOTED ALONE
All 183 sites mutated to `[:1]`, +4 B each, against the full 19-file suite in
goodwe-addon-ci (:ro, -B, fresh rsync per mutant, 0 .pyc before and after - a read-only
mount cannot write one): 183/183 RED. ⚠️ That is TAUTOLOGICAL on its own, because the
pin makes every slice red. So the 51 whose first red was test_peakguard.py were re-run
with the pin NEUTRALISED:
145 of 183 a behavioural fixture separates the site from a slice of itself
38 of 183 ONLY the structural claim. Behaviourally UNGRADED.
All 38 are listed by site with what a `[:1]` there would cost, in the DECISIONS record
and in part 12j itself: 15 are rows and counts on the operator's page, 3 are
`len(targets)` arithmetic beside a write path that IS graded, 6 are allocator internals
ALLOC-01 reaches but not with enough units to separate. NONE of the 38 is closed
behaviourally by this card and none is counted as covered.
THE BEHAVIOURAL HALF, because a shape pin grades an identity and never a meaning
test_validate.py grows a ROSTER SET for the three rules that reduce over
`unit_caps_indexed()`: widths 2, 3, 4, 5 with the one small cap at EVERY position,
14 rosters x 3 rules = 42 runs of the real validator, each asserting the refusal fires
AND names the unit the number came off. Beside it a COMPUTED family table of twelve
candidates - max, first, last, the mean, [:1], [-1:], [:2], [:3], [1:], [:-1], [::2] -
each compared against what the shipped validator actually answered on the same roster.
Only `min` agrees everywhere; a member no roster kills is printed as a NAMED SURVIVOR.
⚠️ A THIRD RULE WAS READING THE FLAT `max_w`, AND IT IS SAFETY-30's
Found by reading the three branches side by side, NOT by the pin - and said that way
because it is the shape of what the pin cannot see. `unit_sign_disagree_w` was graded
against `_number(opts, "max_w", 2000.0)`, three lines under a branch that had already
stopped doing that. Both directions were measured before the fix:
two 2000 W units + a leftover `max_w: 5000` + unit_sign_disagree_w 3000 -> 0 problems.
The site started, the status page said `armed`, and the per-inverter watchdog could
not trip on EITHER unit. Identical sentence to the blocker, one branch over.
two 5000 W units with NO flat key -> REFUSED against a 2000 W default the operator
never wrote, for a threshold both units can reach.
It takes the same `weakest` the branch above takes and names the same unit. At a roster
of one `unit_caps()` is `[max_w]`, so the shipped single-inverter refusal is the same
comparison, the same boundary and the same sentence.
MUTATION PROOF (each: bytecode purged and counted 0 before and 0 after, the edit
re-read FROM DISK with the new text present and the old absent, byte delta printed)
B1 validate min(caps) -> min(caps[:1]) +4 B RED test_validate.py
B2 validate min(caps) -> min(caps[-1:]) +5 B RED test_validate.py
B3 validate caps_ix -> caps_ix[:1] +4 B RED test_validate.py
B4 validate rule 6 min -> min(caps_ix[-1:]) +5 B RED test_validate.py
B5 validate rule 5 msg -> min(caps_ix[:1]) +4 B RED test_validate.py
B6 validate rule 5b msg -> min(caps_ix[-1:]) +5 B RED test_validate.py
B7 CONTROL: weakest -> max(0.0, weakest) +10 B GREEN, as intended
R3a peakguard max(floors) -> min(floors) 0 B RED test_peakguard.py
R3b peakguard max(floors) -> max(floors[-2:]) +5 B RED test_peakguard.py
R3c main call site packs -> self.packs[:2] +4 B RED test_peakguard.py
R4a main pack_floor_soc -> vals[:2] +4 B RED test_peakguard.py
R4b main pack_floor_soc -> vals[-2:] +5 B RED test_peakguard.py
R4c main pack_floor_soc -> vals[:3] +4 B RED test_peakguard.py
R4d main cycle() any(...) -> self.inverters[:1] +4 B RED test_watchdog.py
B1-B6 are caught by the BEHAVIOURAL set in test_validate.py, before the pin is reached.
B7 is a deliberate no-op control: `max(0.0, weakest)` is `weakest` for weakest > 0, and
it stays green, so the battery is not merely red for everything.
R3a is a ZERO-byte delta, so the shipped function was called both ways as well:
packs floors [15.0, 30.0, 20.0], peak_soc_floor 15
binding_soc_floor(...) = 30.0 the 0-byte mutant would answer 15.0 (+15.0 pp)
-> the site would discharge to 15 % while pack-b's own BMS cuts out at 30 %.
⚠️ PRODUCTION IS DEGENERATE ON THE FLOOR AXIS. SPIKE-04: pack A and pack B BOTH floor
at 10 %, so a differing floor exists ONLY in the fixtures. That is the argument for
keeping them, not for simplifying them - collapsing them deletes the only data in this
repository that can see the reduction at all. Restated in config.yaml's example, in the
probe and in the fixtures.
⚠️ INVERTER B DOES NOT PHYSICALLY EXIST (~2026-09-12). Every second-unit figure here -
4600 W FROM ITS DATASHEET, 15 % illustrative, the whole roster SET - is SIMULATED and
labelled. Only inverter A's 10 % is read off hardware.
