33 Commits
Author SHA1 Message Date
noahandClaude Opus 5.5 9b440a46e3 Show hashboards and SHA-256 for VNish models asic-rs does not know
VNish reports the 88-chip S19 as "Antminer S19 (88)", which asic-rs parses
to an Unknown model with no hardware table. The VNish backends sized the
board list from that table, so all three live chains were dropped and the
Health column drew no hashboard icons for hashing miners; the same model also
got an Unknown hash algorithm. Size boards from the chains the miner reports
when the table is missing, and infer the algorithm from the Antminer family
letter (helper shared with the stock backend). asic-rs patch 0034.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-03 09:32:28 -06:00
noahandClaude Opus 4.8 613132035a feat(gui,asic-rs): hashboard yellow-not-red, power-limit column, chip/HB/PSU-serial coverage
Round 21 (three reported items).

1. A hashboard that is still mining near full hashrate but has some failed chips
   now shows yellow (degraded), not red. VNish emits per-board chip failures as
   Warning while the board is mining (Error only when the board is not mining),
   and board_healths() renders a warning-only board as Degraded instead of Down.
   Verified live on 10.4.0.193 (mining ~83%, dead chips -> yellow).

2. New "Power limit" column: the configured power-target setpoint in watts
   (MinerRow.power_limit_w, from a Power tuning target). WhatsMiner shows its
   "Power Limit" (e.g. 2500 W); the value was already scanned but had no dedicated
   column. Numeric (sortable/filterable), Performance category, opt-in.

3. Chip + hashboard model coverage:
   - VNish hashboard model scanned from factory-info board_model (A3HB70601,
     BHB42603, ...) -- 147-unit fleet gains HB model. Both VNish backends.
   - WhatsMiner v1/v2 PSU serial scanned from get_psu serial_no (was hardcoded
     None). 48-unit fleet gains PSU serial.
   - Chip type derived from the detected model for Antminer hardware (VNish/stock/
     LuxOS/Braiins/UMC never report a chip string on the wire), in naming.rs:
     S21Pro->BM1370, S19jPro->BM1362, S19KPro->BM1366, S19/T19->BM1398, etc.
     Only high-confidence mappings; WhatsMiner and unknown models stay blank.
   PSU Vmin/Vmax remain Braiins/ePIC-only (not on the wire for WhatsMiner/VNish/
   Antminer -- confirmed live).

asic-rs patch 0033. 95 unit tests pass.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-10-02 10:09:07 -06:00
noahandClaude Opus 4.8 a4d0cb65f0 fix(gui,vnish): hashrate 0-vs-dash, paused health colours, VNish per-hashboard faults
- Hashrate cell: a readable 0 hashrate now shows "0.00T" instead of a dash. The
  dash is reserved for an unreadable/absent hashrate. New MinerRow.hashrate_known
  records whether the miner reported a value; combined_si/num_value key off it,
  and scan/fixtures set it from data.hashrate.is_some().
- Paused-miner Health icons: the control board (brain) and PSU keep their true
  health (green when fine) instead of turning blue; fans turn blue only when
  actually stopped (0 RPM) and stay green while spinning; hashboards stay blue
  while paused (idle is the expected state).
- VNish per-hashboard faults: DataField::Messages now also reads the summary
  chains (chip_statuses red/orange + per-chain state) and emits one message per
  faulted board, attributed to MinerComponent::HashBoard{idx}, so the Health
  column reds the specific board and its tooltip carries the fault instead of it
  landing on the brain. Applied to v1_2_0 and v1_3_0. patch 0032.

Verified live: VNish 10.7.1.25 -> "Hashboard 2: 1 degraded chip(s)" on board idx
1; healthy 10.7.1.122 -> no messages. 93 unit tests pass.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-24 15:14:20 -06:00
noahandClaude Opus 4.8 a4fcb6e763 asic-rs telemetry: VNish fan%/temps, WhatsMiner chip type, and auto-tuner status
Round 19 #10 telemetry extension, live-verified. Two logical changes (patches 0030
and 0031) shipped together; they populate existing GUI columns plus one new "Tuner"
column.

VNish + WhatsMiner (patch 0030, no core/GUI change):
- VNish v1_2_0 + v1_3_0: fan duty % (summary /miner/cooling/fan_duty) and board temp
  thresholds (settings restart_temp -> dangerous, auto cooling param -> target).
  Live-verified 10.4.1.50 (81 / 90 / 65). No PSU serial exposed -> stays None.
- WhatsMiner chip type (per user, do not drop): v1/v2 summary "Chip Data" merged
  into hashboards -> BoardData.chip_type (first token); v3 chipdataN -> chip_type and
  hash-board ("K40") -> hashboard_model.

Auto-tuner status (patch 0031):
- New DataField::TunerStatus + MinerData.tuner_status (Option<String>), riding the
  existing GetOperatingState trait via a defaulted parse_tuner_status.
- Braiins: performance/tuner-state overall_tuner_state via the BOS+ TunerState enum
  (Disabled/Stable/Tuning/Error/Continuous/Preheat) -- mapping from the bos-plus-api
  proto. ePIC: PerpetualTune -> Off/Optimized/Tuning. Others -> None.
- GUI: new "Tuner" column (Health & status category, auto-hidden when empty).

All values live-verified against the corpus miners (ePIC 10.4.1.6, LuxOS 10.4.1.60,
Braiins 10.4.1.83, VNish 10.4.1.50). Fixture tests assert every parsed value.
ALL_COLUMNS 71 -> 72.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-23 18:10:57 -06:00
noahandClaude Opus 4.8 4aec6d40ad asic-rs telemetry: board temp thresholds -> target/hot/dangerous (LuxOS, Braiins; remap ePIC)
Round 19 #10 telemetry extension. Live-verified against the corpus miners, this
remodels the board temperature thresholds to item #10's target/hot/dangerous scheme
and adds the LuxOS and Braiins backends. Supersedes the ePIC-only critical/shutdown
model shipped in v0.72.0.

