Commit graph

1486 commits

Author SHA1 Message Date
igneum-labs
afa63e344f Counter ASIC 3.0 gates (hash): the fud-ledger entry AP-F8-1 (the fault, the project lead's ruling, the amendment, the split protection, the owed tests); the suite, pairing and G1 lines fill when the jobs land
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 09:41:11 +00:00
igneum-labs
a5b8fee21e Counter ASIC 3.0 gates (hash): the PC 2 G1-only playbook for the amended class v4 packs (AP-F8-1, limited testing by the project lead's word)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 09:40:03 +00:00
igneum-labs
0a62293a95 Counter ASIC 3.0 gates (hash): the class v4 amendment for AP-F8-1 (the project lead: option A, limited testing). The chain draw of class v4 takes a load's source only from registers whose last writer injects (add, sub, xor, mad, shfl, load) or is a rotate (rotl, rotr), never one last written by or, mul or mulhi (candidate_from_words_class, keyed on the era-composed V4_CLASS; no attempts lost, rule (a) unchanged, v2 and v3 and the generator-2 ladder packs byte-identical). Split protection agreed with the node lane: generator stays 4 and the program id of a generator-4 program appends "sub/" || PROGRAM_SUBVERSION_V4 (= 1) as little-endian u16 bytes inside program_id(), so a pre-amendment binary and this one never share an id for one seed; packs carry IGNEUM_PROGRAM_SUBVERSION 1 and program.json sub_version 1, and packcheck refuses a generator-4 pack whose sub-version is absent or other. The seven gate packs re-exported (devnet epoch 0 and eras 0 to 5): id 1a4230699a6b9c60 (was c120d7963abdcd96, now the must-differ vector in tests/recheck.rs), fingerprints Metal = Apple OpenCL 867dbc45cfb36b4d, 2146ecacc8c75a8e, fe52602393f6d3d4, 3b206471a13912b4, c3f03c4a5d7333aa, f1dfd7209f15bb97, 8c194da64fadf31d; the v3 control 73bcbfe8ccf988f1 / 90f794dd556f7a3b untouched. The mixer harness checks the source rule on every load site of an amended program instead of equality with the v3 base
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 09:38:51 +00:00
igneum-labs
bb095461d6 Merge branch 'ca3-v4-uniform' into ca3-v4-amend 2026-10-07 09:33:11 +00:00
igneum-labs
4a4bcac0b9 Counter ASIC 3.0 gates (hash): AP-F8-1 under the windows-union model. The window layer moves the null from 0.115 to 0.160 percent at f = 0.1 percent (1.39x, not 4.05x); the 153x item is the all-ones value of an or-written load source (site 15, or at 61, load at 63), which the era map sends to item 0xca5b92 exactly and whose next seven items are the seven surviving one-zero-bit sources; the popcount model at the measured bias predicts 77,348 reads against 78,479 and the program's S_0.1 percent at 0.58 against 0.52; a static census of 1,024 chain-shaped v4 programs (tools/ca3-v4-uniform, built and run on igneum-build-1): 96.6 percent carry a lossy-sourced load, 48.5 percent an or-sourced one, 4.9 percent an or chain (p3's class); F8's 1.2x gate fails 96.6 percent; the chip ceiling under rule (c) is 1.067x; the two flip options priced; the v5 bound defined
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 09:26:48 +00:00
igneum-labs
670afb8c7d Merge site-ui-4: the EMBER 02 renderer on the home fold and /live (2.0.1 real-data options, the proof inspector, the observatory page)
renderer pack EMBER 02 supplied by the project lead, 7 Oct 2026

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 09:11:43 +00:00
igneum-labs
06f1cd6e86 site-ui-4: live-dag.js 2.0.1, the real-data presentation options (lanes capped at 7 with the own key first, node size from the lane height with a floor, hairline merge edges within two lanes and the first parent only, the others lane smaller and dimmer, a glow and ring on proven blocks); 60 s default window; the demo-feed proof capture beside the live one
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 09:09:24 +00:00
igneum-labs
cf0d4238e5 site-ui-4: the your-blocks tile reads 0 in ash until a key is marked
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:59:58 +00:00
igneum-labs
a4fe3d8e08 site-ui-4: the EMBER 02 renderer on the home fold and /live, the proof inspector, the compact copies for the app (prototype, not deployed)
renderer pack EMBER 02 supplied by the project lead, 7 Oct 2026: src/live-dag.js
and src/proof-core.js adopted byte for byte (sha256 582f2e73... and