REBASED onto
|
||
|
adminandClaude Opus 5
|
8beba0a3e7 |
SAFETY-32 round 4: the same family at the SIBLING ARGUMENT, and the family enumerated
Rounds 2 and 3 hardened the FLOOR axis - the max inside binding_soc_floor(), the roster handed to it, Controller.packs, pack_floor_pct()'s conjunction. peakguard.decide() is handed a SECOND reduction over the same roster beside it, soc=self.pack_floor_soc, and peakguard.py's own docstring names the two as a pair. Three rewrites of Controller.pack_floor_soc were green across all 19 files on |
||
|
glenn schrooyenandClaude Opus 5
|
a4c95f83ce |
SAFETY-32 round 3: a fixture of N packs cannot separate reductions that agree on N
Both blockers were the recurring defect sitting inside the fix written to close it: a check whose TEXT claims more than its FIXTURE can observe. Both survived a green 19-file suite on |
||
|
glenn schrooyenandClaude Opus 5
|
ffbabea8ce |
SAFETY-32 fix round: grade the reduction on packs whose floors DIFFER
AC 6 was not met for the reduction main.py actually hands decide(). The
reviewer removed the per-pack grading twice and the 19-file suite stayed green
both times. No shipped behaviour changes here - app/ is untouched. What changes
is what observes it.
WHY BOTH SURVIVED, and it is this repository's recurring defect landing on this
card's own Decision 2: under any rule-8-valid config every pack's
pack_floor_pct() equals peak_soc_floor, so `floors` is a list of IDENTICAL
numbers and max/min/first/last are four spellings of one answer. The only check
using a cut-out ABOVE the EMS floor - 12c's `_n1_deep` - ran on a roster of ONE,
where every reduction agrees. So no check ever evaluated binding_soc_floor() on
two packs whose floors differ. And the degeneracy is not a defence: this PR's own
argument ("unreachable is a claim about today's rule ordering, and
validate_startup_config() is also sim/inject.py's entry point") justifies writing
the reduction and equally obliges grading it.
* test_peakguard.py part 12c2 - NEW. binding_soc_floor() on a roster whose
per-pack floors ALL DIFFER: four SIMULATED cut-outs 30/25/40/35 %, each
strictly above peak_soc_floor, so max (40), min (25), first (30), last (35)
and the mean (32.5) are five different numbers and the extremes sit at
INTERIOR roster positions. Two packs structurally cannot separate those -
there max==last and min==first - and neither can three; four is the smallest
roster that can. Precondition is its own check (habit 1); the counterfactual
raises ONE pack - the last, neither extreme - past the old maximum and the
answer follows (habit 7); peak_soc_floor is then lifted above every cut-out
to grade the other side of the conjunction.
* test_peakguard.py part 12g - REWRITTEN as behaviour (habit 4). The old
assertion was `"self.packs" in <the unparsed call site>`, which `self.packs[:1]`
still satisfies. It now drives REAL Controller.cycle()s on a two-pack roster,
both packs at the same SoC, and reads PeakDecision.soc_floor - the argument,
echoed back by the shipped decide(). The cut-outs are the only thing varied,
and the SAME PAIR is carried in BOTH ORDERS.
* ⚠️ TWO MORE SURVIVORS, found by this round's own battery and recorded rather
than quietly repaired. (a) The first cut of 12g varied only pack B's cut-out,
so `self.packs[-1:]` (+5 bytes) agreed with the roster on both runs and stayed
green across all 19 files; the swapped pair closes it, and between the two
directions no single-position slice passes all three runs. (b)
binding_soc_floor()'s EMPTY-roster branch rewritten to `return None`
(-43 bytes) was green too - a defensive branch nothing in the suite reached.
Both directions of it are pinned now.
* wired_over_limit() gained a prepare(ctl) hook, inside its retry loop so every
attempt grades the same site.
* In-process mutation rows 22, 22b and 23 record all four survivors.
Nine further source mutations went through the harness on the fixed tree
(binding_soc_floor -> floors[0] / floors[-1], pack_floor_pct conjoining with
min, pack_permissions grading against packs[0], `any` -> `all` on the unknown
floor, Controller.packs reversed, decide(soc=) taking inverters[0].soc, and the
two survivors above). Seven were red on first run; the two survivors are now red
as well. Three of the nine have a ZERO byte delta and still went red, which is
the direction stale bytecode cannot fabricate - and the harness purged anyway.
MANDATORY PROOF, all in the CI image over a tree rsynced to a plain path and
bind-mounted :ro, bytecode entry count printed before and after the purge:
each of the four mutations now takes test_peakguard.py RED, and the unmutated
tree is 19/19 green (twice, plus test_peakguard.py five times in isolation -
card CI-06's flake did not reappear). The zero-byte one was proved to have RUN rather than
assumed: dis.get_instructions LOAD_GLOBALs were [float, pack_floor_pct, any,
min] and binding_soc_floor() returned 25.0 on the 12c2 roster where the shipped
code returns 40.0.
No measurement of inverter B appears anywhere. It does not physically exist.