Core (asic-rs-core):
- MinerData: temp_critical/temp_shutdown -> temp_target/temp_hot/temp_dangerous.
- parse_temperature_limits (defaulted on the existing GetFluidTemperature trait)
  now returns (target, hot, dangerous); parse_data wires all three fields.

Backends (each reads an already-fetched or single command; live-verified):
- LuxOS: new tempctrl RPC; ChipTarget/ChipHot/ChipDangerous (live 10.4.1.60 = 83/93/100).
- Braiins v25_07 + v26_04: configuration/constraints cooling_constraints defaults
  (live 10.4.1.83 = 60/80/90); WEB_CONSTRAINTS already fetched for tuning caps.
- ePIC remap: Shutdown Temp -> hot, Critical Temp -> dangerous, no target.
- VNish (settings restart_temp/param) handled in a follow-up (needs settings auth);
  Antminer/WhatsMiner expose no thresholds -> (None,None,None).

GUI: the two temp columns (Critical/Shutdown) become three (Target/Hot/Dangerous),
right-aligned, Temperature category, auto-hidden when empty. MinerRow gains
temp_target_c/temp_hot_c/temp_dangerous_c; to_row reads them. ALL_COLUMNS 70 -> 71.

Fixtures captured from live: luxminer tempctrl.json, braiins configuration_constraints.json.
Fixture tests for epic, luxminer, braiins v26_04 assert the parsed thresholds. Patch 0029.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-23 17:40:33 -06:00
noahandClaude Opus 4.8 45d9632590 asic-rs telemetry T3-temps: board temperature thresholds (critical / shutdown)
Round 19 #10 telemetry extension. Adds the board temperature safety thresholds
("board dangerous temp" from item #10).

Core (asic-rs-core):
- New DataField::TemperatureLimits; MinerData gains temp_critical and temp_shutdown
  (Celsius).
- Rides the existing GetFluidTemperature trait via a defaulted parse_temperature_limits
  method returning (critical, shutdown) -- no supertrait blast radius; parse_data
  wires both fields in.

Backend:
- ePIC: summary /Misc Critical Temp + Shutdown Temp (already-fetched summary).
- Antminer/WhatsMiner expose no thresholds; LuxOS needs an unwired tempctrl command
  (no fixture) and VNish/Braiins need live verification -> left for a live-verified
  pass, all inherit (None, None).

GUI: two new columns (Critical temp / Shutdown temp), right-aligned, Temperature
category, auto-hidden when empty. MinerRow gains temp_critical_c / temp_shutdown_c;
to_row reads them. ALL_COLUMNS 68 -> 70.

Fixture test for epic asserts the parsed thresholds. Patch 0028.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-23 16:53:57 -06:00
noahandClaude Opus 4.8 fadd851e04 asic-rs telemetry T3-fan: miner-level fan duty cycle (fan speed %)
Round 19 #10 / #8 telemetry extension. Adds the fan duty cycle as a miner-level
percentage -- the normalized fan-load indicator the GUI wanted alongside per-fan
RPMs. No firmware exposes a maximum fan RPM, so a true actual/maximum pair is not
possible; the duty cycle is the closest normalized signal.

Core (asic-rs-core):
- New DataField::FanSpeedPercent; MinerData gains fan_speed_percent (Option<f64>).
- Modeled as a top-level miner-level value (most firmwares report a single duty),
  so FanData is untouched -- no Eq removal / literal churn across its 17 crates.
- Rides the existing GetFans trait via a defaulted parse_fan_speed_percent method
  (no supertrait blast radius); parse_data wires it in.

Backends (read already-fetched commands, no extra round-trips):
- ePIC: summary /Fans/Fans Speed (single %).
- LuxOS: max of FANS[].Speed.
- Braiins v25_07 + v26_04: max of cooling fans[].target_speed_ratio x 100.
- Antminer (pwm via config) and VNish (fan_duty) have no data fixtures -> left for a
  live-verified pass; WhatsMiner is firmware-locked -> None.

GUI: new "Fan %" column (right-aligned, Cooling category, auto-hidden when empty).
MinerRow gains fan_speed_percent; to_row reads it. ALL_COLUMNS 67 -> 68.

Fixture tests: epic, luxminer, braiins v26_04 assert the parsed percentage. Patch 0027.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-23 16:28:14 -06:00
noahandClaude Opus 4.8 3f9fc4a24b asic-rs telemetry T2b: cumulative share stats (difficulty acc/rej, stale, blocks)
Round 19 #10 telemetry extension, tier 2b. Adds the cumulative share statistics
from item #10 end to end: difficulty-weighted accepted/rejected work, raw stale
share count, and blocks found.

Core (asic-rs-core):
- New ShareSummary sub-struct (data/share.rs) with a parse_cgminer_share_summary
  helper reading a cgminer/btminer SUMMARY[0] object.
- New DataField::ShareSummary; MinerData gains share_summary.
- Rides the existing GetBestShare trait via a defaulted parse_share_summary method
  (no new supertrait bound, no per-backend empty impl); parse_data wires it in.

Backends (read the already-fetched SUMMARY, so no extra round-trips; these are
miner-level aggregates -> top-level MinerData fields, not per-pool PoolData):
- Antminer v2020 + v2023_07, LuxOS, WhatsMiner V1 + V2 override parse_share_summary.
- Braiins (no cumulative difficulty), ePIC (current difficulty only), WhatsMiner V3
  and VNish expose none and inherit None -- fixtures confirm.

GUI: four new columns (Diff acc / Diff rej / Stale / Blocks), right-aligned, in the
Pools category, auto-hidden when empty. MinerRow gains the four fields; to_row reads
them from share_summary. ALL_COLUMNS 63 -> 67.

Fixture tests: whatsminer v1 and antminer v2020 assert the parsed values. Patch 0026.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-23 15:51:57 -06:00
noahandClaude Opus 4.8 a60bb34e64 asic-rs telemetry T2a: rolling average hashrate windows (1m/5m/15m/24h)
Round 19 #10 telemetry extension, tier 2. Adds the rolling average hashrate
windows the GUI asked for (1M/5M/15M/24H) end to end.

