From 6ebce7459e94ce2ccf8eeed2e7b337e3f586c75f Mon Sep 17 00:00:00 2001
From: igneum-labs <337424239+igneum-labs@users.noreply.github.com>
Date: Fri, 9 Oct 2026 09:06:47 +0000
Subject: [PATCH] Counter ASIC 3.0 status: the founder's research plan and
 Track B fanned out (B1 to B6), X1's census, traces and hit curves (the window
 concentration 56 to 61 percent, not 72), X0's decider (class v6 costs the
 5090 1.75x class v4) and its op-count band, X3's flat grid, X4's first GPU
 row, the stale-sink class closed on the moving chain, build-1's OOM, the
 Windows canary by KVM, the reset after wave two

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---
 docs/plans/counter-asic-3-status.md | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/docs/plans/counter-asic-3-status.md b/docs/plans/counter-asic-3-status.md
index b34279628..1c6e159e9 100644
--- a/docs/plans/counter-asic-3-status.md
+++ b/docs/plans/counter-asic-3-status.md
@@ -874,6 +874,11 @@ THE MINI'S REBOOT LEG PASS AND THE MINI MINING ON THE MOVING CHAIN (the shipper,
 
 THE WORKERS PAGE PUBLISHED (the build-server lane, 09:50:00 UK, inside 10:30): build.igneum.network/workers.html from master 9c4ed1f0 (the load by process per box, the founder's collapsed panels with the open set per viewer, the mini's miner line, the rig cards' honest report age; no pid number, no machine name, the PCs as rig 1 and rig 2, the Mac as the macOS build host), published from build-1 under /srv/workers/deploy-page.pid after the steward's served-identity guard read the page, shape.js and the served workers.json clean (53 patterns, the hashed private list on the box); the edge sha256 workers.html a904d62f39f9ec3d, shape.js 970e3009627ae718; the edge copies read from the Mac against the private and generic lists (0 hits); the served workers.json guarded on every 20 s tick; the daily edge check in the build user's crontab at 06:17 UTC; the earlier publish at 09:34 refused by the guard on the unfixed page (the guard working); the rig cards to read "agent not reporting since <time>" on the next publish. THE 09:49 PER-BOX READ (the build-server lane, by pid): build-1 (load 25.8): the node lane's two 27f54124 moving-chain nodes, the seed, the hand and the light-reader, the capacity lane's pow mixer fuzz, the hands observer-node and node1, the dn2 seed, the indexer and three observer.mjs, a class-v5 lease self-test; build-2 (2.6): the site lane's gate only, no matrix job (FAULT: X5's design not running, the site lane holding the box); build-3 (0.00): nothing (FAULT: the node lane's consensus fuzz, X5's design); build-4 (9.7): the adversary lane's three jobs (the xbar detail route 18 h 22, the LEC on core32 13 h 44, the mf32full yosys synth at 22 percent under lease pool 24); build-5 (0.02): nothing (FAULT: the TV-02 generators, X5's design); build-6 (0.03): nothing (FAULT: the HEAL lane's fuzz, X5's design, the Token Value generator); build-7 (0.33): the dn4-roll igneumd only (FAULT: the node lane's 2.0.3 flows suites, X5's design); build-8 (13.1): X1's census (three addrsite runs at about 1,440 percent under census.sh, queue pid file build-8-x1.pid), the dn4-roll igneumd; build-9 (5.9): X1's suite (the v7flat test binary, build-9-x1.pid) and X4's CPU differential (build-9-x4.pid), no Token Value generator at that minute. MAIN'S IDLE-BOX RULE (09:5x UK): each other owner starts its job on its box under /srv/queue/pids within 15 minutes or the queue reader reassigns the box to the next entry; the 10:35 line to main names every box with its running pid and job; a box idle for 30 minutes a red against its owner in the steward's board (relayed to the node lane for build-3 and build-7 and the HEAL lane for build-6, the TV-02 lane for build-5, the build-server lane for build-2 shared with X5). THE SEED, HAND AND LIGHT-READER UP ON THE HUBS' CHAIN (the build-server lane, 09:49 UK): the