Every second-pack and fourth-pack figure added here is SIMULATED and labelled at
its definition; only inverter A's 10 % is read off hardware. No hardware was
touched.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU
|
||
|
adminandClaude Opus 5
|
f9e7fbf86d |
SAFETY-32: bms_soc_floor per pack, rule 8 per unit
`bms_soc_floor` describes A BATTERY, not a site. At two packs of different models the cut-outs differ, and one number for both graded the shallower pack against the deeper one's BMS - so a site with cut-outs of 10 % and 30 % under `peak_soc_floor: 20` STARTED, and the guard would command a forced discharge that pack B's BMS, not the operator's floor, is the thing that stops. That is rule 8's own failure mode, reached through rule 8. Scoped to the card's third bullet, per its PM PROOFREAD: the aggregate anti-windup (DEV-02) and per-instance `achievable_w` (CFG-04) were verified as already shipped and are untouched. * config.yaml: `bms_soc_floor: int(0,50)` REQUIRED per `inverters:` entry. Nothing is inherited; the flat key stays the single-inverter shape's own cut-out. A per-entry DEFAULT was declined - a default is one number for every pack, which is this card's defect - and the decision is recorded. * inverter.py: `Inverter.bms_soc_floor`; an entry that states none gets None, not a borrowed number and not 0.0 (the most permissive value the key has). * peakguard.py: `Pack`, `pack_floor_pct()`, `pack_permissions()`, `binding_soc_floor()`. `discharge_permitted()` is unchanged and is still the one place the question is answered. * validate.py: `unit_bms_floors_indexed()` carries the ROSTER index; rule 8 grades EVERY unit and NAMES the one it refuses. Graded with the offender SECOND and with the same two numbers in the other order. * main.py: `Controller.packs`; `decide(soc_floor=...)` is the per-pack reduction of the same option; the floor-limited notification names the pack that stopped the site. `pack_floor_soc` - the emptiest pack, and its None-on-missing behaviour - is untouched. * tools/supervisor_shape_probe.py: RE-RUN and repaired. The required key broke §C and left §D printing PASS on a refusal about a different key. §D is now an isolated counterfactual over an accepted merged baseline, one key varied, asserting the key name. `all cases passed` on 2026.08.0. Stated rather than hidden: under any config rule 8 accepts the per-pack FLOOR reduces to `peak_soc_floor` for every pack. The reduction is written for `unit_caps_indexed()`'s reason and graded on both sides - 1775 rule-8-valid pairs answer the option, the pair rule 8 refuses answers the cut-out. `peak_soc_floor` is still one number for the site, so the "deeper pack needlessly withheld" half is filed as an open question, not closed. Full 19-file suite green. Twelve source mutations through tools/mutation_run.sh: eleven red, one argued survivor. The battery also found `_flat_bms_floor()` defaulting in two places, each covering for the other. No measurement of inverter B appears anywhere: it does not exist yet. Every second-pack figure is simulated and labelled. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
admin
|
67e649ad6b | Merge pull request 'SPIKE-04a: inverter B is 4600 W FROM ITS DATASHEET, and the three documented copies are pinned' (#49) from SPIKE-04a into release/1.0 | ||
|
glenn schrooyenandClaude Opus 5
|
a9993aeda2 |
SPIKE-04a: inverter B is 4600 W FROM ITS DATASHEET, and the copies are pinned
checks / unit (pull_request) Successful in 4m51s
The desk half of SPIKE-04. No control behaviour changes; this is documentation and the check that stops it drifting again. Three places stated inverter B's rating - config.yaml's commented roster example, DOCS.md's, and tools/supervisor_shape_probe.py's "a blob an operator would POST" - and all three said 3000 W, a number nobody had measured or read off anything. The owner supplied B's model and serial on 2026-09-04: a GoodWe GW5048D-ES, serial `55048ESU22AW0992`, max charging and discharging power 4600 W. Serial characters 6-8 are `ESU`, platform 105 / AA55, so SPIKE-01 does not gate this site. All three copies now read 4600. ⚠️ 4600 W IS FROM A DATASHEET AND IS NOT A MEASUREMENT, and the one ambiguity is recorded rather than guessed. The datasheet prints it under BATTERY INPUT DATA, so it is probably DC, while max_w is graded against 0x0518 - the inverter's own AC power. If it is DC the AC ceiling is nearer 4324 W. Neither reading is written in as fact; SAFETY-20's min(max_w, dev_max) already bounds it correctly under either, once dev_max is read off the unit, so `max_w: 4600` is what an operator should configure today. goodwe/NOTES.md §1.1b carries the full record with the date and the method on every figure. A NEW CHECK, BECAUSE A DOCUMENTATION FIX WITHOUT ONE DRIFTS AGAIN. test_validate.py now pins all three documented rosters: findable, in agreement, and equal to 4600 - with an explicit len(...) == 3 guard so that a regex which stopped matching cannot leave "they all agree" green on the two sites still read. It also grades the PROVENANCE beside the number (the datasheet, the 0x0518 it is graded against, and the 4324 W the DC reading would make it) WITHIN 2500 characters of the roster rather than file-wide. ⚠️ The first version of that provenance check searched whole files and was ALREADY VACUOUS on two of them: 0x0518 appeared once in config.yaml and once in DOCS.md before this card, for unrelated reasons, so `"0x0518" in src` was green no matter what happened next to B's rating. That is a check whose data cannot reach the failure it names, in the file that exists to complain about them. It is fixed, and the window is what fixes it - deleting the caveat beside the number now goes red even where the same token survives elsewhere in the file. The windowed version immediately found that the DOCS.md caveat did not name 0x0518 at all; DOCS.md now says it. Four mutations, bytecode purged and counted at 0 before and after each, all RED (two of them 0-byte deltas that change a printed answer): M1 probe's operator blob 4600 -> 3000 0 B RED (value + agreement) M2 DOCS.md's DC/AC caveat deleted -157 B RED (provenance) M3 config.yaml alone 4600 -> 4324 0 B RED (agreement) M4 probe entity renamed, regex finds nothing RED (the vacuity guard) Restored, green again. The probe itself was re-run against Supervisor 2026.08.0's own AppOptions validator with the 4600: all cases passed, exit 0. ⚠️ THE TEST FIXTURES KEEP 3000 W ON PURPOSE AND NOW SAY WHY AT THE LINE. test_inverter.py, test_maintenance.py, test_peakguard.py and test_watchdog.py build a second inverter at 3000 W against A's 5000 W. The only property those checks consume is that B's number is