| 2026-07-23 |
Delivery batch: lib-hosts VR1 arms, D-133 guards, appendix-A pf entry, DoD 8-9 (gauntlet 77 GREEN)
...
lib-hosts vr1-dc0/vr1-dc1 populated from the measured identity table
(tofu-name keys, pinned boot MACs, D-134 octets, per-DC power URIs +
tags) + host_sysid_by_bootmac (the VR1 resolver) + one-selection guard;
dc-selector rewritten, 48 PASS. reenroll-hosts + carve-host-interfaces
gain explicit vr1-* refusals (VR0 flows superseded by D-133 -- populated
arms must not enable them). lib-net vr1-dc0 VID caveat. Appendix-A:
edge pf-never-regenerated symptom (masking trap named). Phase-2 DoD:
items 8 (filter reload + LAN egress probe) + 9 (dc-mirror), item 4
range corrected to D-134 .201-.254. CURRENT-STATE same-commit.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VUjEb7onHpdqFHUnio6iVw
|
Mirrors installed both racks, syncs RUNNING; dc1 edge pf incident resolved; HOME fix (harness 19)
...
Item 9: fresh dc1 edge passed no LAN traffic -- pf ruleset never
regenerated after v4 addressing (set-interface reloads the interface,
not the filter; dc0 masked by 07-20 plugin work). configctl filter
reload -> egress 0%/301. Appendix-A + standup-DoD + script amendment
queued to stage close. Item 10: installs PASS 15/15 both racks, first
sync HOME-under-systemd failure fixed in the unit (T19), reinstalls +
syncs running (dc0 400M/dc1 343M in the first minute). Ledger in-flight
entry added; CURRENT-STATE same-commit.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VUjEb7onHpdqFHUnio6iVw
|
dc-mirror.sh shipped (D-135 item 1): per-DC apt+UCA mirror, harness 18/18, gauntlet 77 ALL GREEN
...
Site-keyed (dc0/dc1 MEASURED rows), check/install/sync; net layer = D-134
utility .4 alias + default route via the DC edge (egress path MEASURED
working; edge DNS answers). debmirror GPG-verified jammy triple + UCA
caracal, nginx on the utility address, daily timer, niced/idle-IO. Install
fails loud on missing packages with the exact apt line (net-first ordering
resolves the no-egress chicken-and-egg). Gauntlet capture + CURRENT-STATE
same-commit.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VUjEb7onHpdqFHUnio6iVw
|
Session bookend: queue-pass CLOSED (ledger summary, changelog archived, scan fix)
...
Bounded GA-R4 close entry; changelog-20260723-queue-pass.md -> docs/archive/
changelogs/; as-executed index row broadened to the window's full scope.
AT CLOSE: ledger-scan false-close defect (detached-amendment status hijack,
D-132 dropped) found by the close reconciliation itself -- fixed number-keyed
with regression cases (48/48; gauntlet 76 ALL GREEN), machine block re-seeded,
scan re-verified truthful. GA-R7 memory review: no changes.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VUjEb7onHpdqFHUnio6iVw
|
Queue pass: post-G12 queued findings delivered (5 script/harness fixes + DOCFIX-198/-199)
...
- opnsense-prep-image.sh: survives bash -s (BASH_SOURCE unbound fix; T3b)
- site-headend-install.sh: version/site-neutral hints; nftables-sec010
idempotent reload preamble; serial-log staging dir in --host-nodes (+check)
- opnsense-bootstrap-apikey.sh: sshd readiness wait + first-scp retry +
ssh -n stdin guard + minter-beside-script usage note (T9/T10)
- DOCFIX-198: phase2 runbook Steps 9-10 SUPERSEDED banners (pod route refuted,
D-103/D-123 amendments) + DC standup definition-of-done section (lands the
staging-dir + SEC-016 snap-refresh re-assert DoD obligations)
- DOCFIX-199: D-112(c) console-bootstrap payload requirements (util.inc et al)
Gauntlet ALL GREEN (76); repo-lint 0 fail. Per-item reverts in
docs/changelog-20260723-queue-pass.md. No cloud mutation this session;
queued: gated live re-assert of the nftables-sec010 files on all three hosts.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VUjEb7onHpdqFHUnio6iVw
|
G12 dc1: dc-rack-net.sh dc1 site-table arm + harness (D-131 sub-1)
...