c2a094a6...), reviewed as third-party code (one same-origin fetch of
/api/live when poll is on, no eval, no CDN, no font download, tokens
read from :root in both themes, checkpoints placed only on an exact
blue score, proof never implies inclusion, a lock only when reported),
its tests/verify.py run on this Mac (73 of 73) and the module mounted
against the real observer feed (state live, only /api/live requested,
zero records promoted). Home: the pack's full mode in the one-screen
fold with the legend row and the newest block's caption, fed by the
page's own reader. /live: the pack in the graph slot, the four miner
stats (blocks in window, your blocks, proven, latest checkpoint), the
All blocks / Selected chain / Your blocks tabs, Pause and Follow, the
window stepper, the inspector column with IgneumProof's shard scene,
the proof shards list, inclusion, miner key, blue score, DAA, parents,
header time, checkpoint, finality, and the three-beat Motion, with
meaning block; the observatory's six tiles, identities, event stream,
block activity, light client and table stay. site.css gains --included
and --excluded. The standalone demo is the pack's standalone.html; the
10-second recording is from this branch's /live.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:59:46 +00:00
igneum-labs
c1d97e776c Income per tier: the three tariffs are USD 0.05, 0.10 and 0.25 a kWh (cheap industrial, US retail, UK retail; the coordinator's answer, 7 October 2026), the table regenerated, the cost per mined IGN shown to three significant figures
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:54:44 +00:00
igneum-labs
c69b1cec58 Launch pack: the thirteen gate rows in testnet-go.md with their checks, the certificate runbook and signing-step specification, the hash-origin runbook, the site-lane text handoff, what needs the owner
docs/plans/testnet-go.md, "Launch gates": LG-1 income per tier (done, re-generate at the cut), LG-2 hash origin on Devnet 2 for seven days, LG-3 signed and notarised installers, LG-4 first-share time per release, LG-5 the text on the site, LG-6 to LG-10 the community gates (first 100 keys, first outside block, first outside-reproduced benchmark, first attested outside pool, first 1,000 independent keys), LG-11 a signed proving customer as a MAINNET gate (the owner, 7 October 2026: a customer paying at a published rate or a letter of intent with a volume), LG-12 the report daily for 90 days, LG-13 the USD 50,000 disclosure prize (approved, staged until the entity address and the publish word). Each with its check, its state and its owner.

docs/plans/launch-pack.md: the Apple organisation and EV Authenticode runbooks (provider, cost, the entity's documents, where the keys live, never on the box), the signing step for the shipper (windows.yml before ISCC and the Inno SignTool directive, build-dmg.sh Developer ID plus notarytool and stapler, the manifest's signed_by and notarized fields, the shipper's fetch and verify checks), the first-share measurement specification, the hash-origin timer on the box and today's reading with its consequences, the handoff block for the site lane (the three wants on the front page, finality on page two, the block sentence, the eight sentences of future.md 8.3 verified and numbered under the label, the journey gate changes, the ledger rows X13, X37, X38, E23, L10), and the nine items that need the owner.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:53:12 +00:00
igneum-labs
6c739c543d Launch pack tools: per-tier income table from the bench rows and TESTNET_1, the daily hash-origin report, the launch-gates check (mission item 10)
tools/launch/income-tiers.mjs renders docs/analysis/income-tiers.md (public) from tools/launch/income-tiers.json: eleven measured cards (the 6 October rented-card rows, the 9070 XT telemetry run, the M5 Max Metal bench, each with its source), IGN a day at 1, 10 and 100 GH/s after the ramp on EmissionSchedule::TESTNET_1 (100 IGN a block, 90-day ramp from 10 percent, monthly 2^(-1/24), the 80 percent producer share), electricity a day and per mined IGN at USD 0.005, 0.02 and 0.10 a kWh, a rig and a pool-user line, the consequences per tier, the owed rows; --check fails CI when the page and its inputs disagree; the test pins the arithmetic (6,912 IGN a day for 100 MH/s on 100 GH/s at the launch rate, day 1 at 10 percent, one step at 2^(-1/24)).