seed (26631, pid 222321) and the hand (26671, pid 223533) since 09:49:10 UK on fresh datadirs with the 5d53a591 kit's igneumd and the three hubs as the only peers, the explorer indexer and the observer-dn4 reader restarted on the hand at 09:49:33 (the hand's EVM answering, the indexer indexing from 1050); the light-reader since 09:47:35 the same way, its 4.6 GB datadir aside, igneum-light-service untouched until the within-ten read; one slip once: the first start at 09:46 used ExecStart lines from a truncated unit file without --appdir, the hand crash-looping on the default datadir's lock for two minutes and the seed writing 337 MB into /home/build/.igneum (left aside), fixed from the backed-up lines at 09:49; the within-ten reader every minute under /run/dn4-restart.pid. MAIN'S DECISION ON THE HOUSE PCs (09:5x UK): rule 33 stands for every public entry; the house PCs are not the public channel's canary any more; the shipper and the relay lane stand up one Windows cloud VM today (Hetzner's Windows image, Azure or AWS, rented by API under the handover gate; CPU-only enough, the engine's CPU miner satisfying the five minutes mining with an accepted block on the moving chain, the shard proved on it or marked as the box's limit) with the relay agent as a service from the first minute, the standing Windows canary box for every future entry; its first rule-33 record by 13:30 UK; the 2.0.3 public entry on its PASS; the house PCs then taking 2.0.3 through the public channel's own apply (ending the old tree, installing the watchdog agent); until then the PCs stay as the founder left them, mining on 2.0.2 public; the AMD and Intel cells NOT RUN with the stated reason, no rig unblocked by a hand. THE STALE-SINK FIX READ ON THE MOVING CHAIN (the node lane, 09:53 UK): kit 6's kept datadir (lp-06's dead island 9b980b2f synced to its end at 19,666 / DAA 39,598, standing with zero relay lines) copied and restarted on build-1 under 27f54124's kit (igneumd 027bb12a) at 09:46:35 with the three hubs as its dial list: three "Sink probe" lines, "IBD started with peer 213.173.98.75:21169", "IBD completed successfully; bodies 6662 ms over 10433 blocks" at 09:50:17; the executed tip 24,852 / DAA 46,799 at 09:50:06 climbing onto the hubs' chain (lp-04 at 26,124 / 50,109); on a moving chain the standing-peer probe (bf60948a) is the stale-sink fix, no further flows change needed for the class; the read to the tip by about 10:12 with the fresh 27f54124 beside it (8,613 and climbing). X2's EQUIVALENCE (the worker lane, 09:52 UK, on fk-x2-4090, the worker built on the pod from 1a938abe4's proto-cuda/nvrtc, sha256 4152c960d1421007): the compacted path (--readback select) equals the uncompacted (--readback full) equals the CPU references byte for byte on all three same-work contexts (every nonce of [0, 3,145,728) at the all-ones target with the day switch by prepare; the 1,048,576-nonce seeded fuzz set on 1,024-nonce chunks with 365 to 599 overflow fallbacks per run and the 8,192-nonce boundary block; the sentinels: exactly 256 hits in a chunk, 257, target 0, the target equal to the job's smallest hash and that minus 1); known-failed first (the counter-reset fault run flagged with 368 stale lines; the driver's self-test flagging drop-257, a wrong digest, stale hits, a missed day switch, a < for <=); the normalized output sha256 per context identical on both paths (prod f278f059 / ad091a0d / 2aa97d5d; fuzz 5f1e591d / 8f39a1b1 / 3a7a7cc4); one finding X2-F1 (both paths, not the compaction): the same job sent twice in a row trips the worker's stale-output detector ("worker fault ... the kernel did not run", exit 3), so a miner or pool re-sending identical work restarts the worker; the meaning: compaction changes no verdict and no digest on the frozen class v6 and class v5 objects; the energy rows running (4 ABBA pairs plus the kernel timer).