Adds the dc1 arm to the standing per-DC rack-legs + node-DNS forwarder
pattern. MEASURED/apex-committed values, network-name keyed (no virbrN):
- vr1-dc1-metal-admin 10.12.68.2/22 (rack MAAS/DHCP; apex-committed rack IP)
- vr1-dc1-metal-admin 10.12.68.3/22 (forwarder listen alias; D-120 static)
- vr1-dc1-provider-public 10.12.64.2/22 (edge-LAN; measured live 2026-07-23)
- upstream region BIND 10.10.0.20 over transit
Harness T15-T18 added; dc-rack-net 18/18, gauntlet ALL GREEN (76),
repo-lint 0-fail. The forwarder pre-empts dc0's commissioning SERVFAIL.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VUjEb7onHpdqFHUnio6iVw
|
| 2026-07-22 |
G12 dc1: item-3 substrate landed + harness reconcile (gauntlet 76 ALL GREEN)
...
Lands the disconnected session's authored tree (vr1_dc1 vars, vvr1-dc1 +
uplink modules, inner root vr1-dc1-substrate/ with 54 pre-pinned MACs,
lib-net dc1 arm + one-selection guard) and delivers changelog item 4:
six harnesses reconciled to the ratified dc1 selector arm, phase-00
PLANES/lib-net parity guard (exit 2, refuses cross-DC planning),
rbd-mirror + radosgw parent-shell selection (dc1 reminder now shows
10.12.84.0/22, not dc0's). Gauntlet ALL GREEN 76
(docs/audit/gauntlet-20260722-g12-reconcile.txt); repo-lint 0-fail.
G12 next gated step unchanged: operator-gated apex --commit.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VUjEb7onHpdqFHUnio6iVw
|
| 2026-07-21 |
cloud-assert A9: vault-kv consumer AppRole auth probe (D-068 item 3 delivery)
...
Delegates to vault-kv-health.sh; skips gracefully pre-Stage-5 (no vault
app). Harness +3 cases (15/15); gauntlet 76 ALL GREEN.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019aLU6j24tjquxgYQaAHgsV
|
Stage-close batch: ledger-scan PARTIAL fix; DOCFIX-197; D-103/D-123 pod-refutation amendments; GA-R7 memory review
...
- scripts/ledger-scan.sh: PARTIALLY-RULED decisions now surface (D-131
blind spot); harness regression fixture, 46 checks ALL PASS
- DOCFIX-197: phase2 runbook Step 11 leg selection corrected to executed
reality (dc0<->dc1 leg; office1 leg carries live transit)
- D-103/D-123 amendments + modules/maas-vm-host header: 2026-07-20
measured pod refutation + per-machine-virsh ruling RECORDED (cited to
captures; no new ruling made)
- GA-R7 memory review DONE: no memory-only facts remain
- Read-only: stale MAAS pod object confirmed (id=4 vr1-dc0-inner, virsh)
for the pending retire-or-keep ruling
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013XRtm5sDQgUsZyTJk3k7dz
|
Step-E netem sudo: scoped NOPASSWD fragment shipped (operator-ruled)
...
scripts/sudoers.d/netem-tc grants exactly the netem-link module's two tc
verbs per measured mesh bridge (virbr5/7/3, re-measured this session;
drifting-ID caveat + fail-closed property documented). 9-case harness;
gauntlet 75 ALL GREEN. Install is operator-only (visudo-checked) and
pending; then the gated placeholder netem run closes G10 step E.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01STHXiHfoxHqq8fGRVvb66G
|
dc-rack-net.sh: D-131 sub-1 delivery -- rack legs + node-DNS forwarder, repo-carried
...