tools/observer/hash-origin.mjs: once a day from the observer's tables, who found the blocks: keys with a block and above dust, attested pools against shared payout addresses (a multi-key machine is not a pool until its operator attests), the project fleet's share from the intake identity lines plus the fleet registry's key file, the ten largest keys, the 2x step since yesterday, the community gates (first 100 keys, first outside block, first attested outside pool at 10 percent for 7 days; the X5 count stays an upper bound until the two observer columns exist). --dry reads only; --write keeps hash_origin_days and hash_origin_reports and live_state.hash_origin; --post goes through the Discord poster's guard. Fixture, a known-finished and a known-failed day in the test. Ran read-only on the live devnet today: 113 keys, 0.76 GH/s, the fleet share wrong until the key file exists.

tools/ci/launch-gates-check.mjs (in pre-push.sh with the two test lines): every row of the Launch gates table in testnet-go.md has a check and every backticked path exists; the site-lane handoff and the income table carry no served-page or export pattern and no em dash; the eight regulatory sentences are present in order under the label. Self-test fires on each.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:53:12 +00:00
igneum-labs
4ce0c34827 site-ui-4: /live as the network observatory (prototype, not deployed)
The reference page's composition in our tokens and words, every number
from the observer: the header row (eyebrow, the title with its ember
full stop, one sentence, two buttons, the state word), six tiles in one
hairline row each with a definition, the block graph panel (our
live-dag.js in the graph slot, the block-count chip, selected chain
only, pause and follow, the inspector column with hash, lane, time, blue
score, DAA, chain, colour word, shards, checkpoint and lock state, the
drag-to-explore line, the zoom control, the legend row and the blue
score range), the three panels (block activity with per-minute bars,
mining identities with share bars and the reader's own key marked when
set, the event stream), the blocks table, our footer. live-dag.js gains
an onSelect hook on click (additive; the app's copy unchanged).

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:30:23 +00:00
igneum-labs
210084b0e5 build server: the remote checkout's clean spares a lane's scratch (AP-H1, the lost-scratch class)
remote-run.sh checkout_tree ran `git clean -fd` on the box mirror before every build from any agent, so the attack rows
lost attack-f3/, attack-f1-venv/ and tools/attack/*/target to each other's builds (7 October 2026, 09:2x UK). The clean now
also spares the fixed prefixes attack-*, scratch-*, target-attack-*, .build-remote.log and every glob in the mirror-local
.igneum-scratch-spare (one per line, # comments, the file itself spared), keeps the target and stamp excludes and still runs
without -x. The clean-tree test asks `git clean -nd` with the same excludes instead of filtering the status list, so a spared
dir is not "not clean". --self-test: a fixed-prefix dir at the root and nested, a declared dir and the spare file survive, an
undeclared dir is removed. tools/ci/scratch-spare-check.sh in the pre-push gate fails when the clean line loses the spare
arguments, spare_args stops reading the file, a fixed prefix goes, or -x appears. docs/plans/build-server.md R4a says how a
lane declares its prefix.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:24:07 +00:00
igneum-labs
d91fe9bbb3 Merge mission: the last research round (five lanes: past, unfinished, future, invent, reinvent) and The Igneum Mission, the closed list of twelve
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:22:14 +00:00
igneum-labs
bbe0acce28 Export exclusion: docs/analysis/mission is internal research (the last mission's five lanes and the closed list), outside the public export list
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:22:09 +00:00
igneum-labs
3ae76a3047 The Igneum Mission: the one page for the project lead and the closed list of twelve (five in flight, seven new, about 290 agent hours, no new consensus rule beyond the three cut branches), the rejected ideas with the line that killed each, what was not run, and the closing line: nothing is added without removing something
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:19:29 +00:00
igneum-labs