Core (asic-rs-core):
- New HashrateAverages sub-struct (data/hashrate.rs): avg_1m/5m/15m/24h, each an
  Option<HashRate> carrying the same unit and algorithm as the current hashrate.
- New DataField::HashrateAverages; MinerData gains hashrate_averages.
- The windows ride the EXISTING GetHashrate trait via a new defaulted
  parse_hashrate_averages method (default None), so no new supertrait bound and no
  empty impl in every backend -- only the backends that expose the data override
  it. parse_data wires it in.

Backends (all read a command the collector already fetches -> no extra round-trips;
each window is filtered to > 0 and the whole struct is dropped when every window is
absent, so a firmware that reports nothing keeps hashrate_averages = None):
- WhatsMiner V1: btminer summary MHS 1m/5m/15m (MegaHash); no 24h.
- WhatsMiner V3: get.miner.status summary hash-1min / hash-15min (TeraHash) only.
- LuxOS: summary GHS 1m/5m/15m/24h (GigaHash) -- full window set.
- Braiins v25_07 + v26_04: miner_stats real_hashrate last_1m/5m/15m/24h (GigaHash).
- Antminer (5s/30m/av only), WhatsMiner V2 (av only), ePIC/VNish (session average
  only), sealminer: no standard 1/5/15 window -> inherit None. This is honest: the
  fixtures confirm those firmwares do not expose the standard windows.

GUI: four new columns (HR 1m/5m/15m/24h), single TH/s value each, right-aligned,
blank when the window is not reported, auto-hidden when empty across the fleet.
MinerRow gains the four ths fields; to_row populates them from hashrate_averages.

Fixture tests: whatsminer v1 and braiins v26_04 gain assertions on the parsed
windows (value, unit, and the expected present/absent pattern). Patch 0025.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-23 15:08:27 -06:00
noahandClaude Opus 4.8 a6e22581d7 asic-rs telemetry T1b: WhatsMiner V3 PSU serial + ePIC chip/HB type
Round 19 #10 telemetry extension, tier 1 sub-batch B. Both ride existing
DataFields (Psu/Hashboards) on already-fetched commands (no extra round-trips)
and reuse the PsuInfo.serial / BoardData.chip_type+hashboard_model fields added
in T1a, so no core or GUI change -- the existing columns just populate for more
firmwares. Patch 0024; fixture tests pass.

- WhatsMiner V3: add a DataField::Psu location (get.device.info /msg/power) and a
  parse_psu that reads model/type, sw/hwversion, and the serial (sn). V3 had no
  PSU parse at all (default None) despite the data being fetched for PsuFans.
- ePIC: parse_hashboards now reads capabilities "Chip Type" (e.g. "BM1366") and
  "Model Subtype" (e.g. "BHB56801"), applied to every board (capabilities reports
  them miner-wide). The capabilities command is already fetched for Hashboards.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-23 14:31:36 -06:00
noahandClaude Opus 4.8 f887affd6d asic-rs telemetry T1a: PSU serial + chip/HB type + best-share
Extend vendored asic-rs to scan more per-firmware telemetry (round 19 #10
follow-up, tier 1 sub-batch A). All new values ride on existing DataFields
(Psu/Hashboards/BestShare) and commands already fetched, so no new core
DataField plumbing and no extra network round-trips. Patch 0023.

Vendored (asic-rs):
- PsuInfo gains `serial`. Braiins v26_04 parse_psu corrected to read model_name
  + fw_version (it used the numeric type code as the model and dropped the fw
  version) and to capture serial_number. ePIC / WhatsMiner v1/v2 report no serial.
- BoardData gains `chip_type` (e.g. "BM1370") + `hashboard_model` (e.g.
  "A3HB70601"); Braiins miner/hw/hashboards populates both.
- best_share (all-time) wired for Antminer v2020 and WhatsMiner v1/v2 from the
  already-fetched summary "Best Share" (those backends had the empty default).
- refresh-checksums --all; existing braiins/antminer/whatsminer fixture tests pass.

GUI:
- MinerRow gains chip_type, hashboard_model, psu_serial; populated in to_row.
- New columns: Chip, HB model, PSU serial. Best share now populates for Antminer
  and WhatsMiner (column already existed).

Feasibility analysis (5-agent, all firmwares) recorded in memory; further tiers
(fan %, rolling hashrate averages, difficulty/shares, temp thresholds, WhatsMiner
V3 / VNish PSU serial) follow. Confirmed genuinely-unavailable: fan max RPM,
WhatsMiner/LuxOS real logs, chip type on Antminer/LuxOS.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-23 14:01:33 -06:00
noahandClaude Opus 4.8 95b8428c12 Round C G3: named firmware presets (LuxOS) end to end
The flagship structured-tuning parity item: list and select a firmware's named
profiles/presets, not just map a watts number onto one.

Backend (vendored asic-rs, patch 0022): LuxOS now implements SupportsPresets
(get_presets reads the live `profiles` RPC) and a real SupportsTuningConfig
(set_tuning_config routes Preset(name) -> profileset, Power -> set_power_limit).
Unit-tested (parse_profiles_to_presets); verified live-read on two production
LuxOS S19j Pro (10.4.1.60/.67). VNish already exposed presets.

GUI: MinerCaps.supports_presets + MinerRow.presets (read live at scan time for
preset-capable miners; ~1 RPC per such miner per 300 s rescan). New "Set
profile..." right-click item opens a selector populated from the profiles
common to the whole selection (so the chosen name is valid on every target),
applied via the new ConfigAction::Preset. Offline fixtures seed demo presets.

CLI: `miner_probe presets <ip>` (list) and `setpreset <ip> <name>` (select).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-22 17:34:26 -06:00
noahandClaude Opus 4.8 e5a6cc2b5e psu: extend PSU-info scan to Braiins + WhatsMiner (R5/G15)
Round 17 R5/G15. The PSU version/Vmin/Vmax columns + tooltip now populate for
Braiins and WhatsMiner, not just ePIC.