 
+THE FOUNDER'S SECOND 1.5x DOCUMENT AND TRACK B (11:1x UK by main's clock, about 10:1x by the Mac's; relayed by main; the lanes opened by the coordinator at 10:1x): "IGNEUM - The 1.5x Research Programme" (the plan sha 269ceaa5824d0914, the package zip c873ac9da231fb26 with check_arithmetic.py and arithmetic_results.json, staged at the scratchpad's 1p5x/, the landing hand landing them beside the brief by 12:00): Track A stays the X matrix; Track B opens in parallel, fanned out: B1 (R15-05) the cryptography lane a4490b2dfe9114a55 on build-6: the exact Blocki and Smearsoll TCC 2025 MTP/DRSample construction (eprint 2025/1456, the FULL paper) as a CPU reference with its complete verifier, domain separation, known-answer fixtures and malicious-prover fixtures (the Dinur and Nadler MTP attacks the first adversary), 13:30; B2 the binding spec lane a717f5c9730229ea6 (documents): the canonical instance input (network, version, epoch, block-template context, trial id) and the reuse, grinding, chosen-instance, partial-evaluation and amortisation questions as proof obligations with the construction's answer or BLOCKED, 12:30; B3 (R15-06) the GPU feasibility lane a8e608abf1567ff87 on rented 5090 and 4090 pods: the exact function with representation and scheduling changes only (batched instances at 32, 128 and 512 MiB, coalesced labels, cooperative processing where permitted), state construction, labelling, commitment and proof creation measured, the under-utilisation trap stated, the first rows 14:00; B4 (R15-07) the theory lane a46974623d72b032c (documents): the composed theorem statement for a cheating prover producing accepted certificates with concrete constants, what the 2025 and the 2024 bandwidth results give and what is new, the reduction losses, R4's encoded-state warning answered, 13:00; B5 the adversary lane: the full-SRAM-on-fresh-state and the near-memory (PIM) designs added to X5's sweep inside its 12:30 line; B6 (R15-10) the node lane: the verification budget (proof bytes, cold CPU, invalid-proof amplification, pool share checks, propagation) as a parameter table against the fleet's node limits, 13:00. The rules for every Track B lane: the work unit never compared by raw hashes per second (an equivalent lottery-contribution and security-budget comparison approved before any number); the internal design goal 1.25x with the guardband stated; every chip figure labelled WITNESS (E_D) or BOUND (L), X5's rows witnesses; ASAP7 screening only; negative results controls; the quantum claim classical; Track C (DEGSample) not started until a Track B deficiency names it; the common brief carrying the rules in its Track B section.
+X1's CENSUS, TRACES AND HIT CURVES (the research lane, 09:5x to 10:0x UK): (a) the census on build-8 (09:49 to 09:56; the harness 1p5x-x1-harness 5264b8152, the frozen census jobs verbatim with the class +flat, 2^28 words): the control seed p18 reproducing the frozen row to the last digit (the instrument the frozen one); the candidate: the object (the genesis pairing, program caf198d6c4b4767b, 2^21 nonces over 8 headers on two lanes) top 0.1 percent 0.001942 = 1.0002x and 1.0003x FLAT (frozen 1.0998x and 1.1005x); the 16 seeds p18 to p33 0.9983x to 1.0018x flat (frozen 1.03x to 1.28x); the min site distinct ratio 0.99989 to 1.00010; the worst index bit 2.88 to 4.30 sigma (band 6); every item of the object's 16,777,216 touched; the top-K cache at 256 MiB 0.415 on every seed (frozen 0.42 to 0.48) and 0.3316 on the object (frozen 0.3540): the candidate's live-dataset access distribution the flat one to the Poisson floor on all 17 programs, the frozen class's window concentration gone and nothing new. (b) and (c) on build-9 (08:45 to 09:01 UTC; the harness fc6bfaf25, the signing pair at 4 GiB, 2^22 nonces, 2^21 train and 2^21 holdout, 256 reads per nonce, 2^29 holdout reads over 67,108,864 64-byte items): the sorted static access probabilities exact from the program's own load sites: the control 2a1d6caab4c24564 three probability levels, max/min 1.43, the hot-set mass 0.3125 / 0.5625 / 0.7812 at stored fractions 0.25 / 0.5 / 0.75; the candidate 34d0b421b17fe803 one level, 0.25 / 0.50 / 0.75 