000a976209 Mission lane 2, unfinished: proof of useful work (every scheme that passed cheap-verify did so by making the work useless; Filecoin at 29 to 36 percent utilisation), the stalled ASIC-resistance programmes, merged mining, finality on proof of work (every one that held was a second set bought with coins), decentralised pools (only Monero's P2Pool at 7.7 percent; a share sidechain about 40 hours on Igneum), the one-click miner apps against Ember, the three-way verdicts
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:18:20 +00:00
igneum-labs
e4efea469d Mission lane 5, reinvent: the miner's dopamine loop from a 105-post sample (the miner's own card at 777x the room median, the network's milestone at 66x), Helium, Chia, Ethereum 2017 and Kaspa 2022 compared, the first month on Igneum per tier (an 8 GB card: a block in 1.6 h at 100 GH/s, never the vote line at 1 TH/s), the 30-day weight as the status ladder, the launch gates, the surfaces table, what not to build
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:15:30 +00:00
igneum-labs
83069944e2 Mission lane 4, invent: proof of following, serving, propagation and state availability judged against class v5 (the 108 ms answer kills every challenge scheme), the in-protocol pool priced (FruitChains shape, 14.6 GB a day at 10,000 miners), heat mode, the 786 B offline finality QR, the self-generated candidates, four persona verdicts, the honest answer
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:08:01 +00:00
igneum-labs
79f3096219 Mission lane 3, the far future 2030 to 2036: HBM4 and the end of GDDR, the chip maker's break-even by node and market cap, the rental deflation curve against the 51 percent table, heat and grid rules by region, zk proving cost to 2036, phone light clients, post-quantum votes at 8,192 voters, regulation of a coin with no issuer, the three-scenario table
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:08:01 +00:00
igneum-labs
f3df738a35 Mission lane 1, the past: every GPU-mined chain since 2011 (31 rows), what each promised and paid, the earnings-curve shape (income halves 60 to 120 days after peak, under USD 0.10 per kWh in 6 to 18 months, 50 to 100 percent of hash gone in 90 days), the three wants and three hates from the archives, the lessons mapped to the design
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:08:01 +00:00
igneum-labs
25a1a310e9 Merge site-ui-3: light-theme wallet captures, the miner light showcase
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:07:55 +00:00
igneum-labs
6785f71a93 site: light-theme wallet captures (0.1.5, rendered from the wallet's mock), the miner showcase frame differs from the hero in light
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:07:53 +00:00
igneum-labs
107090dba9 Merge site-ui-3: the site redesign (simple nav, miner, app, live devnet, one design system), X36, the Discord vanity
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:01:06 +00:00
igneum-labs
583eeea437 site: light-theme app captures, the X9 withdrawal correction (ledger X36), the Discord vanity link
Light theme shows light screenshots on the miner and app pages (the
0.3.16 renders from the RTX 5090 Windows rig's live state, the 390 px
pair included). The X9 never shipped: pre-orders 26 December 2025,
withdrawn mid-May 2026 before any unit shipped, no benchmark; every
public sentence that had it shipping now says so, with k about 0.33
labelled as the claimed, unmeasured core; ledger X36, notes on X34, X35,
M34 and C2. Every public Discord link is https://discord.gg/igneum.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 08:01:03 +00:00
igneum-labs
6e8335d9b9 site: no paragraph above the fold on /live; the v5 capture set (every page, 1440 and 390, dark and light, fold and full)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 07:51:42 +00:00
igneum-labs
7e44748295 site: the simple nav, the miner and app pages, the live devnet redrawn, one design system for every page (not deployed)
Nav sitewide: Litepaper, Miner, App, Wallet, Live devnet, Ledger and the
Download button; the rest in the sheet and the footer. /miner is the
product page for the installers: the shipped app in its frame, the card
picker with the live time to a block, the downloads with sizes, HiveOS,
the fee in full view, three real shots. /app is new: the Igneum Ember app
page by page (Overview, Cards with Ember, the chain drawn live by the
module the app ships, Earnings, Prove, Settings, phone width, what keeps
it mining), light and dark, every shot a real capture or a render from
the RTX 5090 Windows rig's live state with the hostname painted out.