- Braiins (patch 0020): miner/details -> psu_info gives the PSU type (117, the
  same APW12 ePIC reports) + rated voltage min/max. Verified live on 10.4.1.83/.27:
  model 117, Vmin 11.96 V, Vmax 15.2 V.
- WhatsMiner v1/v2 (patch 0021): get_psu -> model + hw/sw version (no DC output or
  rated envelope exposed). Verified live on 10.8.1.214 (v2): model P221C, version
  H20201205. v3 uses a different response shape and no v3 unit was available, so
  it's left as a follow-up.
- Also fixes a pre-existing broken test (test_whatsminer_v1_parse_messages never
  mocked get_error_code -> always returned 0 messages; CI runs only the GUI
  selftest so it was never caught). whatsminer backend tests now 41/41.

SELFTEST PASS; braiins + whatsminer backend suites green.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-22 16:07:53 -06:00
noahandClaude Opus 4.8 49985631d4 antminer: read best-share fleet-wide + Bixbit "Power Estimated" wattage (R7)
Round 17 R7. Flashing Bixbit onto fleet units surfaced two Antminer scan gaps
(patch 0019): the stock v2023_07 backend never fetched best-share at all
(DataField::BestShare had no source), so it was blank for every stock/Bixbit
Antminer; and it read wattage only from the cgminer stats, which Bixbit-based
firmware leaves at 0 (real draw is in the summary as "Power Estimated").

Now maps BestShare/SessionBestShare to the cgminer summary and adds a
"Power Estimated" wattage fallback (stock power still wins when present).

Verified live: Bixbit S19j Pro (10.4.1.70) now reads 2615 W + best-share
178930671; stock S21 Pro (10.4.1.90) now reads best-share 299464394 -- both were
blank before. 36 antminer backend tests + SELFTEST pass, no wattage regression.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-22 13:34:01 -06:00
noahandClaude Opus 4.8 8fe61ec427 psu: PSU info scan + version/Vmin/Vmax columns + tooltip V/A (R5/G15)
Round 17 R5/G15. asic-rs had no PSU telemetry. Adds an optional PsuInfo to
MinerData (model, version, rated voltage min/max, live output voltage/current),
a DataField::Psu, and a default parse_psu backends override (patch 0018). ePIC
populates it by merging the summary "Power Supply Stats" with the capabilities
"Psu Info".

GUI: three new columns -- "PSU version", "PSU Vmin", "PSU Vmax" (Hardware group in
the picker) -- and the R2 power tooltip now shows the PSU output voltage (and
amperage when the firmware reports a non-zero current; ePIC leaves current 0, so
it shows just the voltage).

Verified live on ePIC S19j Pro (10.4.1.6/.10): model 117, fw 22, Vmin 11.877 V,
Vmax 15.182 V, output ~13.2 V. 8 model tests + 18 ePIC backend tests + SELFTEST
pass. Non-ePIC firmware leave the PSU fields blank until they add parse_psu.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-22 12:24:26 -06:00
noahandClaude Opus 4.8 08b3d56864 luxminer: set power-limit profile by name so the default profile is reachable
Round 17 C, found via live tuning verification. LuxOS set_power_limit mapped
watts to a profile but applied it by the profile's Step. The factory "default"
profile has Step "0", which LuxOS profileset reads as an empty parameter
("Missing profile parameter"), so the setter could never select the default and
a miner nudged off it by a watts limit could not be returned to its factory
point. Key on the profile Name instead (patch 0017).

Verified live on a LuxOS S19j Pro (10.4.1.60): set_power_limit(2700)->480MHz/2632W,
set_power_limit(2900)->default/2794W (previously errored). Unit test updated.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-22 11:20:51 -06:00
noahandClaude Opus 4.8 da0d1582be scan: cap identification concurrency at 128 to fix Braiins value dropout (R4)
Round 17 R4. Scanning most values from Braiins OS units failed under a
full fleet scan: a /24 returned watts=0 and no best-share/temps/fw for
Braiins miners (and dropped a few responders entirely), while the same
miner scanned in isolation populated fully.

Root cause: asic-rs picks a scan concurrency of 1000 for any range up to
/22, so a /24 fires all ~254 address probes at once. Over a real link
(the fleet is reached across a WireGuard VPN, and a subnet full of dead
IPs holds every TCP connect open for the whole identification timeout)
that burst starves the per-responder get_data. A Braiins miner fans out
~12 authenticated REST calls; under the burst they drop miner/stats, so
"most values" come back blank. It is not FD exhaustion (nofile is 500k),
not the per-call timeout, and not an auth race.

Fix: cap identification concurrency at 128 in the app (GUI scan +
miner_probe). On the live fleet 128 was the highest cap that stayed fully
reliable -- 192 already dropped responders, 254 blanked a Braiins miner.
A /24 still finishes in ~35s; larger ranges just take proportionally
longer and stay correct.

Belt-and-suspenders in the braiins backend (patch 0016): single-flight
the login (one auth instead of ~12 concurrent) and widen the defensive
get_data timeout 5s -> 15s.

Verified live: 10.4.1.0/24 now populates both Braiins corpus miners
(.27 watts=2748 best=55217110, .83 watts=2749) fully and misses no
responders (65/65). SELFTEST PASS.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-22 09:59:27 -06:00
noahandClaude Opus 4.8 86338062a5 Round 17 B (part 2, GUI): timezone / change-password / factory-reset dialogs + BOS fix
Wire G10/G11/G12 into the GUI (they already had CLI verbs in part 1):
- MinerCaps gains timezone / change_password / factory_reset (probed from the
  matching supports_*() at scan time; intersect + any_control + is_empty updated).
- ConfigAction gains Timezone(String) / ChangePassword(String) / FactoryReset with
  labels + apply arms. Timezone/FactoryReset go through apply_action(&dyn Miner);
  ChangePassword takes &mut self so it runs on the owned miner in the action loop.
- ui.rs: a shared PromptKind (Power/Tuning/Timezone/Password) generalises the
  value prompt (timezone is IANA-validated before firing; password is non-empty);
  factory-reset uses the confirm dialog with a strong WIPES-config/DHCP warning.
  New gated context-menu items: Set timezone, Change password, Factory reset.

