Main, 7 October 2026: a 100 s hook gate on the merge commit against a master that moves every minute lost six pushes in a row.
Self-test: a fixture repository (unstamped branch, stamped branch, stale stamp, wrong tip, plain commit, dirty tree).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Main, 7 October 2026, 15:07 UK: one global exclusive measure flock across unrelated measurements stalled build-1 at load 120 with
free slots (a stopped probe held it 5.5 h; an exclusive waiter queued every new shared taker) and build-2 behind a one-core VDF
bench. lease.sh (installed at /srv/builds/_bin/lease by provision.sh and by hand on both boxes) takes one flock per core for a
pinned measurement and the quiet file for a whole-box one; remote-run.sh takes quiet shared only for unbounded runs, excludes
leased cores from its set, and its keeper refreshes the holder file and calls lease reap (a STOPPED holder of a lease, quiet or a
slot for 5 minutes is killed with a line in _log/reaped.log). Keepers close the lock descriptors they inherit (an orphaned sleep
held a slot and the worktree lock 20 s past the release; the slot self-test had rotted on that since the worktree lock landed).
tools/ci/box-locks-check.sh runs lease.sh --self-test and remote-run.sh --self-test-slots on build-1. provision.sh also carries
the 16 libraries headless Chromium needs (installed by hand on both boxes at 14:5x UK) and a headless self-test step.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
31 runs were queued on igneum-build-1's one runner at 13:15 UK on 7 October 2026 and nothing had concluded since 13:03Z, so no lane could read a conclusion.
- ci.yml: a `changes` job (ubuntu-latest) classifies the push with tools/ci/docs-only-check.sh (docs/, site/, *.md only = code=false; a new branch, a pull request, a force push or an API error = code=true); pow and sims need it and run only on code=true. The site job is unchanged on ubuntu-latest for every push. The classifier's self-test is in the gate.
- provision.sh and runner/register.sh: `--host <ip>` registers another box, forwarding BOX_HOSTNAME, RUNNER_NAME, RUNNER_LABELS, RUNNER_CPUS and RUNNER_JOBS (plain words only); RUNNER_CPUS writes AllowedCPUs into the service drop-in beside Nice=10, so igneum-build-2's runner is bounded like a suite (32 cores). The pool label is igneum-build-1 (both boxes carry it); ci-red marks the box with the record file and the poster, the default labels carry it, and igneum-build-1 got it through the runners API today.
- ci-red.yml runs on the ci-red label, so the red line always lands where the poster reads it.
- CLAUDE.md: the rule reads "read the conclusion when it lands, own a red before the next push"; pushes are never held.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The class (7 October 2026, 11:26 to 12:47 UK): three fork-gate summaries on ca3-v4-node carried the miners' vote-key hashes under "key" and eight CI runs went red on "no secret file names and no 64-hex secrets in the tree" while the pushing lanes saw nothing: the light gate ran only conflict markers and Windows paths.
- tools/ci/pre-push.sh: never_push_checks() (identity grep, no-secrets) runs on every ref from --hook, and inside the full gate where the identity grep already sat; the self-test asserts the wiring; the light gate is about 20 s on the Mac.
- infra/fast-time/lib/redact-keys.mjs: writeSummary() shortens every 64-hex value under a key-shaped field to 8 hex and an ellipsis and refuses a text the no-secrets rule would flag (line named); --self-test and --check; the gate runs the self-test. fork-gate.mjs adopts it on ca3-v4-node.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
the project lead, 7 October 2026 15:02 UK: build-1 at load 139 / 114 / 90 with both slots held and a 1 h 40 min queue while build-2 read 4.5 with
free slots, because the class router pinned each class to its box. Now lib.sh bs_route_spill reads the preferred box (free slots,
1-minute load) with one ssh and hands the job to the other box when the preferred one has no free slot or sits above load 64 and the
other qualifies; neither qualifying queues on the class's own box. The decision travels as BR_ROUTE_* into the JSONL "route" object
for the dashboard. build-2's slots file reads 3; everything on box 2 runs at nice 10 / 32 cores / -j 32, and a bounded run takes
the band its slot owns so three never share a core. Self-test tools/ci/route-spill-check.sh (thirteen cases) in the gate.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The box mirrors carry no infra/, so the test skipped there and the gaps showed only on the Mac and in the class v5 lane's run.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The dashboard lane, 7 October 2026 10:39Z: both slots of build-1 flock-held and EMPTY while two suites ran. Cause: a run from a
worktree without last night's append-mode fix opens a busy sibling's slot file with > on every probe. The holder now keeps its own
line: a keeper re-writes it within BR_KEEP_S (20 s) whenever the file is empty, until release; remote-run.sh --self-test-keeper
(in the gate) truncates a held line and sees it return, and sees nothing written after release; live on build-1 at 11:19Z (the
line came back in 25 s). The first version deadlocked the runner's bare wait with the keeper (build-2's first run hung 15 min
after its test passed): the keeper stops before the wait. The two running suites' -j 90 came from explicit --jobs 90: a bounded
class now clamps it to its cap with a log line (pass --priority gate for the full set). run-from-mac.sh --box N: the host file
was suffixed twice (build-server-2-2) and every box's mirror would have shared one remote name; one remote per box (build-N).