Site-keyed (dc0 rows measured 2026-07-21), runs on the rack via piped
ssh. Legs move from reboot-lost bare `ip addr` into a oneshot unit;
bridges resolved from libvirt NET NAMES at runtime (dc-planes does not
pin bridge names -- virbrN is a drifting ID); dns unit Requires the legs
unit. 14-case harness; gauntlet 74 ALL GREEN. Rack install stays gated.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01STHXiHfoxHqq8fGRVvb66G
|
| 2026-07-20 |
opentofu-validate: cover EXTRA roots (vr1-dc0-substrate, vr1-dc0-maas) -- second blind spot closed
...
The gate validated the outer root + every module and NOTHING ELSE, so both
applied roots shipped while it reported ALL GREEN over them -- the same
false-confidence failure the module loop exists to prevent. Roots are
DISCOVERED (any dir with a *.tf declaring terraform{}), not hardcoded, so a
future vr1-dc1-* root is covered the day it appears. Validated in a copy of
the WHOLE tree, because these roots reference ../modules by relative path
(copying the root alone gives 'Unreadable module directory' -- measured).
Gitignored *.auto.tfvars are stripped from the temp tree (may hold secrets).
VERIFIED the new check can FAIL: a deliberately broken root produced
[FAIL] vr1-dc0-maas before restoring. Harness 14/14, lint 0 fail.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KiUu1oqt76tWvV4vEC3NAr
|
NEW: maas-node-power.sh (+harness 24/24) -- per-machine virsh power, MAC-matched
...
Operator-ruled 2026-07-20 over the pod route. Sets power_type=virsh +
power_address/power_id on already-enlisted machines, matching domains BY MAC
(MAAS renames machines at enlistment, so name matching would map nothing or
the wrong node). Dry by default; every write verified by an actual
query-power-state, since a stored parameter proves nothing about reachability.
Carries the measured reasoning: pods fail on node-vm's pool+volume disks
(domblkinfo 'missing storage backend', reproduced locally), the pod's
discovery role was already done by PXE, and per-node power is the Roosevelt
(IPMI) shape. Gauntlet 72 harnesses ALL GREEN.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KiUu1oqt76tWvV4vEC3NAr
|
FIX false-success: opnsense-plugins.sh apply ALWAYS dry-ran (DRY_RUN:+ expands on 0)
...
Found on the first-ever live run (vr1-dc0 edge, 2026-07-20): ${DRY_RUN:+--dry-run}
expands whenever the var is set and NON-EMPTY, and DRY_RUN=0 is non-empty -- so
every apply silently dry-ran while printing 'OK: profile applied'. Nothing was
ever installed by this script. The harness missed it because every test drove
--dry-run, leaving the live path with zero coverage. Fix tests the VALUE; harness
gains two live-path tests via a stub api client (verified to FAIL 2/2 against the
buggy version before restoring). 20/20.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KiUu1oqt76tWvV4vEC3NAr
|
NEW: opnsense-set-interface-v4 pair (+harness 53/53) -- IPv4 + WAN gateway via the vendor config model
...
Operator-ruled mechanism 2026-07-20 ('Build the v4 setter pair'). Mirrors
the proven v6 sibling: vendor Config singleton (never a config.xml push),
dry-by-default, interface-count guard, fresh read-back verify, sh -s for
the tcsh trap, kernel ground truth. Adds the gateway half (at most ONE
gateway_item, count-guarded) and MEASURES the interface's device from the
edge's own config instead of hardcoding lan->vtnet0 like the v6 sibling.
D-113 amendment RE-MEASURED on 26.7: interfaces/settings/get carries only
offload flags; interfaces/overview/list and interfaces/<if>/get both 404 --
so the amendment holds on two majors. Gauntlet 71 harnesses ALL GREEN.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KiUu1oqt76tWvV4vEC3NAr
|
Deploy step B COMPLETE: rack bootstrapped + enrolled, SEC-010 CLOSED both ends (operator-ruled), OPNsense 26.7 base staged (operator-ruled); ledger-scan fail-open hardening
...