Backend fix (asic-rs patch 0015): BOS send_command treated an empty 2xx body as a
parse error, so change_password reported Ok(false) on success. Empty 2xx is now
Null (success). Found + fixed while live-verifying G11.

Verified live on Braiins 10.4.1.27: full change-password round-trip (root ->
bftest2 -> read with new pw -> restore root), all Ok(true), unit left on root/root.
Timezone shares the CLI-verified set_timezone_config path. Factory-reset is wired +
gated but intentionally NOT live-fired: its behaviour is unrecoverable per firmware
(can revert LuxOS to stock / reset network to DHCP and lose the unit) -- needs a
truly sacrificial console-attached unit. GUI selftest + 89 unit tests pass.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-22 08:58:00 -06:00
noahandClaude Opus 4.8 25c8ccf1ff Round 17 A (part 2, G13): LuxOS pool ops privileged + removepool
The luxminer pool write commands (addgroup/groupquota/removegroup/addpool/
switchpool/enablepool/disablepool) sent privileged=false -> no session_id, which
fails with "Missing session_id" on password-set LuxOS units (all other LuxOS
writes are already privileged). Flipped all seven to privileged=true and added the
missing removepool. asic-rs patch 0014.

Verified no regression on default corpus unit 10.4.1.60 (set_pools_config still
applies). The password-set failure needs a password-set LuxOS unit to verify live.

Round A remaining: G8 (constraints/limits reads) and G14 (telemetry enrichment) are
deferred to the rounds that consume them (G8 -> the tuning/cooling dialogs in C/D;
G14 -> detail-panel/tooltip work) rather than built as unused plumbing now.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-21 22:47:18 -06:00
noahandClaude Opus 4.8 307539543d Round 17 A (part 1): committed write CLI (G1) + best-share reads (G9) + parity roadmap
Executing the aftermarket config+scan parity plan (docs/aftermarket-parity-roadmap.md,
from the ultracode writer/adversary workflow). This is Round A's foundation + first cheap win.

G1 - miner_probe is now a committed config+scan CLI, not a local-only diagnostic. Added
capability-gated write verbs: `light <ip> <on|off>`, `pause`, `resume`, `restart` (alongside
the existing setpl/setpools/led), a READ vs WRITE usage banner warning that write verbs mutate
(use the test-bed miners, not production), and best_share/session_best_share columns in `scan`.
Verified: `light on`/`light off` on a LuxOS corpus unit -> Ok(true) (reversible).

G9 - implement GetBestShare/GetSessionBestShare (were empty impls feeding two dead columns):
Braiins v26_04 reads miner/stats -> /miner_stats/best_share; LuxMiner v1 reads summary ->
/SUMMARY/0/Best Share + Best Session Share. Both reuse already-collected responses (no extra
calls). asic-rs patch 0013. Verified live read-only: BOS 10.4.1.83 = 130601978 (no session on
BOS); LuxOS 10.4.1.60 = 638881749/638881749, 10.4.1.67 = 497802665/497802665 -- exact API match.
(WhatsMiner stock summary exposes no best-share on the tested units; deferred. The roadmap's BOS
source "miner_stats" was a typo for the real "miner/stats" endpoint.)

Also commits the parity roadmap (docs/) as the plan of record for rounds A-F.
Round A remainder (G8 constraints reads, G13 LuxOS pool-op privileged fix, G14 telemetry
enrichment) follows; G13 needs a password-set LuxOS unit to fully verify.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-21 22:36:59 -06:00
noahandClaude Opus 4.8 74b97162f5 GUI round 16 (#1/#2/#3): per-component health attribution + tooltips
Health icons and their hover tooltips now reflect the miner's actual per-component
faults instead of a generic "faulted".

- #1 Each Health icon's tooltip carries the specific fault text for THAT component
  (e.g. "Slot 0 error reading chip id.", "Power current error."), sourced from
  asic-rs's structured MinerMessage.component (HashBoard idx / PowerSupply / Fan /
  ControlBoard) rather than fragile string matching on the summary. New MinerRow
  `faults` field + FaultComponent/Fault types; scan.rs maps messages -> faults.
- #2 A power-supply fault turns the bolt icon red, lists the PSU faults on its
  tooltip, and greys (not reds) every hashboard -- the boards can't be powered or
  tested, so they read "unpowered (PSU fault)".
- #3 When the miner reports board-specific faults, ONLY the named boards are red;
  the rest stay green. A whole-miner stall zeroes every board's hashrate, which
  previously reddened working boards the error codes actually exonerate.

Backend enabler (asic-rs patch 0012): the WhatsMiner V1 backend now sources faults
from get_error_code (like V2/V3; its summary error fields are empty on H6OS), and
a targeted repair fixes firmware that emits an invalid bare-pair error_code array.

Verified live: 10.7.1.231 (PSU) greys its boards + shows the 4 PSU faults on the
bolt; 10.8.1.234 (540/541/561) reds boards 1+2 and leaves board 3 green. Unit
tests cover the attribution and the malformed-array repair; all suites pass.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-21 17:30:18 -06:00
noahandClaude Opus 4.8 75663f4c23 asic-rs: implement LuxOS set_power_limit (via profiles); document ePIC PSU gating
Follow-up to the Round 15 corpus. The toolkit reported supports_power_limit=false
for LuxOS and ePIC; investigated both against the live corpus (S19j Pro).

LuxOS was genuinely under-reporting: it tunes by discrete named profiles that each
carry a Watts rating (the `profiles` RPC gives Step + Watts), applied via
profileset -- pure frequency/voltage tuning, no PSU dependency. Implemented
SetPowerLimit to map the requested watts to the highest profile whose Watts does
not exceed it (or the lowest profile when the target is below all of them). The
pure mapping (parse_profile_watts + select_profile_at_or_below) has a unit test.
Live-verified: set_power_limit(2500) on an S19j Pro picked step -4 (2385 W) and
the miner's active profile became 445MHz/-4.

ePIC is left at false, now with a comment explaining why: PowerPlay's only true
power-limit path is the PowerTune perpetual-tune algorithm, which the firmware
gates on the PSU -- on a fixed-voltage Bitmain APW12 (S19j Pro) it answers "Power
optimization is not supported by this PSU". That is a per-PSU runtime property the
synchronous capability check cannot see, so reporting true would over-advertise on
most reflashed Antminers. Hashrate tuning remains available via set_tuning_config.