unmistakably not A's, and the real pair is a 400 W gap - narrower than slew_w, inside the range where an off-by-one-unit bug and ordinary control arithmetic produce indistinguishable numbers. Collapsing them onto the real values would buy realism those checks do not use and cost them their separating power, which is exactly what test_peakguard.py's own §10g PRECONDITION exists to prevent. Their comments no longer call 3000 W a hypothesis about B's rating; they call it illustrative, and not a placeholder either - the placeholder era ended on 2026-09-04. Also recorded in the changelog for the cards that own them: 94 % battery-to-AC => ~88 % round trip => ~360 W burned continuously to circulate 3 kW, which is the number SAFETY-31 argues its threshold from without having one; < 13 W night self-consumption per unit, which nothing accounts for; and the A/B asymmetry in opposite directions (~33 % more energy, ~8 % less power), which means MAINT-05's 12 h drain timeout was derived on A and should be re-checked against B. ⚠️ Honest survivors: - goodwe/NOTES.md §1.1b, where all of this is actually recorded, is graded by nothing here. Rewriting its 16 occurrences of 4600 to 3000 and turning every FROM DATASHEET label into MEASURED ON THE UNIT leaves all 19 files GREEN. This CI cannot fetch admin/goodwe - private, and GITHUB_TOKEN is minted for this repository only, as .gitea/workflows/checks.yml already states - so the cross-repo check is not buildable here. - No check can see whether 4600 is the RIGHT number. Everything above pins a datasheet reading against itself. Only the bench half of SPIKE-04 can say whether the unit agrees, and the DC-vs-AC ambiguity is the live example. - The constant is still named B_RATING_HYPOTHESIS in four test files, which reads as a hypothesis about B's rating when it is now an illustrative fixture. It is left alone deliberately: renaming ~20 sites across test_peakguard.py and test_inverter.py would collide hard with the open SAFETY-32 pull request, which rewrites 1255 lines of the first. The comment directly above each definition says what the value is and is not. Suite: 19/19 green in goodwe-addon-ci, mounted :ro, bytecode purged and verified 0 before and after. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
admin
|
3934ad7ea5 | Merge pull request 'MAINT-05: one recalibration per pack, and only one pack moves power at a time' (#47) from MAINT-05 into release/1.0 | ||
|
adminandClaude Opus 5
|
905b6fd2b3 |
MAINT-05 fix round 2: the refusal has to survive the NEXT start, and both marker guards have to be visible
checks / unit (pull_request) Successful in 5m14s
REJECTED because the fix round's own load-bearing sentence was FALSE, and the
whole suite was blind to that. `_verify_moves()` produced the red the reviewer
reproduced - that half stood - but DECISIONS §2 claimed "`ok` gates the marker,
so a misdirected migration is not stamped and the next start runs it again ...
a re-run is recoverable". THE REFUSAL LASTED EXACTLY ONE START.
MEASURED, Controller level, ONE persisted store, roster ["garage","shed"], flat
last_completed 2026-08-07, call site still handing `inverter-a` on both starts:
construction 1: ok=False misdirected=4 moved=4 kept=0 marker=<absent>
construction 2: ok=True misdirected=0 moved=0 kept=4 marker=1
history parked under unit/inverter-a/...; marker stamped after ONE restart.
WHY. `_verify_moves()` graded only `rep.moved`. On start 2 the same wrong call
site finds the four keys already present, so rule 1 routes every one of them to
`rep.kept`, `rep.moved` is EMPTY, the roster loop iterates over nothing,
`misdirected` is empty and `ok` is True. Start 3 then takes the
`from_version >= MAINT_SCHEMA_VERSION` early return FOREVER - including after
the call site is fixed. Loud-and-recoverable became silent-and-permanent after
one restart: the exact state store.py's header says must never happen, reached
by the fix written to prevent it.
WHY THE CHECK MISSED IT - the recurring defect, this time INSIDE the remedy.
Part 9b's recovery check was `migrate_maint_keys(_m5, ROSTER[0], ROSTER)` - the
CORRECTED name. It measures recovery along the one path where the operator has
ALREADY fixed the bug. Nothing in the 19 files called `migrate_maint_keys()`
twice with the SAME WRONG `unit`. The check was true and passed for the wrong
reason.
WHAT CHANGED
1. app/store.py `_verify_moves()` grades `set(rep.moved) | set(rep.kept)`.
Sound by construction: every entry in both lists is a `unit_key(unit,
legacy)` built from the SAME `unit` argument, so grading `kept` grades the
same namespace decision `moved` does. A correct call site's `kept` entries
sit under a roster name and pass; a wrong one cannot launder its namespace by
having already written it. `migrate_unit_keys()` passes `roster=None` and
returns before the loop, so it is untouched.
2. test_maintenance.py part 9c - THE CHECK THAT WAS MISSING. Two starts, SAME
wrong unit, ONE store: the marker is still withheld on the second, the
refusal names the same roster, the flat history is still there to recover
from, and a third start with the CORRECTED name still gets a run (which a
stamped marker would have made impossible). Its PRECONDITION asserts the
SHAPE of the second report - `moved` empty, `kept` full - so it cannot
quietly revert to grading 9b's path. Counterfactual beside it: the same two
calls with the right name are accepted, then skipped as `already at schema`.
3. test_maintenance.py part 9d - THE OTHER GUARD ON THE SAME LINE, and a
survivor nothing in the repository could see. `if rep.ok:` is
`not lost and not misdirected`; every check that reached the marker did so on
a store whose `kept` was empty, so
if rep.ok: -> if rep.ok or rep.kept: (+12 B)
stamped the marker on a migration that LOST A MONTH OF HISTORY - defeating
the older, settled `lost` guard as well as the new one - and ALL 19 SUITES
WERE GREEN on it (measured at
|
||
|
glenn schrooyenandClaude Opus 5
|
65cddf35f1 |
MAINT-05 fix round: grade the migration against the ROSTER, not against the name it was handed
checks / unit (pull_request) Successful in 4m47s
REJECTED because a survivor undoes AC 5, and it survived for the recurring
reason: two fixture values that must be distinguishable were EQUAL.
THE SURVIVOR. app/main.py:282 mutated from `self.inverters[0].name` to `UNIT_A`
left the full 19-file suite green. `inverter.UNIT_A == "inverter-a"`, and every
two-pack fixture in test_maintenance.py named its first roster entry
"inverter-a" - so the two expressions were the same string in every check that
existed. main.py's own comment made the distinction load-bearing and nothing in
the repository could see whether that comment was true.
