Conflicts resolved keeping both: engine.rs carries the sweep state (miner-eff) and the node watchdog (reliability);
the STATUS line goes through watchdog::parse_status and still feeds the sweep's rate sample; main.rs declares both
modules; bench-log.md keeps both entries. site/build.mjs keeps master's partial-injected pages and adds the /miners
bench table through the same page() with active: 'miners'; the Miners link is in site/partials/nav.html; sitemap
gains /miners; every generated page rebuilt with node site/build.mjs.
docs/bench-log.md: the fake-worker measurements (slow start one trip and 2.0 s restart, fake-fast guard in under
0.1 s, exit 43 at 8.8 s, CPU re-check stop at 0.5 s, stall guard at 60.1 s with STATUS lines through the silence,
one prepare per epoch with refused retries held; app: zero-rate restart at 79.6 s and faulted at 75.4 s on the
repeat, no-status restart at 90.4 s, silent node restarted at 150.7 s and synced 7.2 s later; no double restart on
the miner's own worker restart). Two defects the harness found are named with their fork commits.
docs/fud-ledger.md and the round-4 review table: M26, M27, X21 Fixed with the commits.
engine.rs: the miner's restart note no longer hides the fault reason on the card.
tools/reliability: the harness matches the miner's stderr lines where they are printed there.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Conflicts resolved keeping both: Settings and SettingsState carry the sweep fields (miner-eff) and dev_fee /
fee_total (dev-fee); the engine's settings snapshot sets both and the DevFeeState line stays.
Decision of 4 October 2026 (evening): the Igneum Miner software takes a visible, switchable 1% dev fee, the norm
for GPU miners; the protocol stays fee-free. The miner side (--dev-fee, the 1-in-100 template counter, the audit
command) is on branch dev-fee of the node fork.
- app: settings.dev_fee (default on) passes --dev-fee 0 to igneum-miner when off; Settings shows the miner's own
"dev fee 1% (1 block in 100) to 0x..." line next to the rewards address with a switch; the engine parses the
miner's start line and its dev-fee block lines (fee_session, lifetime fee_total, an event per fee block)
- packaging/hive: h-manifest.conf, h-config.sh, h-run.sh, h-stats.sh, make-hive-package.sh (igneum-hive-<v>.tar.gz
with the Linux igneumd, igneum-miner and both GPU workers), README with the Flight Sheet, selftest.sh (bash -n,
stub binaries, the three hooks the way Hive runs them, the stats JSON parsed). Hive itself is untested
- infra/cross/build-workers-linux.sh: the NVRTC and OpenCL workers cross-compiled for Linux with zig;
proto-opencl/cl_dynamic.h gains the Linux dlopen branch (libOpenCL.so.1)
- tools/dev-fee/run.mjs: the fee-block test network (two nodes on 29900+, three CPU miners, payouts audit)
- docs/design/miner-dev-fee.md (mechanism, flag, lines, the DEV_FEE_ADDRESS placeholder and the devnet address),
docs/fud-ledger.md E18 and the E5/L9 status line, litepaper "What a miner's hour looks like" paragraph, homepage
miner section note
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Conflict resolved: publish-jobs.sh keeps master's verify command and --tries (the retrying live check) alongside the
build kind's arguments; the usage range covers the merged header.
Conflicts resolved: state.rs keeps both the sweep fields (miner-eff) and the race fields (miner-perf); bench-log.md
keeps both appended entries; publish-manifest.sh keeps master's --override implementation (8082576, the "every
height switch" rule, --verify-only, --tries, the retrying live check) and adds miner-perf's --tuning / --no-tuning
with the carry-over of consensus.override and tuning from the current manifest. One --override case, one parser.