Bootstrap runs 1+2 (stale-lists 404 fixed by proxied apt update):
maas 3.7/stable, rack enrolled, libvirt+nested KVM, SEC-010 loaded
+ persistent on enp1s0; final --check EXIT 0 ALL items
(docs/audit/stepB-check-20260720-final.txt). voffice1-side SEC-010
pin applied per the row's follow-on clause. 26.7 nano prepped on the
rack (measured leaseweb mirror via region proxy); inner-root base
default -> 26.7. ledger-scan: FAIL-OPEN no longer counts as OPEN
(harness 45/45). CURRENT-STATE position/G10/G14 + section-5 updated
in this commit (GA-R1/C1). Open SEC count 8 -> 7.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KiUu1oqt76tWvV4vEC3NAr
|
| 2026-07-19 |
Batch 6: exit runs 1/2/4/5 -- archive exclusion (operator-ruled), adjudication, plan==triple, S2
...
6.1: docs/archive/ joins docs/audit/ as inventoried-not-counted (ruling
quoted in the exit record); record-audit + harness 35/35; per-group
adjudication = ZERO contradictory status-claim pairs on counted surfaces
(the charter's criterion as stated); group floor 9 recorded as re-run
baseline; missed patch file archived; CURRENT-STATE stale-note made
token-free. 6.2: 16 top-level docs (<25). 6.4: fresh capture
outer-plan-20260719-phase6-exit.txt = 6/0/6 == section-5 triple. 6.5:
G9 row names SEC-010's step-B dependency (S2). Exit record:
docs/audit/phase6-exit-runs-20260719.md. Items 3+6 pend operator
execution. G3 cell updated same commit.
Revert: git revert this commit.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VZbUeRE8weySLSGY1odrCi
|
Batch 4 C6: docs disposition -- 24 history docs archived, 16 retained, refs rewritten
...
DISPOSITION MANIFEST (archive, one-line justifications): clientdocs-
workflow-review (v1 tenant-docs era) / D-057-DECIDED-append +
D-057-REVIEW-ITEMS + README-D057-PACK + D-058-renumber (executed v1
decision packs) / D-068-openbao-assessment-DRAFT (supplementary; the
cited analysis retained) / DOCFIX-064-phase08-changelist + docfix-draft-
20260702 (executed/draft) / handoff-20260703 + handoff-20260705
(superseded handoffs) / incident-20260712-triplefault (trap routed to
platform-traps 1b/1c long ago) / netbox-vip-queue + phase-00-maas-
standup-notes + script-quality-findings + session-findings-2026-07-02 +
tenant-cidr-overlap-PLAN + v1-pre-deploy-fixes (v1-era executed/dated) /
repo-lint-nextfree-bug-FINDING (absorbed DOCFIX-105/107; repo_lint.py
pointer updated) / upstream-bug-draft-dashboard-tls (draft, dated) /
model-a-fallback-plan (D-123 Model B adopted; fallback preserved in
archive) / stage3-adversarial-review-20260716 (R3-F register,
dispositioned in 2.10) / dc-dc-replication-DR-seed (absorbed per
workflow) / dc-dc-ipv6-charm-research + dc-dc-netem-and-ula-gua-proposal
(stage 4+ research, refs updated). RETAINED 16 top-level, justified in
the session changelog. Live-surface refs to all moved files rewritten to
archive paths (18 files). G3 cell + ledger + changelog updated same
commit (the L10 fire on a Status-line path rewrite is satisfied by the
CURRENT-STATE touch -- the check worked as designed).
Revert: git revert this commit.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VZbUeRE8weySLSGY1odrCi
|
Batch 1 open: L10 status-coupling flipped WARN -> FAIL (GA-R1/C1)
...
One-constant flip (L10_FAIL = True) per the sweep plan's Batch 0 revert
note ("WARN for one batch before becoming FAIL"). Harness rc
expectations updated for the five firing cases; 42/42 PASS. G3 cell
updated same commit: Batch 0 CLOSED, Batch 1 OPENED by operator.