Captured as asic-rs patch 0011. luxminer + epic crate tests pass.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-21 16:40:38 -06:00
noahandClaude Opus 4.8 9eb6a0f229 asic-rs: fix Antminer locate-LED (and reboot) on legacy web-UI firmware
Round 15 #7. The "flash the light" action failed on old (< 2023.07) Antminer
firmware -- the LuCI web-UI generation, e.g. the Z15/Z15J. set_fault_light
returned Ok(false) on live hardware while appearing to "succeed".

Root cause (confirmed on a live Z15): that firmware's blink.cgi ignores the JSON
body {"blink":"true"} the backend sent (answering 200 with an empty text/html
body, so the JSON parse fails and the error is swallowed into Ok(false)). Its own
blink.html POSTs a form field action=startBlink/stopBlink and reads state back via
action=onPageLoaded -> {"isBlinking":bool}; the action handler also often answers
401/500 even on success, so HTTP status is not a reliable signal.

Fix (v2020 backend): add a form-encoded POST helper; blink() now fires
action=startBlink/stopBlink, ignores the unreliable status, and reads the real
state back via onPageLoaded; set_fault_light reports success only when the observed
isBlinking matches the request. The LightFlashing read is routed to the same path
(get_blink_status.cgi is 404 here) and remapped to {"blink":bool}.

Also switch reboot() from POST to GET to match the same firmware's reboot.html (the
Restart action was affected); not live-fired, to avoid rebooting production miners.

Captured as asic-rs patch 0010. Also teach ci/compute-version.sh that vendor/* and
thirdparty/asic-rs/* are backend changes, so vendored fixes trigger a release.

Verified live: Z15/Z15J now return Ok(true) with the light genuinely toggling;
VNish (4 control boards) and WhatsMiner (H6OS/H616) unchanged; antminer tests pass.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-21 15:19:05 -06:00
noahandClaude Opus 4.8 0d19e958f0 GUI round 15 (part 1): pool dropdowns, aligned health icons, export, more
Twelve of the round-15 items (the code items; the live-hardware flashing/
testing and the settings/localization/logs work follow separately):

#1 Configure-pools: three pool rows (address, account, worker, password),
each field with a "+" dropdown of presets (top-11 BTC pool stratums;
worker-name placeholders) plus user-addable, persisted entries. Worker
picks insert at the cursor rather than replacing. Writes up to three pools
per miner via set_pools_config, worker expanded per unit.

#2 PIC detection: replace the broken SSH i2cdump (it read 0x48, the temp
sensor, not the PIC) with a model lookup -- no firmware exposes PIC over the
network, so infer it from the model (noPIC: S19k Pro, S21 family, T21,
explicit noPIC; PIC: rest of the S19 line).

#4 Zoom: an editable percent field in the status bar (type an exact scale)
and Ctrl +/- / Ctrl+0 to zoom in/out/reset.

#5 Health icons align in fixed slots (power, control, fans, hashboards);
miners with fewer fans/boards leave a gap so hashboards still line up. The
max fan/board counts are computed parametrically from the fleet.

#6 Light: the locate-light column is on by default (last), shown as a
green-over-red beacon flashing quickly instead of a single amber dot.

#8 The bottom-left status line is now scan-only; operation progress lives in
the Health column (per miner) and failures in the bottom-right indicator.

#9 / #10 / #11: drop the Credentials-dialog blurb; move the scan-ranges
format help to a tooltip; make Restore-to-stock a plain confirmation (no
typed keyword, no warning wall).

#12 Z15 fans: the v2020 backend now reads fans from higher fan slots
(the Z15 reports fan3/fan4 with fan1/fan2 = 0). Vendored patch 0009.

#14 Export: an Export button writes every scanned data point to a CSV, one
value per cell (lists expanded to indexed columns) -- complete and
machine-readable, unlike the on-screen table.

#15 Health tooltips are per-icon: hovering a specific fan or hashboard shows
just that component's state, with errors attributed to the board that names
them in the message.

New wx-rs bridge surface: widget tooltips, save-file dialog, insert-at-cursor,
status-bar zoom entry + Ctrl+/-, the light beacon + alignment-gap sentinels,
and per-icon cell tooltips (bridge patches 0002/0003 regenerated). Unit tests
(worker templates, PIC map, CSV export, health) + headless selftest pass.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-21 14:43:00 -06:00
noahandClaude Opus 4.8 ecdb2fab9d GUI round 13: health-icon column, Z15J algorithm, Credentials + Columns dialogs
Five items from the round-13 list:

#4/#5 Z15 hashrate / unknown algorithm: a Z15J reports its model as
"ANTMINER Z15J", which asic-rs' alias table did not recognise, so it
deserialised to Unknown and its Sol/s hashrate was left un-bucketed as
"unknown hashrate". Add a first-class Z15J model (Equihash) in the makes
crate, plus a family-letter fallback in the v2020 backend so other
un-tabled Antminer models still resolve an algorithm (Z->Equihash,
L->Scrypt, S/T->SHA-256; ambiguous K/DR stay Unknown). Live-verified on
the Z15/Z15J fleet. Carried as vendored patch 0008.

#3 Component-health column: a new Health column (power, control board,
each fan front-to-back, each hashboard) drawn as coloured icons -- grey
unreachable, red down, yellow sub-optimal, green healthy -- replacing the
Errors, Boards and Status columns. Left-clicking the header cycles the
sort through the component facets (overall/power/fans/hashboards); the
active facet shows as a tag beside the chevron. Errors move to the cell
tooltip. Four operator-supplied Font Awesome icons registered via the
existing SVG-in-cell renderer (new SHAPE_BOLT/BRAIN/FAN/CHIP sentinels).