exactly; over 256 accepted chain-shaped seeds the control's hottest half min 0.531, mean 0.611, median 0.594, p90 0.688, max 0.781, the candidate 0.500 on all 256; the hit fraction at 0.25 / 0.5 / 0.75 on the holdout stream: the control's window-model static 0.3125 / 0.5625 / 0.7813, train-ranked static 0.2717 / 0.5240 / 0.7662, LRU 0.2506 / 0.4883 / 0.7064, the oracle in-sample 0.3775 / 0.6519 / 0.8640 (an upper bound); the candidate 0.2500 / 0.5000 / 0.7500 on both static policies, LRU 0.2457 / 0.4808 / 0.6996, the oracle 0.3657 / 0.6396 / 0.8563; the brief's conditional cache calculation re-done measured: at the half store the misses from 0.4375 (the control's windows, its best realisable static set) to 0.5000 (the candidate), plus 14.3 percent on the signing program, plus 28.5 percent at the 256-seed mean (0.389 to 0.500); the brief's 72 percent (plus 78.6 percent) near the 256-seed p95, not at the object. THE MEANING: removing the windows makes every static and LRU policy see the uniform stream, but the frozen object's own window concentration is smaller than the 72 percent the hybrid row assumed (56 percent on the signing program, 61 on average), so the hybrid's real edge on the frozen object is already near its uniform-store lower bound (the adversary's 2.30x against 2.69x) and X1 can take at most that gap: an activate-bound hybrid's rate at the half store falls 1/0.4375 to 1/0.5, minus 12.5 percent on the signing program, minus 22 percent at the seed mean (X5 owns a). The GPU rows: the 5090 (fk-x1-5090b, the host cap 400 W) ABBA pairs candidate against control rate minus 0.14 percent, joules plus 0.14 percent at 1 GiB, minus 0.10 / plus 0.10 at 4 GiB (g = 1.001), the self-test PASS on all 8; the second 5090 and the two 4090s by about 10:20. The X1 GPU kit (the kit lane, 09:54 UK): /srv/artefacts/packs/packs-x1-20261009T085304Z.zip sha256 88f5abf4036a035f5da89d0e7bade516614e9437fe6c2efd18668bd155b9b1c8 (1,127,251 B, 72 files; built on build-1 at the x86-64-v3 baseline by the class v6 kit recipe from branch x1-kit 55a2365d3; the four X1 packs (the candidate 0x34d0b421b17fe803 "+flat", the control 0x2a1d6caab4c24564, generator 6, log2 28 and 30), the Linux CUDA worker f5f3846c and OpenCL 371dbb27, the pack tools; the CPU reference's emulation check on the ds28 candidate PASS, 96 of 96 vector lanes).
+X0's DECIDER (the hash lane, 09:59 UK, one session on the rented 5090 fk-x0-5090c, Vast 54992886, driver 580.105.08, the gen-6 Linux worker 804a6f7f, 250 batches of 2^24, nvidia-smi 1 Hz, stock only, the provider refusing -lgc): class v4 v4-devnet-epoch0 (0xa785001687d8688a, the 7 October numerator's pack) 141.963 MH/s / 500.0 W / 3,522 nJ per hash / e370fb2080b7dbb1 (the card at its 500 W limit); the frozen class v6 hl-v6-all (0x9d40978601a7df2a, the house knee rows' pack) 70.900 / 445.1 / 6,278 / 59e6708e46f1e87c; the signing object hl-v6-all-cs (0x2a1d6caab4c24564, generator 6) 70.650 / 457.1 / 6,470 / 01f51b9d4805e5e6 (its first CUDA fingerprint, self-test PASS); the ratio class v6 over class v4 in joules per hash 1.78x (the kit pack) and 1.84x (the signing object), the rate halving (2.00x); the house rig's same pair 1.68x at stock and 1.73x at the 1,300 MHz knee, so across three sessions 1.7 to 1.8x: class v6 costs the 5090 about 1.75x class v4's joules per hash and half its rate, the 7 October numerator (2.33 microjoules at the lock) understating the frozen object's card cost by about 1.7x; the evidence build-1:/srv/artefacts/tas/x0-pins/x0-run-fk-5090c/ (run-x0.log 9bc40685, smi.csv 7ab9f16f, SHA256SUMS.txt); the row document and registry batch 5 to land. X0 AMENDED at 25fe02109 (the adversary lane, 10:03 UK): what the decider does to r0 turns on one fact asked of the hash lane by 10:30: whether the class v6 program carries about twice the instructions per hash (then the chip's shadow scales with the card's and r0 holds: the board 1.5x, the hybrid 1.93x, the die 2.44x same node) or the same instructions at a structural card cost (then every ratio rises 1.7x: the board 2.6x, the hybrid 3.3x, the die 4.2x); X0 carries the band with the first reading as the default (the connected-state lane measured the 64-register