Revert: git revert this commit (restores WARN severity).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VZbUeRE8weySLSGY1odrCi
|
Batch 0.4: extractor status scan keyed to the legal token set (GA-F10 / vocab H1)
...
record-audit.py now knows the ratified Option A vocabulary: legal set
OPEN/BLOCKED/HELD/CLOSED (OPEN was previously unscanned -- a post-
migration blind spot where OPEN-vs-CLOSED stage contradictions would
have gone undetected); DONE/TO RUN/RUNNING added to the token scan.
New report section 6 classifies every distinct token (legal/illegal/
event) and flags illegal tokens plus event words (VOID/REOPENED/
SUPERSEDED) minted in status position (**Status: line or table row);
prose event words stay legitimate row-history vocabulary per H1.
Report-only: the 0/1/2 exit contract stays contradiction-driven, so
Batch 6.1 semantics are unchanged. Live sanity run (scratch --out, the
Phase-1 capture untouched): 12 distinct tokens, 163 flagged counted
occurrences -- Batch 2.5's migration work-list. G3 cell updated same
commit (GA-R1 rule 8).
Harness: tests/record-audit +9 checks, 33/33 PASS; repo-lint 0 fail.
Revert: git revert this commit (restores the 8-token scan and the
section-5-only report).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VZbUeRE8weySLSGY1odrCi
|
Batch 0.3: ledger-scan wrap-aware next-free exclusion (GA-F15)
...
The per-line "next-free"/"reproduc*" exclusion went blind when a quoted
next-free value wrapped its hyphenated token onto the following line
(happened twice live inside the 2026-07-18 audit -- CURRENT-STATE
section 9). An excluded line now suppresses itself AND the immediately
following line; the window is exactly one line and resets at every file
boundary. Chose this over assignment-site counting: DOCFIX/BUNDLEFIX
have no single register file to key assignment sites on (unlike D
headers). Live counters re-verified unchanged post-fix: D=130,
DOCFIX=197, BUNDLEFIX=052. CURRENT-STATE section 9 marked FIXED + G3
progress cell updated same commit (GA-R1 rule 8).
Harness: tests/ledger-scan +3 checks (wrapped quote excluded, wrapped
repro example excluded, two-lines-below still counts), 44/44 PASS;
repo-lint 0 fail.
Revert: git revert this commit (restores the per-line grep exclusion
and the section-9 "recorded, NOT fixed" wording).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VZbUeRE8weySLSGY1odrCi
|
Batch 0.2: ledger-scan SEC reporting repointed at security-ledger.md (GA-R4/F3)
...
Scan output now emits open COUNT + ID list + register-of-record pointer;
row-level dispositions dropped (they live in docs/security-ledger.md
ONLY). Session ledger's machine-derived SEC bullet re-seeded to pointer
+ count form. G14 evidence cell repointed at security-ledger.md same
commit, and its count corrected 9 -> 8 per GA-R1/C2 (measurement wins:
the cell's own enumeration and the scan both measure 8; divergence noted
in the cell, not silently harmonized).
Harness: tests/ledger-scan +4 checks (count+IDs surfaced, pointer
present, disposition text absent), 41/41 PASS; repo-lint 0 fail.
Revert: git revert this commit (restores per-row scan output, prior
ledger bullet, and the 2026-07-18 G14 cell incl. its count error).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VZbUeRE8weySLSGY1odrCi
|
Batch 0.1: repo-lint gains L10 status-coupling check (GA-R1 rule 8 / C1)
...
L10: a change set modifying state-bearing surfaces (docs/audit/ incl.
ga-rulings.md; design-decisions.md **Status: lines; session-ledger.md
gate rows) without touching docs/CURRENT-STATE.md is flagged -- pending
working-tree set when dirty, HEAD commit when clean. WARN this batch
(L10_FAIL=False); flips to FAIL at Batch 1 open per the sweep plan.
Divergence noted: the sweep plan's paraphrase listed "CURRENT-STATE gate
rows" (vacuous -- such an edit touches CURRENT-STATE by construction);
implemented per the ratified C1 utterance, "ledger gate rows"
(docs/audit/ga-rulings.md:129). G3 evidence cell updated same commit
per the rule this check enforces.