THE FAILURE IS REAL. `build_roster()` names a declared roster's entries from the
config and numbers a nameless one POSITIONALLY (`inverter-1`), so "inverter-a"
is not even the fallback. On ["garage","shed"] with a flat last_completed of
2026-08-07 the mutant migrates into unit/inverter-a/, pack 0 reads back None,
and both packs come up believing they have never been recalibrated.
AND IT FAILED SILENTLY, which was the worse half: `_verify_moves()` read back
the keys it had just WRITTEN, under the namespace it had just been HANDED, was
satisfied, and MAINT_SCHEMA_KEY was stamped - so the next start skipped the
migration too. A postcondition that reads back the namespace it was handed
cannot detect being handed the wrong one.
WHAT CHANGED
1. THE POSTCONDITION IS GRADED AGAINST THE ROSTER. `_verify_moves()` takes the
names the MACHINES read and reports a namespace that is not among them as
`misdirected` - a new, differently-named failure from `lost`:
* `lost` is "the store took the write and dropped it" - a storage fault,
with a harmless-default case (a dropped `last_completed: None` loses
nothing, because absent and never-recalibrated are the same statement);
* `misdirected` is "the store did exactly as it was told, by a caller that
was wrong". There is no harmless-default version of it - a write nobody
reads is not information whatever its value - so it fails `ok` alone.
`ok` gates the marker, so a misdirected migration is NOT stamped and the next
start re-runs it. Asymmetric on purpose: a re-run is recoverable (rule 1
makes it a no-op, rule 2 left the flat keys to recover from), a silent skip
is not. Part 9b MEASURES the recovery rather than asserting it.
2. `migrate_maint_keys()` TAKES THE ROSTER IN PLACE OF AN `n_units` COUNT. The
count was `len(self.inverters)` at the only call site, so the roster carries
it, and a count beside the names is a second opinion about how many packs
there are.
3. THE FIXTURES CAN NOW REACH THE FAILURE THEY NAME. Parts 9 and 11 run on
["garage","shed"], each behind a PRECONDITION check that the first name is
not UNIT_A - so an edit that makes them equal again goes red instead of
making the claims vacuous. Part 11's fixture date is relative to `now`, not
a literal, so "history preserved" and "history lost" cannot converge as the
file ages past `interval_days`.
MUTATION LEDGER - all red, purged tree, none at +0 B, full detail on the PR:
M1 main.py inverters[0].name -> UNIT_A -16 B 4 checks red
M2 store.py drop `roster=names` from _verify_moves -14 B 5 checks red
M3 store.py drop the misdirected.append -37 B 5 checks red
M4 store.py ok: drop `and not self.misdirected` -25 B 3 checks red
M5 fixture garage -> inverter-a (the trap itself) +4 B 5 checks red
M6 store.py parts[1] -> parts[-1] +1 B 1 check red
FOUND AND NOT FIXED - `migrate_unit_keys()` HAS THE SAME HOLE AND IT IS A LATCH.
It runs BEFORE `build_roster()` (each Inverter restores its latch in its
constructor), so it is handed the constant UNIT_A with no roster to grade
against. On a declared roster a pre-DEV-02 `sign_tripped: true` lands under
unit/inverter-a/sign_tripped, which no unit reads - and
`_grade_store_migration()`, whose whole job is to re-arm exactly this, is
DEFEATED BY THE KEY THE DEFECT WROTE: its "is anything namespaced latched" test
is satisfied by the misdirected value itself. Measured, with the flat-config
counterfactual, and pinned as a running check (test_maintenance.py part 13)
whose assertions state the WRONG behaviour on purpose. It wants its own card:
the remedy is in inverter.py and SAFETY-02's semantics, and a silent expansion
into a safety latch is how a rejected PR becomes a broken one.
Suite: 19/19 PASS, purged tree, bind-mounted :ro in goodwe-addon-ci.
No hardware was touched; every fixture here is SIMULATED and pack B's 3000 W
remains a marked hypothesis.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU
|
||
|
adminandClaude Opus 5
|
1aef19e995 |
MAINT-05: one recalibration per pack, and only one pack moves power at a time
checks / unit (pull_request) Successful in 4m48s
§12's monthly cycle drains a pack to empty and refills it to true its coulomb
counter. There was exactly one `Maintenance` and one un-namespaced history, so
at two packs only the first was ever recalibrated - and nothing prevented the
cycle draining pack A while the loop charged pack B, which is two power-moving
phases on one grid connection pulling against each other.
* ONE `Maintenance` PER ROSTER ENTRY, each with its own schedule, phase and
history under `unit/<name>/...`. `store.migrate_maint_keys()` moves an
install's flat history into the first pack's namespace under
`migrate_unit_keys()`'s three rules - shared in code via `_move_flat_keys()`
and `_verify_moves()`, not restated - on its own marker, because unlike
DEV-02's it runs only where the roster has more than one pack. The flat keys
are kept LIVE as the rollback view.
* ONE MAINTENANCE SLOT PER SITE (`maintenance.arbitrate()`). At most one pack
in a power-moving phase - drain, charge AND hold. The holder is DERIVED from
the packs' own stored phases, never persisted, so a restart mid-drain
re-derives it. Several packs due in the same hour: the slot goes to the one
that has gone longest without a completed cycle; roster order breaks a
numeric tie and nothing else. Nothing preempts a cycle in flight.
* THE DEFERRED PACK IS NOT STARVED AND IS NOT A FAULT. A deferral does not
stamp `last_start_attempt` and does not move the phase, so neither the 12 h
drain nor the 36 h charge timeout is reachable by waiting.
* The claim is filed on the pack whose state machine produced it, and stays
SCOPE_UNIT - ARCH-01's rule.