build-2's first green run: a suite at nice 10 on 32 cores, jobs 32, 986 s cold.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The build-server agent has igneum-build-2 and -3 on order (suites and benches; proving and the fast-time nodes). The
installer takes a host argument (N reads ~/.config/igneum/build-server-N, or build@<ip>); the pusher drops the Mac and
PC facts on every box it has a host file for, pulls each box's file and publishes boxes[] (box stays the first for the
old shape); the page draws one server section and one crew card per box, reads every box's Caddy feed in parallel
(build, build-2, build-3.igneum.network) with the edge copy filling any box that does not answer, and merges every
box's builds into the lanes, the timeline and the analytics. Nothing changes for build-1 until the host files exist.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Main's order of 7 October 2026 (igneum-build-2, AX162-1, and igneum-build-3, AX102-1, on order). lib.sh: bs_box_file N and
bs_route <class>; a class whose box has no host file yet falls back to box 1 and says so. run-from-mac.sh --box N <ip> provisions
igneum-build-N and writes build-server-N. tools/build-remote.sh --box N overrides the route; --priority gate always runs on box 1;
the proving crate routes to box 3. README.md: the kind map and the project lead's rule, no mining on any Hetzner box, ever (nodes, builds, tests,
benchmarks and CPU proving only; the pool's fast-time network mines on rented GPU pods, never on build-3). capacity/run.sh refuses a
job that would start igneum-miner mine or a GPU worker, whatever SEQUENCE says.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Main's order of 7 October 2026 after a load of 190 on 96 threads (a release join bench and the 0.3.19 app gate starving each
other, 'builds' of 16 minutes). tools/build-remote.sh resolves a class from the cargo subcommand and --priority: test and bench are
the bounded class (nice 10, the last 32 cores, -j 32) unless --priority gate (nice 0, the full set, the box's own jobs rule);
builds and checks are unchanged. remote-run.sh applies renice and taskset to the command's subshell, caps the jobs, lets a queued
gate (gate-pending-<pid>) take the next slot ahead of suites and benches, and prints nice and cores in the RESULT line and the
JSONL line (nice, cores, priority). The slot label carries '; kind=<k> nice=<n> cores=<c>' before '; agent=', so the dashboard's
job card shows why a job is slow. --plan prints the resolved class without the box; tools/ci/build-kind-default-check.sh (in the
gate) holds the five shapes. Smoke on the box: a suite at jobs=32 nice=10 cores=32, a gate at nice=0 cores=96.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
docs/design/genesis-forward.md names the three fields, their defaults, the digest change and the gates; spec 03 W1 and W5
amended and the W5 implementation row filled; spec 01 the cache rung beside the schedule; spec 04 the class-group VDF's
quantum fallback flagged, not sized; infra/fast-time/key-succession.mjs (3 nodes, one CPU miner each: the window fills,
a succession carried once on every node, refused twice, scheme 1 refused by every node; the known-failed case
--expect no-succession first); override-60x.json carries the four new keys at never.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
docs/plans/hands-on-build-1.md section 6: the move of 6 October 2026 23:06 to 23:22 UTC, observer node, observer, node 1, unload,
each hand's first executing line, digest eada4bda MATCH, IBD and acceptance, the open readback (the Mac's Miner app has not
started its own node since 26610/26611 were freed). lib.sh: the shipper's class from the 0.3.17 tree, the fork's igneum-exec
embeds proving/igneum-prove/elf/*.vk by include_bytes! five levels up, which is no path dependency; every .rs in the trees that
travel is scanned for include_bytes!/include_str! paths leaving the crate's repository and their directories join the overlay;
proving/igneum-prove/elf is the fixed fallback for a fork build. move-hand.sh: igneumd --version exits 1 (tolerated; it ended
the binary step before the override was written), and the launchd pid lookup used \s in macOS awk (printed 'pid none').