Harness: tests/repo-lint T34-T41 added, 42/42 PASS; live lint 0 fail,
1 legacy WARN (unchanged).
Revert: git revert this commit (check + tests self-contained; restores
the prior G3 cell).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VZbUeRE8weySLSGY1odrCi
|
| 2026-07-18 |
GA Phase 1: deterministic record inventory + measured evidence (freeze upheld)
...
- scripts/record-audit.py + tests/record-audit/ (24/24; gauntlet 69 GREEN):
doc census, status-claim extractor, reference integrity. Exit 0 == zero
contradiction groups (the Phase-6 re-run criterion).
- docs/audit/record-inventory.md: 179 files, 858 claim rows, 12 contradiction
groups (plan counts, version pins, stage statuses), reference-rot counts.
- docs/audit/outer-plan-20260718.txt: captured outer plan -- 7 add / 2 change
/ 7 destroy; voffice1 seed volume forces replacement. THE citable
plan-count source from here forward.
- docs/audit/env-snapshot-20260718.md: measured host/guest/toolchain state
(vcloud, -136 bookend done, autostart enable, tofu 1.12.4, MAAS 3.7.2,
LXD 5.21.5).
- docs/audit/grounding-audit-20260718.md: findings register GA-F01..GA-F14
incl. the operator provenance rulings on the two memory files and the
07-16 06:41 churn-mechanism correlation.
No fixes applied; no existing file touched. Findings only, per the charter
freeze. Phase 2 (CURRENT-STATE.md rebuild) starts from this evidence base.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018ZV3uqyU9oWoFTcwiE4a21
|
D-129: build the OPNsense edge plugin profile-installer + sharpen pins
...
Autonomous processing of D-129 items that need no operator keyboard. Repo-only;
NOTHING applied to the cloud (live plugin install stays classifier-gated).
- NEW scripts/opnsense-plugins.sh + tests/opnsense-plugins/ (17/17): named-profile
installer over the D-113 firmware API. Installs ONE plugin at a time + polls
upgradestatus (the firmware-lock lesson), delegates auth to opnsense-api.sh,
warns os-qemu-guest-agent != a working agent, offline --dry-run. Closes the
"build the reusable installer" pin.
- D-129 + dc-dc-phase2 pin: installer recorded as BUILT (use it, not build it);
qga channel logged as an OPT-IN-var design (default OFF -> no office1/DC0-plan
side-effect; dmacvicar 0.9.8 mechanism-to-verify); os-git-backup operator ask.
- Gauntlet ALL GREEN (68), repo-lint 0-fail.
Revert: git rm the script+harness; revert the doc/runbook edits.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01JUBUyinU4sBAT4HV1b9vSz
|
| 2026-07-17 |
D-126 base-leg oneshot (site-baseleg.sh) + D-128 operating model; post-restart sweep
...
vcloud post-restart health sweep all-green (host on 6.8.0-136, patch+reboot done;
libvirt/pools/autostart clean; Office1 stack self-recovered end-to-end). Sweep
surfaced a recurring break: office1-local is an ISOLATED libvirt net, so the host's
base L3 leg (10.10.0.10/24 on the bridge + compose route 10.10.1.0/24 via 10.10.0.20)
lives nowhere in the XML and drops on every reboot -- the un-persisted foundation
D-126 assumed but never owned.
- NEW scripts/site-baseleg.sh + tests/site-baseleg/ (24/24): root Type=oneshot unit,
After=libvirtd (+virtnetworkd for modular hosts; ordering MEASURED here = monolithic
libvirtd), re-adds the leg at boot. Bridge DISCOVERED from the stable net name (never
a baked virbrN, PATTERN-1); apply idempotent + root-guarded; check = session-verify.
INSTALLED + enabled this session; boot-race UNPROVEN until an actual reboot.