* `validate.py` rule 6 grades `maintenance_charge_w` against the WEAKEST cap on
the roster now, as its own comment said it must once `maint_unit` stopped
being `inverters[0]`.
* At a roster of one nothing changes, and it is asserted: no new store key, the
migration does not run, the four cycle events are the shipped strings, the
banner is the bare phase.
* RECORDED NOT FIXED: a drain pins its pack while the loop's target is a site
claim, so below the pin the allocator refuses the split and the site parks at
0 W. Pre-existing at
|
||
|
admin
|
2d025eebf7 | Merge pull request 'SAFETY-30: grade each inverter against its own power, and the site's total against the meter' (#46) from SAFETY-30 into release/1.0 | ||
|
glenn schrooyenandClaude Opus 5
|
ae21b71824 |
Merge release/1.0 into SAFETY-30
checks / unit (pull_request) Successful in 4m39s
CFG-04 (#45) merged first and moved the base. Two conflicts, both resolved as the union of the two sides: - app/main.py: an import-block collision. CFG-04 added validate_unit_ranges to the .validate import; SAFETY-30 added AGAINST_METER, AGAINST_OWN_POWER, SIGN_OPTION_DEFAULTS and SignWatchdog to the .watchdog import. Neither side dropped anything, so the resolution keeps both lists whole. - CHANGELOG.md: two adjacent new entries under ## Unreleased. Both kept. Both sides also independently rewrote large parts of test_peakguard.py (60 and 47 insertions against the merge base) and git merged them textually, so the full suite was run on the MERGE RESULT rather than inferred from the two parents: 19 passed, 0 failed, twice. One caveat recorded rather than hidden: test_peakguard.py failed once during the first sequential full-suite run and then passed 10/10 in isolation and in two further full runs on this same tree. It passes on both parents too, so this is the pre-existing flakiness CFG-04 already disclosed, not a merge defect - it is timing- or load-sensitive. It sits on the required CI context and can therefore block an unrelated merge at random, which wants a card of its own. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
admin
|
18f052a071 | Merge pull request 'CFG-04: an operator can declare a second inverter, and the shape was measured against the Supervisor' (#45) from CFG-04 into release/1.0 | ||
|
glenn schrooyenandClaude Opus 5
|
d5c1577212 |
SAFETY-30: pin WHICH thresholds each sign detector is armed with, not just what the numbers are
checks / unit (pull_request) Successful in 4m48s
REVIEW FOUND A MUTATION SURVIVOR THAT UNDOES THE WHOLE PRODUCT OF AC 5, and it
is a coverage defect rather than a behaviour one - the shipped code arms the
detectors correctly today. In `_arm_sign_watchdogs()`, collapsing
inv.sign_wd.arm(self.sign_cfg if one else self.unit_sign_cfg, ...)
to `arm(self.sign_cfg, ...)` throws the re-derived 250 W / 85 s per-unit pair
away and arms every unit with SAFETY-02's meter-facing 300 W / 180 s - and left
ALL 19 FILES GREEN. Reproduced here before fixing anything: fresh per-mutation
tree, no __pycache__, mounted :ro, graded through tools/mutation_run.sh; -31 B;
19 pass, 0 fail.
WHY NOTHING CAUGHT IT. `grep -rn "sign_wd.cfg" test_*.py` found exactly ONE
threshold assertion, part 7's N=1 arm at (300.0, 1.0) - which the mutation
PRESERVES. Nothing at a roster of two read `inv.sign_wd.cfg` at all. The ledger
mutates the VALUES in SIGN_OPTION_DEFAULTS (M10 250->300, M11 325->180, both
correctly red through the config pins) and never the WIRING that delivers them.
Part 3's behavioural checks cannot close the gap either, and the comment now
says why: B is reversed at ~2000 W with short holds, so it trips under EITHER
pair, and a fault big enough and long enough to trip under both thresholds is
evidence about neither.
THE FIX - test_watchdog.py part 3, on the real two-inverter Controller. Every
`inv.sign_wd.cfg` equals the per-unit pair, `site_sign_wd.cfg` equals the site
pair, and each detector's `against` is its own instrument. Graded against
LITERALS, because `== self.unit_sign_cfg` cannot tell "the per-unit pair" from
"whatever this Controller was handed". Both pairs are printed.
⚠️ AND THE FIXTURE HAD TO CHANGE FOR THE PIN TO MEAN ANYTHING - the recursion
worth naming, because it is the same defect class as the one being closed.
`_sign_site` set `sign_disagree_s`, `unit_sign_disagree_s` AND
`site_sign_disagree_s` all to 1 s, with two of the three magnitude gates at
300 W. A bare equality there discriminates on `min_w` alone: a mis-wiring that
delivered the right magnitude with the WRONG HOLD compares EQUAL to the pair it
was supposed to have. A check comparing two things that happen to be equal
cannot see the difference it names (MUTATION.md's third failure mode). The
three pairs are now pairwise distinct IN BOTH COORDINATES - 250 W / 1 s per
unit, 320 W / 2 s site, 300 W / 3 s shipped - a PRECONDITION asserts that
distinctness so a later edit collapsing them goes red rather than quietly
vacuous, and a second PRECONDITION asserts the literals are the pairs this
Controller actually holds.
VERIFIED BY BREAKING IT. Three mutations, each on its own freshly copied tree
with no __pycache__, mounted :ro, through tools/mutation_run.sh:
M14 arm() handed self.sign_cfg at every roster -31 B RED (was the
survivor; the AC 5 WIRING check names the wiring)
M15 right min_w, SHIPPED hold - hold-only mis-wire +78 B RED, same pin
M16 right min_w, SITE hold - hold-only, and it still
trips inside the drive +83 B RED, and the
ONLY red in test_watchdog.py
M16 is the one that says the hold half of the equality bites on its own.
