1380 Commits

Author	SHA1	Message	Date
admin	3decdf94a8	poster refresh 2026-10-21: export NOT taken in (6-file folder, 4 unasked text rewrites -> R-1201); OMV guide matches Ferenc's PDF (installer >= 1.33.0) ... Part A checks + label diff + PNG crops in documentation/audits/poster-refresh-2026-10-21/. Guide: HDD_PATH trouble item and "remote data area" sentence removed, first-days line added, operator note dated. Register 178 -> 179. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012qRErfCoiTkvDK9N5XHbzb	2026-10-21 10:00:54 +02:00
admin	19064ef1a7	register: R-1200 state word	2026-10-21 07:41:59 +02:00
admin	31d975c997	rehearsal 2026-10-20: B5 night (db/second drive/off-site PASS, data identical; whole-box not due → R-1200; host reboot = harness); due checks R-1019 closed, R-895 read, R-435 moved	2026-10-21 07:38:10 +02:00
admin	0d3f72e463	rehearsal 2026-10-20: journal — push held by due-checks gate	2026-10-21 02:58:05 +02:00
admin	c4262a6f6c	rehearsal 2026-10-20: night db copy OK; watcher logs recovered after an autostash froze them (harness slip)	2026-10-21 02:44:16 +02:00
admin	a600d68d50	rehearsal 2026-10-20: night log copier added before the night (read-only)	2026-10-20 23:39:50 +02:00
admin	830ed9714e	rehearsal 2026-10-20: report draft (night/morning parts pending); W26 marked fixed (unreleased)	2026-10-20 23:27:47 +02:00
admin	16188df64c	rehearsal 2026-10-20: 44 more screenshots off the browser container, gallery rebuilt; A4 browser spike result	2026-10-20 23:16:20 +02:00
admin	0d4bfe7ca6	rehearsal 2026-10-20: pre-night household snapshot (byte compare vs manifest), B7 plan written before	2026-10-20 23:09:29 +02:00
admin	938afb7416	rehearsal 2026-10-20: B10 ten stuck points; capability map A15 updated (still PARTIAL: R-1192); clock tool for #7/#8/#10	2026-10-20 22:41:46 +02:00
admin	c0b70b15ac	rehearsal 2026-10-20: post-night scenario plan written before any result	2026-10-20 22:24:21 +02:00
admin	ce8cf2c477	rehearsal 2026-10-20: B6 #1 README, journal to 20:14, B5 night expectations written before the night	2026-10-20 22:18:53 +02:00
admin	4a557f54a5	rehearsal 2026-10-20: B6 #1 power cut mid install, #6 off-site blocked, #14 kept data — evidence	2026-10-20 22:13:50 +02:00
admin	92de2d324e	Rehearsal: B6 #12 DNS dead an hour (as expected; household told nothing); R-1199 (P2 VERIFY) the household's ticked alarms reached only the operator (1 of 9 mails); R-1189 (b) demo-hp journal: three skipped nights	2026-10-20 21:50:47 +02:00
admin	12090df2d9	Rehearsal: R-1197 (Nextcloud 413 over ~1 GB by WebDAV), R-1198 (a later backup disk is never offered as the whole-box target); R-1194 facts	2026-10-20 20:50:36 +02:00
admin	434fc7cc02	Rehearsal: B6 #3 disk full (as expected: warned at 89 %, refused before writing, old copies kept, healed); #12 DNS dead started	2026-10-20 20:43:29 +02:00
admin	53d962d95b	Volunteer guide (HU+EN): the console's „linked" text replaces the stale „known bug: the code stays on screen" note; „Mentés most" takes ~2 min with eight apps (measured in the 2026-10-20 rehearsal)	2026-10-20 20:12:20 +02:00
admin	836b47f1ec	Rehearsal: B6 #16 English pass, #9 two-at-once (as expected), #13 second drive (not offered as backup target), #4 drive pulled (handled well; last uploads 0 bytes → R-1196 P3); B9 W24-W29	2026-10-20 20:07:30 +02:00
admin	bb187f3503	Rehearsal: B8 all seven restore paths pass (NFC/NFD: both names return); B6 #2 power cut mid-restore → R-1195 (P2, proven live: the app runs on the backup's data, the page says only interrupted); B6 #5 slow line; register 173 -> 174	2026-10-20 19:43:31 +02:00
admin	9db8aacdad	Rehearsal: R-1193 corrected and narrowed to P3 (the operator mailbox shows host_stale at 13:33 and app_oom; the household was told nothing); R-1192 second freeze + the app_oom hint; R-1194 (P3 VERIFY) fresh box's felhom-sshd config invalid; B6 #5 slow-line result	2026-10-20 19:10:42 +02:00
admin	78278502ac	Rehearsal 2026-10-20: second freeze and plug pull, memory watch, Mentés most, B6 #5 started; B9 words W7-W23; tools	2026-10-20 18:32:20 +02:00
admin	2228b9f6b4	Rehearsal: R-1192 (P1) the 8 GB minimum box froze ~1 h on the family's first afternoon, plug pulled to recover; R-1193 (P2) the hub showed it OK throughout; register 170 -> 172; STATUS P1 line and a note on D5	2026-10-20 16:38:27 +02:00