/live is redrawn to show the system working: the step scene, six live
cells, finality in plain words, miner lanes, provers paid, recent blocks
landing, the light client in one line, the DAG below. site.css carries
the shared product components (frames, download list, lines, your-card
line, phone download bar); yourcard.js reads the network rate. The
ledger and wallet pages no longer overflow at 390 px.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 07:44:15 +00:00
igneum-labs
c7149bdd6a Ledger decisions: the project lead approves the emission schedule, the signing bonus, the ladder, the cryptanalysis engagement and the testnet genesis re-cut (7 October 2026, 09:3x UK)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 07:36:12 +00:00
igneum-labs
02d7f5ab81 Observer sync follows GitHub: the deploy-key probe captured before grep (ssh -T exits 1 under pipefail); plan 5c done
the project lead added the read-only deploy key on 7 October 2026 (SHA256:51ice3W8...). The sync still said 'mirror' because ssh -T to GitHub
exits 1 after its greeting and observer-sync.sh runs under pipefail, so su ... | grep -q reported failure although grep had
matched (the same line by hand, without pipefail, said authenticated; shown under the unit's environment with systemd-run -v).
The probe's output is captured first. First github pass 07:30:54 UTC: the clone moved from the mirror's d79f4a6 to origin/master
8b878ba4, the observer restarted on it.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 07:31:42 +00:00
igneum-labs
8b878ba43c Merge site-ui-3: the one-screen home page is back
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 07:14:26 +00:00
igneum-labs
fc77e88683 site: the one-screen home page is back (84b349e), with the wording that landed since
The owner's word this morning: the 56 px header, the eleven-word
statement, two buttons, the step scene full bleed, three facts, the
downloads and one line to the ledger. Kept from the nights since: the
X31 testnet sentence, the X35 chip range with k stated, the Discord
line. The RandomX correction (X34), G4, C2 and X8 stand on the
litepaper; the ledger notes and the text check say so. The app showcase
moves to the miner page in the next step.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 07:14:24 +00:00
igneum-labs
3cc5e5c2f0 Hands mover: on 0.3.18 trees the restart read-back takes powEngine and the blockrate object from igneum_getNodeInfo; a stub answer stops the sequence
The shipper's read-back for 0.3.18 (7 October 2026): igneum_getNodeInfo exists again and must answer powEngine igneum-pow (a stub
is a FAIL) with a blockrate object, which the mover prints; a tree before 0.3.18 answers -32601 and the engine is still read from
the binary's igneum-pow source paths. The parser is checked against the three answer shapes.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 06:25:12 +00:00
igneum-labs
081f832b01 Hands plan: the 0.3.17 hotfix run log (hands and seed on 5899f603, read-backs, times, the pairs held for 0.3.18)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 05:33:45 +00:00
igneum-labs
fd4785d2bc lib.sh: bs_pin gives an empty pin for a tree without rust-toolchain.toml (a failed sed under pipefail ended the caller silently)
The b3c228fa builds under the 0.3.16 app tree 5d118f58 (no rust-toolchain.toml yet) died right after the pairing line with no
message: sed on the missing file failed, pipefail carried its status into the assignment, set -e ended the script. A guard and a
tolerant pipeline; checked alive against a tree without the file.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 03:44:26 +00:00
igneum-labs
db6475def5 Build tools: whole-body blocks (the edited-while-running class, with its check), the igneum-pow pairing line, move-hand.sh restart with its readbacks
Three classes from the 0.3.17 night. (1) bash reads a script incrementally: tools/build-remote.sh was edited while a four-minute remote
build ran, the running copy continued at shifted bytes and died with a syntax error after the build had succeeded on the box; the
four long-running tools (build-remote, cross-remote, workers-remote, move-hand) now keep their body in one brace block ending in exit,
parsed whole before a line runs; tools/ci/whole-body-check.sh (in the gate, self-test with a block-less copy) holds the shape.