Unmutated tree: 19 pass, 0 fail, purged and fingerprinted.
Ledger rows M14-M16 and the write-up in DECISIONS/SAFETY-30-per-unit-sign.md
section 5; a CHANGELOG bullet.
NOT IN THIS PR, so it gets a card rather than a quiet fix: a stuck per-unit
power sensor would false-trip the new per-unit check at N>=2. Already-merged
SAFETY-31 has the identical exposure on the same fixture. Pre-existing N>=2
class, unreachable at the deployed roster of one.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU
|
||
|
glenn schrooyenandClaude Opus 5
|
22c016fadd |
CFG-04 fix round: the probe now measures the key it is cited for
checks / unit (pull_request) Successful in 4m33s
Review blocked this on ONE finding, and it was in the EVIDENCE, not the code.
The design and the shipped behaviour are unchanged.
THE INSTRUMENT DID NOT MEASURE WHAT FOUR DOCUMENTS SAID IT MEASURED
tools/supervisor_shape_probe.py section B validated the RECONSTRUCTED LIVE
options blob - which is missing many required keys that have nothing to do with
this card. Its refusal therefore named whichever of THOSE the validator reached
first: the committed transcript recorded `meter_freeze_s`, the reviewer's run
printed `p1_phase_export_entities`, and a run today printed `sign_disagree_w`.
The string those documents quoted - `Missing option 'inverters' in root` - was
never produced by the instrument.
And section B asserted a bare `not ok`. Any refusal at all satisfied it, so
deleting `inverters: []` from config.yaml left it printing PASS: the probe could
not detect the one regression it is the committed evidence for.
THE UNDERLYING CLAIM IS TRUE AND IS NOW MEASURED PROPERLY
Section B is an ISOLATED COUNTERFACTUAL now: the full shipped `options:` block
(expect accept), and the SAME block with `inverters` removed and nothing else
varying (expect refusal). One key differs, so the refusal has nowhere else to
come from - and the assertion is on the KEY NAME IN THE REFUSAL STRING, so the
line goes red if the refusal is ever about a different key. The probe also exits
non-zero when any case fails; without that a transcript is only evidence to
someone who reads it.
The old live-blob case survives as section B2, relabelled for what it is: an
incomplete reconstruction whose RAW half is refused for unrelated missing keys,
kept because its MERGED half genuinely demonstrates what App.options is.
VERIFIED BY BREAKING IT, WHICH IS THE WHOLE POINT
`inverters: []` deleted from config.yaml, probe re-run against
ghcr.io/home-assistant/amd64-hassio-supervisor:2026.08.0:
== B. THE LOAD-BEARING ONE: an options blob with no `inverters` key ==
FAIL the shipped `options:` block validates AS IT STANDS - ...
Missing option 'inverters' in root in GoodWe RS485 Controller
...
== 2 CASE(S) FAILED ==
exit 1. Key restored, same image, same command: `== all cases passed ==`, exit
0. The transcript in DECISIONS is that restored run, re-run today, not the old
one.
THE FOUR MIS-CITATIONS
DECISIONS/CFG-04-inverter-roster.md section 4's "section B shows both halves"
sentence and its line-103 transcript; CHANGELOG.md; test_inverter.py's part-9
comment and its part-12 mutation row; and the PR description. All now attribute
the string to section B and say plainly what section B2 does not show. The
part-12 row additionally records that the probe was blind to that mutation until
review, because a mutation table that lists a defeat nothing catches is the
failure this repository keeps finding.
MUTATION.md gains it as the sixth instance of "the check's data cannot reach the
failure it names" - the first found OUTSIDE the suite, in a committed evidence
artifact, because nothing in that document had ever been pointed at tools/. Three
habits with it: assert on the CONTENT of a refusal rather than that one happened;
vary ONE thing, from an ACCEPTED baseline; and treat a committed transcript as a
check subject to the mutation battery.
TWO NON-BLOCKING FINDINGS, ALSO FIXED
validate.py rules 5 and 6 NAMED THE WRONG INVERTER. `unit_caps()` SKIPS an
entry with no usable `max_w`, so caps position i stops being roster position i
the moment one entry is malformed - and rule 5 fed `caps.index(weakest)` to
`_unit_label()`, which indexes the roster, while rule 6 paired `caps[0]` with a
hard-wired `_unit_label(opts, 0)`. Right number, another inverter's name: an
operator sent to the wrong device by the message that exists to stop that. New
`unit_caps_indexed()` carries the roster index with the cap. It is latent -
rule 9(d) refuses such a config in the same pass - but that was a property of
today's rule ordering, not of the code, and validate_startup_config() is also
sim/inject.py's entry point. Part 10 gains a check that asserts the fixture
really does desynchronise the two lists, then that both messages name the right
unit, then the counterfactual: the OLD expressions, on that same config, still
evaluate to the wrong names.
DOCS.md called a downgrade "safe". True of the SCHEMA - the older schema drops
the key with a warning - and incomplete about the ADD-ON: rule 9(e) requires the
flat `setpoint_entity` be cleared when a roster is declared, so a config that
downgrades cleanly is by construction one whose flat actuator is blank, and the
older version has nothing to write to. Confirmed by running release/1.0's own
validate_startup_config() on that blob: one refusal, naming setpoint_entity.
Recoverable and loud, but an outage until a field is refilled. The note says so
now, and tells the operator to write the entity down before downgrading.