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Main's hold of 6 October 2026 (23:xx UK): the hands move only on the shipper's 'publish 2 live: digest <x>' line, each hand read back by
commit string and consensus digest. move-hand.sh binary now derives --node from the fork tree behind the Mac's node1 launchd agent
(igneum-wt-ship0315/vendor/igneum-node-0315 at f1ea7a38 tonight; the box's igneumd rebuilt from it is sha256 7f0bde70...), warns when the
Mac's override still has 13 fields (publish 2 writes the sixteen-field object when it restarts the hands), and after each hand's first
executing line prints the box binary's commit string and the hand's digest against the Mac hand's last digest (MATCH or DIFFER, DIFFER
stops the sequence). Dry run clean against the live Mac.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The fleet's 22:09 UK incident (a Mac-side pkill -f <log file name> matched nothing, the roll-everything script lived on and wiped a held box) and the day's two pgrep self-matches are one class. The check flags pgrep -f / pkill -f with a plain literal (every one on a line), any pgrep/pkill on a file-name shape, and ps | grep with a literal; it allows the bracket form, -x, -F pidfile, kill $(cat pidfile), a variable and a full path; 11 banned and 16 allowed shapes in its self-test; 0.15 s over the tree. The 25 pkill -f sp1-gpu-server inside bash -c bodies (which matched the calling bash) are pkill -x; the other 11 literals take the bracket form; prover-socket-check accepts both. Row R in the record; the CLAUDE.md rule names the check and covers pkill and file names.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The box's 34 red rows of 6 October classified (docs/analysis/ci-failures-2026-10-06.md section 6): 22 iterations, 12 in three real classes (instant deaths with nothing kept, a shared worktree directory, an unread dry run). remote-run.sh now: pre-flight (subcommand, manifest, -p package, --features) refuses in a second with exit 3 and a class; the last 400 lines of every run kept in /srv/builds/_log/runs; a class on every row (compile-error, link-error, test-failure, instant, no-test-matched, slot-timeout, no-dir, preflight-*); a cargo test whose filter matched no test exits 3; a per-worktree lock in checkout and run mode; every red row appended to /srv/ci-red/red.jsonl as source box. red-watch.mjs never posts a box row alone and sends one digest a day (counts per class with each class's guard); the timer runs tick. Shared group cired on the box so the runner and build append to one file. Shown in a sandbox on the box: pass, failing test, empty filter, bad package, bad feature, missing subcommand, compile error, broken manifest, two concurrent runs of one worktree.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- pow-fuzz: list saved mismatches from the directory, not an ls|node pipe (pipefail ran
the || echo too, giving invalid JSON [][]); drop the dead inner accumulation line
- sync-fuzz: merge the override with python, not node (node rounds the u64::MAX activation
fields to a float the Rust parser rejects); retention-period-days 2 (the 2-day minimum);
grep -c without || echo (pipefail doubled the count)
- clippy-audit: cap_cargo_t (timeout cannot run the cap_cargo shell function); grep -c fix
- all jobs sync the checkout unconditionally and lib.sh defaults the branch vars so a
standalone job or smoke never trips set -u on CAP_NODE_BRANCH
Smokes on igneum-build-1, all 0 panics: pow-fuzz 98 rounds / 19,600 programs / 78,400
units; sync-fuzz 142,883 requests, node alive, every kind disconnected or answered;
sim-sweeps 5 rows; model-sweeps 7 sections; clippy-audit 69 branches, 1,192 warnings,
1 error (ca2-coord), 0 advisories.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
pgrep -x git over the whole machine kept the lock whenever any git ran (the pre-push hook's own git push, another agent's build) and the checkout died with "index.lock: File exists". The fact is the lock's age. Class Q in docs/analysis/ci-failures-2026-10-06.md with the GIT_DIR leak of the previous commit.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
168 non-green runs since the first workflow run were classified (docs/analysis/ci-failures-2026-10-06.md): 102 were tree checks that finish in under 25 s on the pushing machine, 40 were GitHub-side refusals nobody saw.
tools/ci/pre-push.sh is the one list of fast checks; ci.yml's site job calls it with --ci and the pre-push hook with --hook (full gate for master and release-*, structural checks for other refs; never writes into the worktree). tools/ci/export-exclude.txt lists research documents outside the public export list, pruned by identity-check.sh and by the mirror's sync.sh (self-test: an excluded path may quote the patterns, an exported one may not); polish.md and this record are its first entries, which makes master green. tools/ci/windows-paths-check.sh (colon, trailing dot or space, reserved names, over 240 characters) runs as the pre-commit hook on staged paths and in the gate. tools/ci/red-watch.mjs plus the red job on the box's runner record one line per failed master or release-* run to /srv/ci-red/red.jsonl; igneum-ci-red.timer posts each once to the updates channel. pow and sims read IGNEUM_CI_RUNNER for the box. no-foreign-tree-writes.sh no longer exits silently on its warning pipeline under pipefail.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
infra/build-server/capacity: igneum-capacity.service (user build, Nice 19, SCHED_IDLE,
CPUQuota leaving 8 threads free) runs run.sh, a controller over a priority queue of jobs
that pauses (SIGSTOP) the instant a build slot or the measure hold is taken (5 s poll of
/srv/builds/_locks) and resumes after. Jobs, each with a dry-run and a 10-min smoke:
pow fuzz (continuous, seed base advances), sync-request fuzz against a throwaway pruned
node on loopback (the Horizon 28-unwrap gate, igneum-p2p-probe sync-fuzz), GHOSTDAG and
finality sweeps seeds 1 to 1,000, model.py sweeps cached, clippy+audit per recent branch.
Summaries to /srv/workers/capacity.json; the worker dashboard gains a Background lane
(collect.mjs doc.background, page renderBackground). Night battery stops and restarts the
layer. docs/plans/build-server.md section 8. igneum-pow fuzz tests and ghostdag_sim.py /
attacks.py gain a seed-base knob for the continuous sweeps.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