admin	ee743477b4	Rehearsal 2026-10-20 B2-B4: bound (operator passphrase), claimed, recovery code, family (accented names refused), second disk initialised, 8 of 9 apps (Jellyfin refused on memory), first uploads froze the box ~1 h (hub said OK); screenshots gallery	2026-10-20 16:34:30 +02:00
admin	ad6397308c	Rehearsal 2026-10-20: guide §3 quotes the console the box really shows; bind-page screenshots (codes never shown); app upload and gallery tools; R-1189 (d) (e) confirmed in code	2026-10-20 13:17:29 +02:00
admin	d750a90d0e	Data-loss hunt: R-1188 (c) proven — a leftover .felhom-restoring blocks every retry of the second drive's whole restore	2026-10-20 13:11:32 +02:00
admin	a942ae90d6	Data-loss hunt: R-1185 proven (an invalid dump replaces the good one, no error); R-1182 confirmed in code; DST replay of the controller scheduler (once each on 25 Oct; spring 02:30 dump collides with the 03:30 copy)	2026-10-20 13:07:39 +02:00
admin	20a77d367b	Data-loss hunt: R-1180 proven against production code (the nightly rsync --delete mirror runs while a restore holds the backup flag and deletes the files being restored)	2026-10-20 13:02:50 +02:00
admin	a57a8363df	Data-loss hunt: R-1191 (P1, proven live on 9202) — a .fab import makes a family-only app public (stranger 401 -> 200); R-1187 narrowed; register 169 -> 170; STATUS P1 line	2026-10-20 12:58:30 +02:00
admin	bffe2d9dea	Data-loss hunt 2026-10-20 C4: register 153 -> 169 (opened R-1175..R-1190: two P1 — shares second copy pruned by another drive's group, box rebuilt after the host delete cannot reach its off-site history; R-1177 proven LIVE on 9202 — a failed local restore destroys live data with no copy); STATUS: the P1 line and decisions D22, D23	2026-10-20 12:31:34 +02:00
admin	43e5a5496d	Data-loss hunt 2026-10-20: four reviewers' candidates, C2 verdicts, proofs (D1 shares second copy pruned, K1 escrow unreachable after host delete, R5 whole restore leaves app stopped while saying untouched, T9 hub jobs run twice on 25 Oct); rehearsal A5 manifest, B1 notes, B9 words, tools	2026-10-20 12:13:19 +02:00
admin	1d7071e8fd	Status view, part 3 (the gate): scripts/status_view_gate.py in repo_gates.py (fast: pre-push + CI) — size outside decide, one Updated paragraph, English outside quotes, cited rows open, no decided row in decide, the P1 and every operator-owned row on the page; WARN-FIRST until 2026-10-27, then it refuses; 39 decoy cases (red-proved per check and on both sides of the date); filed R-1174 (P4, Process & tooling) + its due check 2026-10-27; R-919 owner cell -> operator (the upload and the look); register 152 -> 153 ... Claude-Session: https://claude.ai/code/session_012qRErfCoiTkvDK9N5XHbzb	2026-10-20 11:59:48 +02:00
admin	9ec5f76a88	Status view, part 2 (the page): STATUS.md rebuilt from the register — 107 lines / 31,695 bytes -> one screen (50 lines, about 5,400 characters outside the decide table); decide table: 8 open decisions (13 decided rows out, rulings checked in CLOSED-ITEMS); do list checked against the hub and DooPlex (7 done items out, 7 'still needed? yes/no'); old page byte for byte in STATUS-HISTORY.md ... Claude-Session: https://claude.ai/code/session_012qRErfCoiTkvDK9N5XHbzb	2026-10-20 11:59:41 +02:00
admin	f84f94d987	Rehearsal 2026-10-20 (Tester4) + data-loss hunt: baselines, B1 install evidence, A6 drawn order, tools, first two reviewers' candidates ... No release, deploy or publish. Off-site is OFF for Tester4 and the owner passphrase is needed from the operator (bind blocked after 2 of 5 tries).	2026-10-20 11:57:55 +02:00
admin	2954b0be3a	Status view, part 1 (register pass): register 156 -> 152 (closed R-1121 file gone, R-1118 retirement done, R-231 in git + green run; R-993 merged into R-991); R-132 narrowed (rotation done, hash field left), R-1049 -> WATCHING (fix live since hub 0.151.0); hub + local read evidence ... Claude-Session: https://claude.ai/code/session_012qRErfCoiTkvDK9N5XHbzb	2026-10-20 11:43:58 +02:00
admin	5c487313fa	Night brief part 2 (run by day 2026-10-20): register 160 -> 156 (closed R-1001, R-1011, R-1167, R-1070, R-1010, R-893, R-1169; opened R-1171..R-1173; R-1019/R-462 narrowed, R-49 READY -> D21, R-1083 re-stated); 07 §6.3 option A as the design; STATUS morning list; report part 2; P4 Immich plan, P7 release evidence	2026-10-20 11:06:00 +02:00