19/19 green in the goodwe-addon-ci image.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU
|
||
|
adminandClaude Opus 5
|
f000547518 |
SAFETY-30: grade each inverter against its own power, and the site's total against the meter
checks / unit (pull_request) Successful in 4m38s
THE DEFECT, in one line. `inv.sign_wd.evaluate(stood, self.grid, ...)` compared
the command THIS controller put on ONE actuator against what the WHOLE SITE is
exchanging with the network. At a roster of one those are two views of the same
thing - specs.md §6.2's identity - and that is what makes the check sound and
what earned it the best record in the safety table (NOTES.md: 93 % agreement
while healthy against 4 % during runaway, four observed episodes, zero false
positives). At two inverters they are not, and it fails in BOTH directions:
FALSE TRIP inverter A commanded -1500 W while the meter reads +1500 W is
evidence of nothing when inverter B is discharging 3000 W exactly
as told. Nothing is faulty and the highest-value check fires.
MISSED the converse, and it is worse. A unit commanded +1500 W and
MEASURED at -1500 W, beside a sibling at +1500 W, leaves P1
exactly where a healthy site leaves it. The summation destroys the
evidence and no threshold recovers it.
DEV-02 had already shipped the per-INSTANCE half - one SignWatchdog per
inverter, namespaced latch, a trip on B no longer latching A. What was left was
the COMPARISON, and it is two questions rather than one:
PER UNIT this unit's standing command against ITS OWN measured power
(`Inverter.batt`, holding register 0x0518, per-instance since
DEV-02). Conclusion: "the actuator I am driving is not doing what I
told it" - a fact about one device, so the remedy stays the per-unit
`failsafe` SET that arbiter.CLAIM_SCOPE has classified SCOPE_UNIT
from the day it was written. That entry read "the evidence is
site-wide and the remedy is not"; the first half is now simply true.
SITE the SUM of the standing commands against the meter, as a new
`sign_site` SET at P_SAFETY on the site actuator, latched under flat
`site_sign_tripped` / `site_sign_reason`. This is the 2026-08-20
fault day and NO per-unit check can see it, because every inverter
may be faithfully executing its share of a number that is itself
wrong. It stops every inverter and the banner names the site.
WHAT IT IS NOT, said here because two guards that overlap without saying so is
how a false-alarm generator gets built. It is not SAFETY-31: circulation.py
compares the units AGAINST EACH OTHER and catches them pushing power THROUGH
each other, which nets to zero at the meter and about which no single command is
evidence. This compares each unit against ITS OWN COMMAND. They overlap in
exactly one scenario - a unit reversing while its sibling pulls the other way -
and test_watchdog.py asserts on that fixture that both fire and what each
concludes. It is also not the saturation freeze, which is a MAGNITUDE test
(|command - measured| > 500 W) that shapes the loop and fires on a merely
saturated unit; this is a DIRECTION test that latches and parks one.
THE FOUR NUMBERS, RE-DERIVED AT TWO INVERTERS - one carried with a justification
and one refused. 1890 closed-loop runs on admin/goodwe's TEST-19 two-master rig
in which NOTHING is faulty, swept over the whole range config.yaml accepts for
`gain` and `slew_w`, across load reversals with write staggers and flapping
loads, and RUN AT BOTH ROSTERS so the N=1 arm is the control:
unit_sign_disagree_w 250 `saturation_w / 2`. The check gates on both sides,
opposed, so the smallest gap it can see is 500 W -
exactly where compute()'s own anti-windup test
already says the inverter is not following. It
cannot fire on a divergence the control law calls
ordinary tracking. Worst legitimate opposition at
250 W: 10.0 s; the curve is nearly flat, because a
unit's own power chases its own command.
unit_sign_disagree_s 85 8.5x that 10.0 s, 2.8x the firmware's own 30 s
link veto (inverter.LINK_STALE_S), under half the
180 s site hold.
site_sign_disagree_w 300 CARRIED. Its argument is the meter's noise floor
and the 15 W deadband; the roster touches neither,
and the campaign agrees - the gate opens in 994 of
1050 runs at two inverters and 955 of 1050 at one.
site_sign_disagree_s 325 NOT CARRIED, and this is the number the card
exists to not assume. The worst legitimate
crossing at 300 W is 110 s with one inverter and
215 s with TWO, because the command must travel a
rail that is the roster sum at a `slew_w` that is
not - reachable at gain 0.05, slew_w 50 and a
2.2 kW load reversal, all schema-legal. 180 s
would have shipped a false trip. 325 s is 1.51x
the measured worst, inside the 900 s billing
quarter and inside every episode of the fault day.
AT ONE INVERTER NOTHING CHANGES, and "nothing" is asserted rather than assumed.
The two questions are the same sentence there, so only the detector that shipped
is armed: same comparison against the meter, same 300 W / 180 s, same reason
string byte for byte, same namespaced latch and flat rollback mirror, same
`failsafe` claim, same banner, same single status row - and the site detector is
NEVER CALLED, counted on the object over twelve cycles.
VERIFIED BY BREAKING IT. Thirteen mutations on fresh per-mutation trees with no
__pycache__, mounted :ro, with the battery refusing to report at all unless the
baseline is green. Twelve red on the first pass. M4 - the site half armed at N=1
as well - SURVIVED, because the check read `disagree_since is None` and the unit
detector trips first, parks the inverter and leaves the site standing at 0 W:
the assertion ran, the sentence was true, and the data could not reach the
failure it named. Replaced with a call counter on the object; M4 is red now.
Recorded as a survivor in DECISIONS/SAFETY-30-per-unit-sign.md §5, with the
first battery's own RED BASELINE - a shared mutation directory read mid-copy -
discarded rather than reported.
RECORDED, NOT FIXED: the site comparison is defeated by a square-wave load whose
half period is about one control cycle, which holds it opposed for ~1000 s. The
N=1 control arm is what makes that interpretable - it is identical at the roster
this add-on ships, so it is a property of comparing a lagged command against a
meter and has nothing to do with the second inverter. It wants an alternation
test rather than a number, and a card of its own.
SAFETY-24's lo-only asymmetry proof re-run unchanged: 1296 combinations, 0
negative floors, `hi` +inf on every claim, every claim on the SITE actuator.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU
|