- docs/design-decisions.md: D-126 amendment (base-leg durability) + NEW D-128 (operating
model -- Claude on the vcloud jumphost; Plane-1 substrate on vcloud / Plane-2 executes
on voffice1; workstation tailnet is the human path; D-107 untouched).
- Skill invariant + routing row (SKILL.md), Cross-cutting bullet (dc-dc-deployment-workflow.md),
session-ledger + changelog.
Gauntlet ALL GREEN (67 harnesses), repo-lint 0-fail. Nothing merged to main (stage-close).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QMKdXCquMnh1jKH2sWXPMg
|
| 2026-07-16 |
D-126: durable rootless agent access to VR1 site VMs (shell + forward) + DC creds prep
...
Establishes durable, rootless jumphost->site-VM access so the recurring
sudo-route friction stops (and generalizes to the DCs). Operator ruled Option A
(SSH via systemd --user / ssh_config ProxyJump), advisor-conferred.
Shell layer (FOUNDATION, EXERCISED): scripts/site-ssh-config.sh generates
~/.ssh/config.d/vr1-sites ProxyJump aliases (voffice1, office1-netbox,
office1-tailscale, office1-opnsense). All four aliases proven live from the
agent's session after the operator installed with the one-key office1 env key --
key read from IdentityFile, no creds path in any command (clears the guard hook).
Forward layer (convenience, reviewed-but-unexercised): scripts/site-forward.sh
holds a systemd --user SSH local-forward -> http://127.0.0.1:8000 for office1-netbox.
Secrets boundary: the key is operator-supplied via --key (a PATH, never read/baked);
generated fragments name that path and are written locally (not committed).
DC prep: per-env-key isolation (operator ruling) -- each DC gets a DEDICATED key in
its own ~/vr1-dc<N>-creds/ folder, minted at deploy. Folders pre-created (0700);
creds-manifests/vr1-dc{0,1}.manifest declare the keypairs. DC ssh aliases deferred
until their IPs are measured post-apply.
Also: D-124 dc0 rack tfvars written locally (opentofu/d124-rack.auto.tfvars,
gitignored) -- the independent `tofu plan` unblocker. vr1_dc1_rack_* NOT written
(no ruled dc1 addressing; variables absent) -- STOP flagged for an operator ruling.
Tests: site-forward 38/38, site-ssh-config 30/30; repo-lint 0-fail (1 legacy WARN).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ck6xh3jWQi5b3Su8Dx1LEH
|
D-125: Model B per-DC ISP egress (bridge-in, single-NAT) -- resolves OBS-3 design gap
...
OBS-3: Model B nesting put vr1-dc0-wan (a libvirt NAT) inside vvr1-dc0, whose
only leg is the SEC-010 FORWARD-dropped transit -- so the per-DC simulated ISP
had no egress. Operator ruled bridge-in (single-NAT) over double-NAT.
- modules/wan-bridge (NEW): libvirt forward=bridge onto a pre-existing host
bridge (bridge={name}, schema measured from the provider, not guessed). No
subnet/NAT. Sibling to site-wan so every NAT user stays untouched. Covered by
opentofu-validate S3 (standalone PASS).
- OUTER root: vcloud ISP NAT vr1_dc0_uplink (the single NAT) + a 2nd IP-less
uplink NIC on vvr1-dc0 + the br-vr1-dc0-wan bridge declared in netplan (transit
stays NIC1 so SEC-010/--transit-if keep keying on it). New HELD var
vr1_dc0_uplink_cidr (no default; office1-netbox apex, hard rule 2).
- INNER root: vr1_dc0_wan site-wan(NAT) -> wan-bridge onto br-vr1-dc0-wan;
OPNsense WAN re-addresses into the vcloud ISP /24 (edge config, D-113, HELD).
- Bootstrap (site-headend-install.sh): --uplink-if/--wan-bridge; --check verifies
the bridge exists + uplink ENSLAVED (fail-open guard, same class as the transit
check); SEC-010 rationale corrected (bridge-in has no inner NAT) + br_netfilter
CONSTRAINT (keep the FORWARD-drop interface-scoped, NEVER global). Harness 49->54.