(2) A fork build pairs with the igneum-pow of the igneum worktree it sits in: a fork at 12153428 under a master worktree failed in
kaspa-pow four minutes in (no chain_program_shadow; master's igneum-pow predates release-0.3.17's); build-remote.sh says the
pairing on its first line ('pairs with igneum 6f8d7a7e (detached): igneum-pow 0.2.0') and the JSONL line carries pairs_with.
The first version of that line used '[ -n ... ] && echo' inside an assignment's $( ) and set -e ended the script on the false
status; fixed. (3) move-hand.sh restart <hand> [--digest <hex>] [--go]: after binary installed a release, restart ONE unit and read
it back (first exec line, commit string in the running binary, digest against the wanted one, igneum_getNodeInfo powEngine over the
node's loopback EVM RPC); the digest readers tolerate a missing line.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 03:12:51 +00:00
igneum-labs
3720c40bce glibc ceilings per artefact class: hive and rig 2.31, seed and linux 2.35, native unchecked; proven in ubuntu:20.04 and 22.04 on the box
Main's order of 7 October 2026 after RunPod's Ubuntu 22.04 canaries (glibc 2.35) refused the box's GLIBC_2.38 binaries and HiveOS
turned out Ubuntu 20.04 based (2.31). The table lives once, in tools/ci/glibc-ceiling-check.sh (--class, --ceiling-of; self-test
covers the table, an unknown class and a 2.34 need against hive); tools/build-remote.sh --ship hive|rig|seed|linux (default seed)
and tools/workers-remote.sh --class (default rig) build with zig at the class's glibc and check against it. provision.sh installs
docker.io (user build in the docker group) for the proof. Proof: fork 3bfe346f at class hive, igneumd and igneum-miner need
GLIBC_2.30; the workers at class rig need GLIBC_2.17; all four run in ubuntu:20.04 (ldd 2.31) and ubuntu:22.04 (ldd 2.35) and
print their version or usage lines (the OpenCL worker its own no-libOpenCL message, the binary running in a GPU-less container).

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-07 00:35:18 +00:00
igneum-labs
01aa9fae83 Deploy key: the plan's section 5c (the project lead's four steps and the public half) that 9215d624 named but did not carry; observer-sync.sh runs git as build
The previous commit's plan edit aborted on a changed anchor, so section 5c was missing; written now with the public half
(ssh-ed25519 ... igneum-build-1 observer read-only) and the CI-runner row closed. observer-sync.sh ran git rev-parse as root
against the build-owned clone (safe.directory refused it, 'up to date at' with no commit); every git call runs as build now.
Verified on the box: one sync pass prints 'source: mirror (the deploy key is not accepted by GitHub yet ...)' and the commit.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-06 23:49:20 +00:00
igneum-labs
9215d624d2 Observer clone: a read-only deploy key made on the box, GitHub preferred over the mirror once accepted; the project lead's steps in the plan
Main's order of 7 October 2026. install-hands.sh creates /srv/observer/.ssh/deploy_igneum (ed25519, user build, mode 600; the private
half never leaves the box and nothing prints it) and the ssh alias github-igneum-observer; observer-sync.sh tests the key with
ssh -T on every pass, pulls from GitHub when it is accepted and from the mirror otherwise, saying which. docs/plans/build-server.md
5c: the four steps for the project lead (print the public half, Settings > Deploy keys > Add, read-only, start one sync pass) and the public
half itself. Verified on the box: key created, alias written, one sync pass reads 'source: mirror (the deploy key is not accepted by
GitHub yet)'. The CI-runner row is closed (the box-work agent's runner service).