window at under 1 percent of the card's energy, so the structural reading has no measured support yet); X5's six designs on their boxes (build-8's launcher relaunched at 10:01, the container's missing work-root variable). X3's FIRST PAIRED ROW (the Ember lane; the rented 4090 at stock, board power, the frozen control via the kit zip 4 bb66a546 pack hl-v6-all, worker f5f3846c, the anchor fingerprint e8f4f3289c6ee1fc equal to the CPU on the 4090 and the 5090; on an island, not a network): the control (base, batch 2^22) 31.252 MH/s at 250.0 W = 7,999 nJ per hash; tuned lb4-w4 at batch 2^26 31.302 MH/s at 251.6 W = 8,038 nJ (rate plus 0.16 percent, energy plus 0.49 percent per hash: no gain); 0 rejected shares over 12 shared 2^28-nonce windows (204 finds equal); the 4090 screen of all 17 race variants x batch 2^20 to 2^26 flat within 0.1 percent (the kernel memory-latency bound at 2,775 MHz SM, 250 of 450 W), so the in-container knobs carry no g; the only Ember lever with energy in it the clock/power ladder (the recipe's 5090 lock row: minus 31 percent nJ at minus 0.3 percent rate), which rented containers refuse: those rows NOT RUN; the rows to 11:00, the best setting's 30-minute stability run to 11:30. X4's FIRST PAIRED GPU ROW (the k lane, 10:0x UK; fk-x4-4090 at stock, worker 804a6f7f, 250 x 2^24, power.draw at 1 Hz from 8 s in): all four texts passing the pack self-test (96 of 96 vector lanes) with fingerprint e8f4f3289c6ee1fc, the GPU equivalence holding; pass 1 against the frozen hl-v6-all (35.080 MH/s, 285.4 W, 8,134 nJ per hash, 93 regs): direct 35.021 (minus 0.17 percent), 7,839 nJ (minus 3.6 percent), 182 regs; generated 35.053 (minus 0.08), 7,824 nJ (minus 3.8), 142 regs; prefix 35.056 (minus 0.07), 7,905 nJ (minus 2.8), 154 regs; pass 1 alone possibly the card warming up, the reversed pass 2 (D C B A, the frozen pack last) ending about 10:08 deciding it, g the mean of both passes; the knee NOT RUN (the container refusing -lgc).
+THE STALE-SINK CLASS CLOSED ON THE MOVING CHAIN AND BUILD-1's OOM (the node lane, 10:05 UK): kit 6's stale-sink kept datadir under 27f54124 reached the moving tip at 09:54:37 UK (26,324 / DAA 50,430 against lp-04's 26,312 / 50,413 read in the same second), one capture (lp-04's chain's seed, no refusal, no strike, no lost proof), five sink probes in all: PASS against a moving reference, the standing-peer probe closing the stale-sink class, no further flows change; the fresh 27f54124 beside it reached DAA 12,657 by 09:52 and then build-1 ran out of memory (the hands' two fresh kit-2 syncs from 09:49, the light-reader and the node lane's two nodes took the box to 120 GB used / 5 GB free at 09:53; both the node lane's nodes and the reader gone by 09:55 by the OOM killer's shape, dmesg root's), the fresh read VOID on memory, the fleet lane's kit canary on build-8 the fresh read of 27f54124 on the moving chain; build-1 at 101/23 GB after; the node lane's rule: one read node of its own on build-1 at a time and none while the hands sync; the build-server lane asked for the dmesg read, the light-reader's state (restarted through its unit only when the seed and the hand are within ten), the memory rule in the queue file. The node lane's queue entries up since 10:03:42 (build-7-node-flows.pid: the 2.0.3 flows suites on 27f54124 continuous; build-3-consensus-fuzz.pid; build-9-pow-fuzz.pid; driver loops on the Mac, ssh orchestration only, every build and run on the box); the HEAL lane's build-6-exec-p2p-fuzz.pid up at 10:05 (the exec, p2p-flows and p2p-lib suites each pass from the foreign-seed-capture-2 worktree 0156e250, build-6 shared with X5's hyb75); the D02 lane's build-3 flagged idle at 09:53 (its model under a second, its next run the TV-01 reconcile at 13:00; the box free beside its short runs); the steward recorded the D02 batch at 000ec2b2 (TV-01, TV-03, TV-04 NOT RUN in progress), the hashed identity list fix 780df1d7 (a path entry's login only), TV-04 on the map 7ad28f75, the same-work recorder class f5a560b5 (one record per case, cell and run; an older run kept as evidence). THE WINDOWS CANARY BOX (10:0x UK): no provider rents a Windows VM by API from the Mac (Hetzner