#2 Credentials: rename the "Set scan credentials" button to "Credentials"
and make the per-firmware defaults the main dialog (the global "all
firmwares" login is folded in as the top row).

#1 Columns dialog: narrower; new Alphabetical and Categories sort buttons
(via a Column::category grouping and a wxbr_reorder_set_order bridge fn);
one mouse handler that does both drag-to-reorder and drag-to-rubber-band-
select (the old BEGIN_DRAG binding had silently disabled rubber-band
selection); Ctrl+A select-all and Ctrl/Shift+Up/Down extend-selection.

wx-rs bridge patches 0002/0003 regenerated. Unit tests + headless
selftest pass.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Djv3oncM4LxTS5tb61Ako6
2026-09-21 11:06:05 -06:00
noahandClaude Opus 4.8 f1fb2f8623 GUI round 12 (#8b, part 2): WhatsMiner control commands on the older V1 backend
asic-rs already implements the full WhatsMiner proprietary control+monitoring
surface for V2/V3 (adjust_power_limit, set_led, update_pools, reboot, power_off/
power_on + all reads), so the modern fleet was already covered. The gap was the
V1 backend (firmware < 2022-07-29): it declared no control commands, so older
units could only be monitored and the GUI's capability actions never appeared.

V1-era firmware speaks the same classic btminer token write API as V2 (the round
12 #8b part 1 access-probe handshake proved V1 accepts get_token + AES), so the
V1 backend now reuses V2's proven RPC client for writes and implements the same
control commands: set_fault_light, set_power_limit, update_pools, restart, pause,
resume (HasAuth routes the admin password to the write RPC).

Verified live on a V1 unit that the token AES key (password) is accepted; the
write path is byte-identical to V2's live-validated one. Some V1 units keep their
API in read-only mode and answer writes with Code 45 "can't access write cmd"
until an operator enables write access via the WhatsMiner Tool -- a per-device
setting; the backend now fails those gracefully instead of hiding the command.
40 whatsminer crate tests pass; GUI self-test passes. Patch queue adds 0007.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Y8h447BPyJR4fioYZvQ9bh
2026-09-20 11:13:01 -06:00
noahandClaude Opus 4.8 c8d5cd71c4 GUI round 12 (#10): Z15 realtime hashrate + control board, from data we can read
Stock Antminer Z15/Z15J were mis-handled: their control board showed blank and
their realtime hashrate looked absent next to the target. Root cause: the version
detector only read summary.cgi, which the Z15 does not serve (404), so version
detection returned nothing and AntMiner::new fell through to the newest backend,
which mis-parsed the miner. Fixes (all in the vendored asic-rs antminer crate,
verified live on 10.5.1.x):

- Backend routing: fall back to the cgminer :4028 `version` RPC's CompileTime when
  summary.cgi is absent, so a 2020 Z15 routes to the v2020 backend instead of the
  default newest one.
- Control board "read, not guess": the Z15 has no miner_type.cgi, so read the
  control board from the kernel build string in get_system_info.cgi
  ("Linux 4.6.0-xilinx-..." -> Xilinx). The ControlBoardVersion field now carries
  both the subtype and the kernel; the parser prefers the explicit subtype and
  falls back to the kernel marker (xilinx/zynq -> Xilinx, aml/meson -> AMLogic).
- Malformed cgminer stats: the Z15 emits a stats array with no commas between its
  objects (invalid JSON serde rejected, which dropped every stats-derived field).
  Insert the missing separators before parsing (a no-op on valid responses),
  restoring expected hashrate, fans, hashboards and uptime. Expected hashrate now
  gets the same Equihash Sol/s reinterpretation as the realtime rate, so both read
  on one scale (e.g. 144k / 320k) instead of expected reading ~1e6x larger.

Live result: Z15/Z15J now report control board "Xilinx" and realtime/expected
hashrate on the same scale. asic-rs antminer crate tests (35) pass; GUI self-test
passes. Vendored source + refreshed checksum committed; patch queue adds 0006.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Y8h447BPyJR4fioYZvQ9bh
2026-09-20 08:59:53 -06:00
noahandClaude Opus 4.8 8f7aab3f80 GUI round 11 (live-miner items #5/#6b/#8/#11): control-board scan gaps, VNish partner build, CVITek PIC bus
All root-caused and fixed against the live fleet (read-only scans; no device
state changed). asic-rs changes go through the vendored tree + thirdparty/asic-rs
patch queue (0003-0005) with refreshed checksums, per ADR 0003.

#5 (XIL/CVITek + WhatsMiner blank control board):
- VNish reports the board in short form in /api/v1/info "platform": "cv"
  (CVITek), "xil" (Xilinx), "aml" (AMLogic); AntMinerControlBoard::parse matched
  only the full names, so CVITek/Xilinx boards were blank. Added CV/XIL.
- Old M30S firmware (< 2022.7.29 -> v1 backend) omits summary's "CB Platform"
  and v1's parse mandatorily stripped an "ALLWINNER_" prefix. Added get_version
  /Msg/platform as a fallback source and made the strip optional.
- Verified live: S19k Pro cv->CVITek, S19 xil->Xilinx, M30S+ VE40->H6OS.

#6b/#11 (VNish partner / shared-devfee build):
- The partner build shows as /api/v1/info "build_name" ("WildRose",
  "vnishreturn"). The VNish backend (v1_2_0 + v1_3_0) now reads /build_name and
  combines it with the /fw_version number, so firmware_version reads e.g.
  "WildRose 1.2.7". Verified live on S19j Pro / S19k Pro.

#8 (hashboard PIC detection):
- pic_probe tried i2c-1/i2c-0 for CVITek, but a live VNish CVITek S19k Pro
  exposes the hashboard MCU on i2c-3 (i2cdump 0x48 = period-8 -> PIC). Broadened
  bus_candidates (cvi->[3,4,1,0], aml->[1,3,4,0], bb->[0,1,3,4], else->[1,0,3,4]);
  probe_pic stops at the first bus that classifies. The classify logic was
  already correct + unit-tested; only the bus map was wrong.