The first Windows update over the air (0.3.3 to 0.3.4) sat on 'the installer is starting' with no helper log on
PC 2. The helper launched by a remote job with the same arguments ran at once. Fix in 0.3.5; the Windows machines
take 0.3.5 by hand once more.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
src/sweep.rs (new): the cap steps 100% to 50% in 10% steps clamped to the card's limits, per-step rows (mean draw
from nvidia-smi power.draw, mean worker interval rate), the choice (best MH/W, ties to the higher rate then the lower
cap), the nvidia-smi power parser, a state machine on an explicit clock (15 s settle, 60 s hold, 30 s cap readback
limit), the elevated helper scripts (one administrator prompt per sweep: a command file polled by one elevated
process, self-restoring after 20 idle minutes), the unsupported reasons (Apple silicon, AMD). 9 unit tests with
PC 1's recorded RTX 5090 numbers (575 W default, 460 W cap, 290 W draw, memory temperature [N/A]).
Engine: scheduler (once after install, then weekly; one card at a time; only while the card mines, after 120 s
steady, never under a remote job hold, a pause, or inside 600 s of the hour boundary), the cap-mode probe (direct
when the engine runs elevated, else the helper), abort on any fault (card leaves mining, worker error, GPU 90 C,
job, pause, quit) with the cap restored, the chosen cap held and recorded, SWEEP table lines in the app log,
--sweep mode (sweep every supported card, print the table on stdout, leave the caps, quit). Cap floor 50% (was 60).
A readback that matches the asked cap now counts as applied (PC 1 showed "cap NOT applied" for hours at 460 W).
Dashboard: live eff MH/W on each tile, the sweep line (phase, last result, or why unsupported), Sweep now / Stop /
Unpin, "pinned" and "chosen by the sweep" on the cap line, the Settings toggle, the cards-page note, slider min 50.
A cap moved by hand pins the card: the sweep records but does not change it.
PC 1 measurement: relay/playbooks/sweep-5090.ps1 (a run job, elevated, miners stopped; a second engine with --sweep
in a scratch data folder, RESULT SWEEP lines) and docs/plans/miner-eff.md with the publish command. Not published.
Untested on a card.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
New job kind `build` (jobs.rs, jobbuild.rs, jobrun.rs run_build): free-space check on both sides (20 GB), the
build-inputs zip by sha256, setup inside the distro as root (mingw-w64 posix, clang for bindgen, protoc, zstd, the
Windows rust target; idempotent), sources extracted with the target dir persisting under /root/igneum-build,
cargo build --release native and for x86_64-pc-windows-gnu, cargo test for the manifest's packages, binaries
zstd-compressed and sent to the relay (fn=upload, Blob PUT, fn=drop; 50 MB each) with sha256 in RESULT lines,
STAGE lines with UTC times, a 40-minute default budget and per-stage caps, the Linux side killed on a cap. The
app's runner stays serial (one Active at a time), so a build never overlaps a shard job; nothing stops the miners.
From this version an unknown job kind is skipped by the app (parse_lenient) instead of rejecting the whole file;
the signer stays strict.
Mac side: packaging/windows/push-build-inputs.sh packs a fork worktree, app/igneum-app, brand/icons and
proto-cuda with a manifest (branch, commit, dirty, builds, tests) and the sha256; publish-jobs.sh add --kind build;
tools/build-job.mjs packs, publishes, watches, fetches, checks both sha256 per file and the PE header of every exe
(plus verify-exe.py on igneum-app.exe), and places the binaries where push-inputs.sh, make-payload.sh and the
cloud-devnet scripts look. relay.mjs drop <file> --body carries the body.
Tested on the Mac: 33 app tests (6 new) and the signer's 21; cargo check for x86_64-pc-windows-gnu; the packer
(7.9 MB zip, no target dirs); the publisher against a scratch folder with the rebuilt signer, the old signer
refusing the kind, a bad job refused at signing; the fetch path against the live relay with a real exe (sha256
and PE pass, a wrong sha256 refused; test items deleted). Not run on a PC: the job itself. docs/plans/build-job.md
has the first job for PC 1 and the rollout order (0.3.4 must be on the PCs before a build job is published).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Log drawer: the Logs button shows its open state (Close logs, chevron); a Newest button appears when the user scrolls
up and follow re-engages at the end; a time ruler with error and swap marks, an HH:MM:SS jump box, last error and last
swap buttons; substring search with a match count and highlights; copy the lines in view; the height drags from a grip
(140 px to 70% of the window, remembered); a window under 700 px tall opens a 180 px drawer. The list is virtualised
(19 px rows, only the rows in view in the DOM; up to 20,000 lines kept; wrap mode under 5,000 lines). Polling runs only
while the drawer is open and the window is visible.