Cloud: Linux images only, the Server 2022 install ISO 8637 needing a console hand; no Azure or AWS CLI or credential); the relay lane's route, a console install being a manual input under the founder's rule: Windows Server 2022 (4 vCPU, 8 GB, 80 GB, CPU-only) built unattended under KVM in a container on build-7 (/dev/kvm, Microsoft's public evaluation ISO, an autounattend with OpenSSH and auto-logon, the virtio drivers, ssh through a forwarded port on the box; the relay id win-canary with its secret minted and the bundle baked with the watchdog agent, the workers intake key issued, the bootstrap and the relay-task canary written; the pid line /srv/queue/pids/build-7-winvm.pid 2504745 since 08:58:41Z); the clocks: the VM answering ssh by 12:30 UK, the agent registered and the first workers report by 13:15, the first rule-33 record (2.0.2, the shipper's three changes for a CPU box) by 15:00; the shipper's cpx31-with-ISO order withdrawn (its correction once), the fleet lane released from fk-win-canary; the founder's folder: #1836 and the PC 2 note withdrawn, #1860 "PC 1 and PC 2: nothing to do". THE MINI's HOLD (the shipper, 09:57 UK): nothing armed to restart the mini (its launchd jobs: the app's own, the supervisor which never reboots, mini-tip, mini-collector, the relay agent; the snapshot script's go file removed); every future canary's reboot leg NOT RUN "founder's hold" unless he says otherwise; the mini at 09:57:14: app 2.0.2, the node synced at DAA 50,580 with 7 peers (the feed 50,573), the card between work, uptime 12 minutes. THE RESET AT 09:56 UK (the fleet lane): wave two done at 09:54:57, 40 of 40, no errors; lp-4090-43's stage running; of 92 prepared boxes at 09:55, 8 synced=true, 11 past DAA 46,800, the median DAA 17,991 (a fresh 4090 syncing in about 15 minutes), the island count high (28 tips among 35 mining boxes) until the syncs land; the hubs lp-04 DAA 50,495 (84 peers), lp-05 50,507, lp-06 50,520 still its own fork (the 09:50 reset did nothing: the forwarded scp of the 5 GB snapshot lost its connection and the sha check refused to touch the box; retried); the hub-1 archive stream restarted at 09:55 (13,039 files, 12.7 GB, hub-1 free 6.9 GB). THE 1.5x BRIEF LANDED (the landing hand, 09:56 UK): box master 37ee2667140411cd (the brief under docs/plans/igneum-2.0-master/1p5x/ with its README carrying the owners and clocks and the permanent rule, the ten BST and the one rig-name replacements with both sha256s, the master README line, the token-value README's ratification line citing d3686955); thirty-seven landings through it; X0's document gating.
+
 A SHARED-DEVNET FACT FROM THE FLEET (not this lane's, with the shipper and the infra lane): the Hetzner live seed 188.245.5.161:26611 is still on the old override object (digest eada4bda) 1 h 40 min after the 0.3.20 sweep (the fleet never touches Hetzner nodes, so it was outside the sweep); the 0.3.21 wipe canary c22-1 took five digest-mismatch rejects from it; an app with the packaged peers is refused at the seed and syncs through node1 and the hub only, a fresh joiner with only the seed cannot join, the 14 voters and the hub are unaffected; the owner puts the floor file ov16-floor-900000.json (sha 294f1f80) and the c4459193 pin on it. 0.3.21's STAGING (the node lane): the order dry-merges onto 55768f88 with nothing moving to 0.3.22; the late-join fix is 52e96c94 (70e4601e rebased onto 55768f88, exec suite 33 green with both new tests); f067f7c1, b0444f51 and 437f0438 merge clean in order; 2e32d5f6's one conflict (DST_ADDRESS beside pool-finish's DST_BINDING in consensus/core/src/finality.rs) kept both; the live-file digest eada4bda after each (every switch at never); the staging waits on the shipper's sweep-end word; the re-pin held. PC 2 DOWN AGAIN (main, 16:5x UK): the founder takes PC 2 down for cable work (PC 1 back but his desk); both PCs out of the sweep's waves, each updates on its poller on return; no PC job to PC 1; the Windows G1 completed before the outage, nothing reruns. 