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-06 23:47:10 +00:00
igneum-labs
49ca718c9c rust-toolchain.toml: one pin for every side (1.99.0, the two cross targets); the mismatch refusal reads it
Main's order of 7 October 2026. The repo root carries rust-toolchain.toml (channel 1.99.0, targets x86_64-pc-windows-gnu and
x86_64-unknown-linux-gnu); rustup resolves the nearest file walking up from the crate, so the fork worktrees under vendor/ are
covered, and the fork's master carries its own copy (vendor/igneum-node 37f1206b). lib.sh bs_toolchain_check reads the pin
and refuses a build when the pin, the Mac's rustc as resolved in the crate dir, or the box's rustc differ (the message says
the three commands that align them); run-from-mac.sh passes the pin to provision.sh as RUST_TOOLCHAIN. The 1.99.0 toolchain
with both targets is installed on the Mac so no agent's build stalls on rustup's auto-install. Verified: the Mac resolves
1.99.0 in a fork worktree, the box runs 1.99.0, the check prints 'pinned 1.99.0 by rust-toolchain.toml'.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-06 23:44:20 +00:00
igneum-labs
fff352e52f workers-remote.sh: the host's /usr/include stays out of the zig build (the OpenCL headers through a CL-only symlink dir); both workers proven at glibc 2.36
The previous commit's 'the workers at 2.36' was not yet true: the first zig build died on __isoc23_strtol because -I /usr/include
for CL/cl.h put Ubuntu's glibc 2.39 stdlib.h in front of zig's bundled 2.36 headers. Fixed; proof on the box (6 s):
igneum-worker-cuda 6,759,272 B sha256 690c8e91..., igneum-worker-opencl 307,264 B sha256 329fb6a9..., each needing GLIBC_2.34
and libc only, the ceiling check passing. The two static flags zig ignores are dropped on the zig path (zig links its libc++ in).

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-06 23:40:14 +00:00
igneum-labs
7089aa161f Build server: zig and cargo-zigbuild on the box; build-remote.sh --ship builds glibc 2.36 Linux artefacts for seeds and HiveOS; the glibc ceiling check
Main's order of 7 October 2026 after a seed took 14 restarts and three minutes down on a glibc 2.39 binary. provision.sh:
step_zig (zig 0.17.0 from ziglang.org, sha256 from the official download index) and cargo-zigbuild 0.23.4 in the cargo tools.
tools/build-remote.sh --ship [--glibc 2.36]: cargo zigbuild --target x86_64-unknown-linux-gnu.2.36 with zig as the C/C++
toolchain (the Mac's infra/cross/build-linux.sh recipe), artefacts from the target-triple dir, each checked by
tools/ci/glibc-ceiling-check.sh (need at most the ceiling; self-test fires on 2.38 against 2.36, passes 2.34 and 2.36;
--symbols reads a saved objdump -T text so CI needs no ELF tools). tools/workers-remote.sh builds the two GPU workers with
zig at 2.36 by default (GLIBC=native for clang). Proof on the box: fork 3bfe346f igneumd needs GLIBC_2.34 (47,023,120 B,
sha256 345dfb95...), igneum-miner GLIBC_2.34 (9,248,808 B, d09dc27b...), 3 min 17 s cold through zig; the workers at 2.36.
Rule: anything that ships to a seed or a HiveOS rig is built with --ship; a plain build (glibc 2.39) is for the box and
Ubuntu 24.04 hosts only.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-06 23:37:33 +00:00
igneum-labs
2750361894 Merge remote-tracking branch 'origin/master' into HEAD 2026-10-06 23:35:37 +00:00
igneum-labs
89300aafe6 Merge branch 'gpu-fleet' into HEAD 2026-10-06 23:33:54 +00:00
igneum-labs
db1cebd1f8 Fleet: the last two process patterns (a top-level alternation; the canary's node path by its basename)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-06 23:33:54 +00:00
igneum-labs
9e6a325128 Merge branch 'gpu-fleet' into HEAD
# Conflicts:
#	.github/workflows/ci.yml
2026-10-06 23:33:28 +00:00
igneum-labs
22ef7c3f97 Fleet: the last process patterns in the bracket-first form; the fleet's own pgrep check withdrawn (master's kill-by-name check covers the class)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-06 23:31:55 +00:00
igneum-labs
d41cb3876c Merge branch 'gpu-fleet' into HEAD 2026-10-06 23:29:39 +00:00
igneum-labs
238fccfce6 Fleet scripts: no path into another worktree (sys.path from the file, the ship tree from the environment)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-06 23:29:39 +00:00
igneum-labs
6aad8d8a53 Merge branch 'gpu-fleet' into HEAD 2026-10-06 23:28:55 +00:00