admin	bcaa9576b2	Night 2 evidence: P2 R-893 option A live proof on 9202 (forced failure put back exactly; normal restore OK; teardown); P5 R-462 coverage and currency list	2026-10-20 10:47:37 +02:00
admin	301f4e9e92	Night 2 (2026-10-20) evidence: P1 hub 0.162.0 deploy, installer 1.35.0 publish, agent 0.161.0 release+delivery+host_busy live; P2 R-893 red-proofs; P3 R-1019 classifier before/after	2026-10-20 09:32:53 +02:00
admin	7d7868e5e5	Night 2026-10-19 Part 10: ep0's first weekly prune matched the dry runs (only tester-1's 10-04 copy removed; every household has a rule) -> R-1162 closed, register 161 -> 160; R-895 and R-435 due dates moved 10-20 -> 10-21 with the reason; P10 evidence	2026-10-20 06:00:58 +02:00
admin	189eb82eb0	Night 2026-10-19: register 161 -> 161 (closed R-921, R-953, R-986; opened R-1168..R-1170; R-1001/R-1011/R-1010/R-1070/R-1167 READY, R-893 READY with the measured design note -> D20, R-1083 gap); morning list in STATUS; 03/05/07/11 updated; R-893 evidence	2026-10-19 21:44:10 +02:00
admin	f92fa5a8c8	nodes.md, nightly-rotation.md: scratch guest 9202 takes its address by DHCP (.114 until 2026-10-19, now .121); read it from the guest	2026-10-19 21:30:39 +02:00
admin	15f7428871	security_scan_gate.py: govulncheck (called vulns) + gitleaks on the pushed commits in every repo's fast set and CI, warn-first, with an accepted-hits file and 8 decoys (R-986); first scan triaged: 0 real secrets, 0 called vulnerabilities; evidence for R-953 and the 0.335.0 release	2026-10-19 21:03:23 +02:00
admin	a28f162c66	Day 3 2026-10-19: passphrase locks once the box uses its own key (D19; hub 0.161.0, controller 0.334.0 floor), every ep0 household gets a prune job (tenantsync v1.3.0), demo-hp address clean; register 163 -> 161 (closed 3, opened R-1167)	2026-10-19 16:34:29 +02:00
admin	224fe41e06	Day 2 2026-10-19: ep0 keep-weekly 2 (D18), passphrase measured (R-1159 READY, D19), hub 0.160.0 deployed, controller 0.333.0 floor, gofmt gates; register 167 -> 163 (closed 6, opened R-1165, R-1166)	2026-10-19 12:20:37 +02:00
admin	e2c35007b5	Website git-sync in-cluster: R-1163 closed with evidence, R-1164 opened (P4); report	2026-10-19 11:51:54 +02:00
admin	fd75003e17	webpage: the four git-syncs read Gitea in-cluster; sidecars retry forever (--max-failures=-1); gate 6 keeps the --repo values identical ... On 2026-10-19 09:24 the public path to Gitea failed once and the sidecar exited (max-failures defaults to 0). Anonymous in-cluster fetch proven first (main and installer-v1.34.0 resolve to the public hashes).	2026-10-19 11:38:44 +02:00
admin	db2371c6f5	Day 2026-10-19: agent 0.160.0 (released, not vouched), controller 0.332.0 floor (one sign-in R-1100, escrow confirm R-1151), hub 0.159.0 deployed (R-1150, R-1154), R-1153 not a defect, old-restic check passed; register 168 -> 166 (closed 10, opened R-1155..R-1162)	2026-10-19 05:38:38 +02:00
admin	200ab6530b	hub 0.159.0 to the manifest (operator yes in advance, 2026-10-19); P1 delivery evidence (agent 0.160.0 on all three hosts)	2026-10-19 04:10:23 +02:00
admin	5622ce2735	Day 2026-10-19 P1 evidence: agent 0.160.0 release, demo-hp delivery, backup + cap, escrow confirm/abandon, R-1093 on the format route	2026-10-19 03:52:34 +02:00
admin	afc939bb3c	R-887 closed: 381 CI jobs since the S21 change, 0 lost (the 3 re-runs' first attempts ran 47-49 s; control: the known lost attempt ran 11 m 48 s); register 168 -> 167	2026-10-19 03:42:11 +02:00
admin	b505ce4ba0	hub v0.159.0 (R-1154, after security review): the enroll reuse path re-issues nothing, it warns the operator ... The first draft (5121536f, unpushed) re-issued the claim code and the off-site login on a reuse whose old box was stale or switched off. A security review flagged it: the caller holds only the household's retrieval passphrase, and both re-issues are operator session-only actions (05 section 17.5, 09 decision 252, R-1055/R-1059). Now the reuse path classifies the old box and re-issues nothing: gone -> host_reinstall_over_old_host (warning, mailed to the operator: delete the old host and re-run, or re-issue by hand); live -> host_reenroll_while_live (info). The mint path after the operator's delete is unchanged and still re-issues. The once-per-silence record and the claim engine comment edit are reverted. 05 section 8, README and CHANGELOG updated; tests red-proved.	2026-10-19 03:29:44 +02:00