Records: D-125 (new); D-122 amendment (forward-pointer -- WAN realization
superseded, intent stands); SEC-010 (br_netfilter + hardened --check);
model-a-fallback B->A delta item 8 + revert; platform-traps 1g (bridge +
br_netfilter + nested-foreign-MAC trap).
NOT closed: the L2-NAT-of-foreign-MAC path is a DEPLOY-TIME gate (bridge-in
equivalent of the depth-4 boot gate); fallback = double-NAT. HELD: vr1_dc0_uplink_cidr
+ OPNsense WAN re-address. Both roots validate; gauntlet ALL GREEN (63); repo-lint
0 fail. Present-only; NOT pushed.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ck6xh3jWQi5b3Su8Dx1LEH
|
Stage-3 review sweep C+D: Model B two-root reshape + SEC-010 hardening
...
Phase C (Model B reshape, D-123 operator ruling):
- outer root (opentofu/main.tf): vvr1-dc0 resized to whole-DC containment VM
(108 vCPU / 425984 MiB / 3221225472000 bytes; expose_nested_virt=true;
single Office1 transit leg); node/plane/wan/opnsense modules removed from
vcloud level (MOVED markers), storage-04 added per R-3.
- inner root (opentofu/vr1-dc0-substrate/): NEW second root, provider
qemu+ssh to vvr1-dc0; reuses ../modules/* verbatim to build 9 nodes +
6 planes + wan + opnsense inside the containment VM. dmacvicar/libvirt
0.9.8 pinned (.terraform.lock.hcl committed). Both roots tofu-validate.
- variables.tf: vvr1_dc0_vcpu/memory_mib/disk_bytes with derivation comment.
Phase D (SEC-010, R3-F02): site-headend-install.sh --host-nodes node-host
bootstrap installs qemu-kvm+libvirt, persists+verifies nested=1, writes a
transit FORWARD-drop (nftables inet sec010) enforcing DC-LOCAL under Model B.
Gate hardened (advisor catch): --check now verifies the keyed transit
interface EXISTS, else nftables oifname loads clean but matches nothing
(fail-open). Harness 49/49.
Records: D-103/107/114/121/122/123/124 amendments (append-only);
model-a-fallback-plan synced; ledgers + Phase-C changelog.
KNOWN GAP (OBS-3, operator-confirmed 2026-07-16): under Model B the inner
vr1-dc0-wan NAT is nested inside vvr1-dc0 (single transit NIC), severing the
simulated-ISP egress that Model A had at vcloud level. DC0/DC1 ARE meant to
have ISP connections -- egress path resolution is follow-up design work, not
resolved by this commit.
Present-only; not pushed. Gauntlet ALL GREEN (63); repo-lint 0 fail.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ck6xh3jWQi5b3Su8Dx1LEH
|
Review sweep A+B + R-3 storage-04: coherent record, whole-host calculator, R-3-compliant Model A fallback
...
Phase A (record amendments) + Phase B (scripts/dc-dc-whole-host-budget.py + harness) of the
Stage-3 adversarial-review sweep, plus the R-3 storage-04 addition (Model A layout now
3+2+4 = 9 nodes/DC). This commit is the R-3-COMPLIANT Model A state and becomes the re-cut
`model-a-fallback` anchor (the prior tag at 87a7a8a was R-3-stale -- 3 storage -- flagged by
the Model B design cross-check). NO cloud mutation; nothing applied.
- design-decisions.md: D-103/D-107/D-114/D-115/D-121/D-122/D-123/D-124 amendments (rulings R-1..R-4).
- scripts/dc-dc-whole-host-budget.py + tests/ (13/13): replaces the lost optc-calc.py (R3-F06).
- opentofu/main.tf: locals.vr1_dc0_nodes += storage-04 (R-3); tofu validate Success.
- security-ledger SEC-011 closed; session-ledger rulings + sweep progress; model-a-fallback plan.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ck6xh3jWQi5b3Su8Dx1LEH
|