Screens: one type scale and spacing scale in app.css, tabular figures everywhere, the bottom bar in pieces (machine name
truncates first, address always short, uptime dropped under 1100 px), tile captions carry their full text as a title,
syncing shows an ETA from the measured block rate, the engine-away state names itself on the node tile, short-window
rules, the key sheet scrolls on a short window, compact settings card rows on two lines, only a block that just arrived
flashes (not the whole strip on first paint).
Performance: the minute grid is drawn once into an offscreen layer; the strip redraws twice a second and at the display
rate only during a flash; nothing draws and the state poll drops to every 5 s while document.hidden. ?debug=1 prints a
budget line every 5 s and exposes window.__igneumBench.
tools/ui-mock: a stand-in engine (node tools/ui-mock/server.mjs) replaying recorded, scrubbed API responses with
scenarios (syncing, nodedown, nocards, integrated, clock, paused, biglog, fresh, job, nvidia), so UI work never
touches a running miner.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Review round 4 X21 and the project lead's levers 5 and 6 (4 October 2026 evening):
- src/watchdog.rs: CardWatch (no status line for 90 s, or hash rate 0 for 60 s while the node is synced: one
restart, then the card is marked faulted with the reason in the UI and the log; the miner's own worker restarts
suspend both rules until ready, so there are no double restarts; exit 43 counts as a watchdog restart) and
NodeWatch (our node silent for 120 s is restarted in-process through the restart kind the remote jobs use, with a
growing delay). State machines with no clock; 11 unit tests replay recorded STATUS and WORKER FAULT lines.
- engine.rs reads STATUS mismatched=, faults= and the WORKER FAULT lines onto the card (faults, mismatched, message);
a faulted card is not restarted until the user changes its settings or resumes mining; other cards keep mining.
- site/miners.html (build.mjs, scrubbed like /bench, rows from site/miner-bench.json): card, generator version, best
MH/s, MH per watt where measured, miner version, date, source, measured by the team or reported by the fleet. In the
navigation beside the engineering log on every page and in the sitemap. One line says there is no other miner to
compare with.
- tools/reliability: fake-worker.mjs (the serve protocol, misbehaving on command), run.mjs (miner guards on a private
test network, ports 29950+) and app-run.mjs (the app watchdog end to end, ports 29960+).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The engine turns a worker's race line into one TUNING {json} line in the app log (card model as the worker names it,
driver, arch, program class loads and wide loads, every variant's MH/s, winner, gain, the card's power cap and
draw, MH per watt), which the existing intake receives; the card state carries the variant for the dashboard and
one event per race. tools/tuning.mjs aggregates the records from miner_logs per card model (median MH/s or MH per
watt, at least 3 samples, de-duplicated per race) and writes tuning.json; publish-manifest.sh --tuning puts it in
the signed manifest (and now takes --override for consensus.override; both are carried over from the current
manifest when not given, --no-tuning drops it); manifest.rs parses it; ota.rs writes <app data>/tuning.json and
removes it when the manifest drops it; procs::spawn takes an environment and every miner starts with
IGNEUM_TUNING_FILE, which its worker reads at every prepare. Dry run of the publisher against a scratch folder:
tuning and override written, carried over, dropped, signature verified.