0.3.21's SECOND GATE LINE on 55768f88 (sha256 279b1b690e854fc9): the ten-minute mixed-version gate beside the 5899f603 pair, 13:37:40Z to 13:47:52Z, SUMMARY PASS (one digest b0afb2ee on five nodes; 223 new and 381 old blocks accepted by the old hub, 0 rejected; counts equal at 319, 486 and 604 through both clean joins and the restart step at 13:45:22Z; no panic); the node lane's two lines on 0.3.21's first candidate complete, in plan 6.9 on ca3-v4-node; the fleet's set on it (the bare-child 12 GB line, the wipe, the kept read, the cases) is the fleet's. 0.3.21's FIRST GATE LINE on 55768f88 (sha256 279b1b690e854fc9, the string read back; pairing igneum-pow 8c728ca3 at byte 5): the digest gate 13:35:41Z to 13:37:19Z SUMMARY PASS (a89be8a7 on both binaries with the peers; db9a85f9 refused, no peer; the live file's eada4bda unmoved); the ten-minute mixed-version gate from 13:37:40Z, line about 13:50Z. The 0.3.21 order as the shipper sent it: 55768f88; f067f7c1 and 70e4601e; b0444f51; 6eb21fc9; db28d331; then the re-pin from 8bdcbdd8 on the coordinator's word; suites between, the digest read after every one; the mirror's release-0.3.20-node back at the pin c4459193, release-0.3.21-node open at 55768f88. THE LATE-JOIN COMMIT (N9's second half, the node lane): 70e4601e on the box mirror as branch proof-hold-fix, from c4459193, two files (igneum/exec/src/proving.rs, protocol/flows/src/v10/proving.rs); the gap was the fetch side on the joiner (the served record ran the native check against the joiner's trailing exec state before anything was stored, the check refused it, the proof was never held, the body rule read "not held" for 20 s and failed the IBD); the fix holds the proof by hash before the checks (the pool entry still needs them) and the serve side says when it holds fewer than asked; the exec suite 32 passed at 13:26Z with the known-failed shape first, the flows check green 13:28Z, igneumd on build-1 at the 0321 worktree path built 13:32Z, sha256 17649eeb2f7d1290, string read back; with the testnet lane (the resume form, B alone); it joins the 0.3.21 staging as its own commit. THE WIPE CANARY ON c19-1, c4459193 (sha 45be9b02d1b002f5, string read back): FORM END rc 0 at 13:50:53Z. Wipe synced 13:35:50Z (57 minutes, inside the 98-minute class); mining 13:36:00Z to 13:47:07Z, 66 mined, 66 accepted, 0 rejected, isSynced true at the tip throughout; the hub holds 41 of its blocks in its last 700 with 0 rejects (13:47:09Z); the restart on its kept datadir at 13:47:15Z: the old process stopped at once (the new process's first lock line seven seconds after the marker; the watchdog held nothing, the b7cc37e7 fault closed), synced again at 13:48:39Z after 84 s, 109 templates read with max 3,432 ms and 0 timeouts; the kept read on pool-1's 0.3.17 copy on the same pod passed at 13:38Z (the rewrite line once, a clean second start). The pin's set on c4459193: the digest gate PASS, the mixed-version gate PASS, the wipe canary PASS, the kept read PASS, the restart PASS, the 12 GB line proves and verifies (paid is a race, not a gate); CASES END from c20-1 (about 14:50Z) is the last pin line. THE INTEROP FACT stands from the void run: the 5899f603 hub accepted 235 object-byte-5 blocks from the 8097d600 node with 0 rejected, one digest on all five nodes on the live sixteen-field file. The gates: the digest test and the kaspa-pow vector test (the amended devnet epoch-0 id 1a4230699a6b9c60 must equal, c120d7963abdcd96 must differ, the v3 control unchanged) on the box; the mixed-version Devnet 2 gate (the amended 0.3.20 node beside a 5899f603 node for ten minutes on the live file without the v4 fields) after the Mac build; the fresh-join canary the 0.3.20 cut's |
 | Main's rulings (7 October, morning) | no generator change to v4 on the live devnet; the record's null is the window model with numbers, sent by the hash lane to the attack-pass lane so AP-F8-1 re-gates against it; a fault beyond the model (a low-entropy source at site 15) stops at the coordinator with the two options priced (a 0.3.19 class amendment before the flip, or the flip held at the floor), nothing shipping without the founder's word; the tighter tail, an acceptance bound on the hot-set share, is a CLASS V5 item (sent to the v5 lane a6410f3b8abefb762 with the 64-seed census as its gate; the bound's number follows from the model) |