The GUI picks up #5/#6b automatically through asic-rs (CPU/Board columns now
populate for VNish CVITek/Xilinx and old M30S; FW Version shows the partner
build); #8 is fixed in the GUI's own pic_probe. The miner_probe diagnostic's
local-only mutation modes are left uncommitted as before.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Y8h447BPyJR4fioYZvQ9bh
2026-09-19 09:07:57 -06:00
noahandClaude Opus 4.8 8eb9a2caad GUI round 10 (#8): Z15/Z15J Equihash hashrate shows the correct SI prefix (k, not T)
Root-caused and verified against the live fleet. An Antminer Z15 shows its
hashrate as e.g. "288T" when it is really ~288 kSol/s ("288k"). Equihash miners
report a Sol/s rate in the cgminer "GHS" key, but asic-rs AntMiner parse_hashrate
takes the unit from the key name ("GHS" -> GigaHash), scaling the value 1e9x too
large. The value is a raw Sol/s count, so for Equihash use the base Hash unit;
the toolkit`s SI-prefixing then yields "288k".

Matched by algorithm, and by "Z15" model name for the Z15J (which asic-rs leaves
algo Unknown). Fixed in both the v2020 and v2023_07 AntMiner backends, via
asic-rs patch queue 0002. Verified live: Z15 288T -> ~288k, Z15J 138T -> ~110k.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Y8h447BPyJR4fioYZvQ9bh
2026-09-18 15:18:12 -06:00
noahandClaude Opus 4.8 9e2b453eb5 GUI round 9 (#4 v2): WhatsMiner v2 power-limit now reports success on the apply-restart
Follow-up to 3530d67a. Verified on the live fleet that the v2 (firmware
< 2024-11-01) "adjust_power_limit" command DOES apply the new watt limit --
confirmed the set value settles via the read-only :4028 "Power Limit" after the
miner restarts to re-tune. But the restart drops the connection mid-command, so
the read times out; the code then reported that as a failure even though it
applied.

Both v2 and v3 set_power_limit now treat a transient RPC error after the command
(is_expected_write_error -- read timeout / connection reset, the helper the
pause/resume paths already use) as success. A genuine rejection is a
StatusCheckFailed and still errors. v1 miners keep the "not supported" bail.

Verified: v2 M30S+ (fw 20240930) set 3500 -> :4028 settles to 3500, restore 3600
-> settles to 3600 and the miner resumes hashing.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Y8h447BPyJR4fioYZvQ9bh
2026-09-18 15:01:42 -06:00
noahandClaude Opus 4.8 3530d67a5c GUI round 9 (#4): fix WhatsMiner power-limit (v3 auth/param) + asic-rs patch queue
Round-9 #4 -- "the power-limit on the WhatsMiners doesn't apply properly" --
root-caused and verified against the live fleet, then fixed in vendored asic-rs
via a new patch queue (ADR 0003).

Root cause: asic-rs routes a WhatsMiner to its v3 (WhatsMinerOS) backend when
the firmware date is >= 2024-11-01. The v3 privileged power-limit write
(command "set.miner.power_limit" on port 4433) authenticated against account
"super" but hashed the token with the WEB/admin password, and sent the watt
value as a float -- so the miner rejected every write with "invalid json or
password is wrong". asic-rs then swallowed that (Ok(data.is_ok()) -> Ok(false))
so the GUI showed "refused" with no reason. (Older miners, fw < 2024-11-01, use
the v2 "adjust_power_limit" path, which restarts the miner on a change -- see
below.)

Fix (thirdparty/asic-rs/patches/0001):
- v3 set_auth: the "super" account's token password is a SEPARATE v3 credential
  (factory default "super"), NOT the web/admin password carried by `auth`; use
  "super". The web password still drives the v2 (admin) path.
- v3 set_power_limit: send the watts as a bare integer (a float trips the
  firmware's "invalid json" parser).
- v2 + v3: propagate the RPC error with `?` instead of swallowing it to
  Ok(false), so a real failure reaches the round-9 bottom-right error indicator
  with the miner's own message.

Verified on a live M30S+ (fw 20241108, api 3.0.1) via a read-only 4028
capture/restore harness: the fixed request returns {"code":0,"msg":"ok"};
asic-rs set_power_limit now returns Ok(true) and the miner applies the limit
(it restarts to re-tune, ~10 min ramp -- expected WhatsMiner behaviour). Matches
the pyasic and HashSource whatsminer-api-3 references. All test miners were
returned to their prior configured limit.

asic-rs patch-queue mechanism (ADR 0003 updated): asic-rs is cargo-vendored, so
in-place edits are documented as a numbered patch under thirdparty/asic-rs/
plus a checksum-refresh helper and scripts/apply-asic-rs-patches.sh for
re-vendoring. The committed vendor/ tree has the patch applied and its
.cargo-checksum.json refreshed, so builds need nothing extra (CI builds fresh;
a note documents the local `cargo clean -p` cargo-caching gotcha).

Known follow-up: the v2 backend (fw < 2024-11-01) reaches the miner via
"adjust_power_limit" but a changed value restarts the miner and the in-restart
status check reports failure; whether it applies correctly needs its own
verification. The error-propagation change surfaces that honestly meanwhile.

Compile-checked (release + tests); 79 offline tests pass.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Y8h447BPyJR4fioYZvQ9bh
2026-09-18 14:10:05 -06:00
noah fcc70c04f6 M1 primary view: wx-rs patched+extended (Linux/GTK3 build, widget bridge), vendored wx 3.3.3 + asic-rs in thirdparty/, UI-SPEC table+KPIs+dialogs, build scripts 2026-09-15 13:59:55 -06:00
noah a2b65bb47f Vendor asic-rs 0.8.1 (280 crates, offline builds) + wxWidgets 3.3.3 (pristine + patch queue); UI spec + ADR-0003; M1 on asic-rs 2026-09-15 10:41:00 -06:00