docs/plans/miner-perf.md: the signed jobs for PC 1 (fetch the race build of the NVRTC worker, then
relay/playbooks/race-5090.ps1 with the miners stopped: 17 variants, 3 rounds, twice) with the exact publish
commands for the main session; not published by the agent.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
PC 2 on 0.3.3: 'wsl -d Ubuntu-24.04 -u [user]' from the app fails with getpwnam([user]) and the default user has no
cargo, while the project lead's own session has both in a distro of the same name: WSL distros belong to the Windows account,
and the engine runs under a different context than the interactive session. So the prover path is self-sufficient
inside the Ubuntu the app sees: the wsl-prover probe and the shard-benchmark job run as root (the job's wsl_user or
IGNEUM_APP_WSL_USER first, root second, the distro default last), and the shard job runs the Linux host the
payload ships next to the app (wsl2\bin\igneum-prove-host, cuda feature) directly for each fixture (shard 0 in
shard mode, the blocks in block mode, results under <app data>\prove\igneum-prove-wsl2\results); params.build
= true keeps the package's prove-shard.sh path. Every job logs the account context first (Windows user, SID,
elevated, the signed-in console user, then 'wsl user <id> uid <n> home <h>' and nvidia-smi inside the distro),
the engine logs it once at start, and the dashboard (Settings and the job strip) says 'The app runs as X
(elevated). The signed-in user is Y; WSL and its tools belong to that account.' when they differ.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
PC 2 on 0.3.3 logged 'needs wsl-prover' although the toolchain is there as user [user]. Every negative probe now
lands in the engine log with its exit code and the first 200 characters of output (it reaches the intake). The
probe sources ~/.cargo/env and sets ~/.cargo/bin and ~/.sp1/bin itself (a login shell from a console-less process
need not), runs as the job's wsl_user or IGNEUM_APP_WSL_USER first and the distro default user second, and a
payload with wsl2\bin\igneum-prove-host next to the app meets the requirement when the distro answers.
publish-jobs.sh: --requires none or "" publishes an empty list (none was a literal requirement, "" fell back
to the kind's default).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
d5398f3 switched only wsl: BSD sed dropped the alternation. Now every Command::new in src/jobrun.rs goes
through platform::tool (R4.3.3).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The job runner launched its helpers by bare name; binary planting into a writable PATH entry fed an elevated
prompt was the round-4 finding. Fetched files were already sha256-checked before use (fetch and shard-benchmark
refuse a job without a 64-hex sha256 at parse time, fetch_file verifies before the rename).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Every Windows executable now ships with the coin icon Explorer shows and the version block Properties shows
(CompanyName Igneum, ProductName Igneum Miner, FileDescription per exe, 0.3.0, LegalCopyright Igneum contributors),
like the Mac app and DMG already do.
- igneumd.exe, igneum-miner.exe: packaging/windows/embed-resources.sh now takes the worktree and target dir
(defaults vendor/igneum-node-v4 and vendor/igneum-node/target-integration) and relinks with a linker shim first in
PATH instead of `cargo rustc -- -C link-arg`: cargo rustc takes one package and unifies features differently from
the two-package cross-build (300 crates differ), and a configured linker is fingerprinted and rebuilds everything;
the PATH shim changes neither. .rc files moved to 0.3.0.
- igneum-worker-cuda.exe, igneum-worker-opencl.exe: proto-cuda/nvrtc/build-windows.sh compiles the two new .rc
files with windres and links the objects; the icon is made with make-icons.py if missing.
- igneum-app.exe: app/igneum-app/build.rs runs windres on resources/igneum-app.rc for Windows targets and links it
(cargo:rustc-link-arg-bins, no crate dependency); it fails the build if the .rc version drifts from Cargo.toml.
- packaging/windows/resources/verify-exe.py: small PE parser that checks the .rsrc section, icon group and version
strings on the Mac; both build scripts call it, and it checks every exe inside the packages.
- proto-opencl/cl_dynamic.h: clGetEventInfo added to the run-time loader table (4714b51 added the call in host.c
without it, so the one-click OpenCL worker no longer linked under IGNEUM_CL_DYNAMIC).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
windows.yml: parse job (every .ps1 through the Windows PowerShell 5.1 parser with a negative fixture, PSScriptAnalyzer
as warnings, parenthesis check of every .bat/.cmd with a negative fixture), build job (engine on the MSVC target,
window host through BUILD-APP.bat as it is, payload through make-payload.sh in Git Bash, installer through
build-installer.ps1, smoke run of both exes, launcher DRY_RUN, three artifacts for 90 days).
push-inputs.sh publishes payload-inputs.zip (node, miner, workers, NVRTC DLLs) to the downloads host from the Mac;
fetch-ci-artifacts.sh pulls the green run's installer and payload back into the downloads folder.
Host: --version and --help, version.h shared with host.rc. Launcher: DRY_RUN=1 prints the plan and starts nothing.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
