388 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
jschoubben	3a486014fc	Prove the operator's answers end to end before anybody uses them (hq ADR 0259) ... The bus the controller composes and raises, at its commit, for one machine; the mesh's runtime serving the router and the Telegram channel on their own credentials; a fake Bot API as the phone. It links the operator's account with the code shown on the desk, refuses an approval in the hour after a link, performs an approval exactly once with who and how in the warrant and the router's record, refuses a replayed tap and a redelivered warrant, tells the operator when another account answers, drops a tap from a group, does nothing on Decline and on expiry, offers no approval while an agent can become root where the channel runs, and has the bus refuse every forged warrant, ask and tap. live-acceptance.sh is the same after rollout, with a drill the operator approves on the phone.	2026-10-09 11:08:42 +02:00
mesh-admin	d9173d49b8	Merge pull request 'Register R336: a send held for the bus's planned step is told to the operator' (#73) from replays/336-a-send-held-for-the-bus-step into main	2026-10-09 07:29:35 +00:00
jochen	fb5bb1e3de	Register R336: a send held for the bus's planned step is told to the operator ... hq issue 336 is a core issue and resolves with its replay. TestReplay336 in the controller, merged with #173 (b13a0f7), fails on its first parent and passes on it: prove R336 says PROVED.	2026-10-09 09:25:29 +02:00
mesh-admin	aaca73afb2	Merge pull request 'Register R335: a passed gate is not failed by the first machine's later reports (hq issue 335)' (#72) from replays/335-a-passed-gate-is-kept into main	2026-10-08 19:54:35 +00:00
jochen	06fa70d081	Register R335, so a passed gate failed by later reports fails before a core merge ... hq issue 335: a build that passed its gate was failed and put back when the first machine's later reports went quiet. Proved: fails on 8170fc5^1, passes on 8170fc5.	2026-10-08 21:52:19 +02:00
mesh-admin	5981f0ae85	Merge pull request 'Register R327: a verb that only reads opens no bus connection of its own (hq issue 327)' (#71) from fix/327-a-verb-reads-on-the-serving-connection into main	2026-10-08 19:38:52 +00:00
jochen	c9582ce041	Prove R327 against the merge of its fix	2026-10-08 21:33:28 +02:00
jochen	e99c2b22cb	Prove R327 against the fix rebased on the dead-letter merge	2026-10-08 21:32:25 +02:00
jochen	617f95f2bf	Prove R327 against the fix as reviewed	2026-10-08 21:32:25 +02:00
jochen	06e86ccd05	Register R327: a verb that only reads opens no bus connection of its own (hq issue 327)	2026-10-08 21:32:25 +02:00
mesh-admin	386e306642	Merge pull request 'Register R330: a message a consumer gave up on is kept (hq issue 330)' (#70) from fix/330-a-message-given-up-on-is-kept into main	2026-10-08 19:27:04 +00:00
jochen	f91164863b	Prove R330 against the merge of its fix	2026-10-08 21:15:36 +02:00
jochen	54db7c3458	Register R330: a message a consumer gave up on is kept (hq issue 330)	2026-10-08 18:43:43 +02:00
mesh-admin	96b05fbd85	Merge pull request 'Register R331: a tool check is asked on the link the node-engine holds (hq issue 331)' (#69) from replays/331-a-tool-check-on-the-live-link into main	2026-10-08 15:17:24 +00:00
jochen	3a3317b487	Register R331, so a tool check refused as no link fails before a core merge ... hq issue 331: every declared tool check read no link to the bus is open while the node-engine's link was up. Proved: fails on 824911d^1, passes on 824911d.	2026-10-08 17:11:04 +02:00
mesh-admin	24bc736393	Merge pull request 'Register R325: an assignment made while its module's build runs (hq issue 325)' (#68) from fix/assign-says-why-a-module-is-not-there into main	2026-10-08 14:58:10 +00:00
jochen	33e63ee5b4	Point R325 at #150's head after its second review ... The fix moved to 8adb7f1 on main fe00fec; proved again there.	2026-10-08 16:46:21 +02:00
jochen	5c128c1c18	Point R325 at #150's head after its rebase and review ... The fix moved to dd78357 on main 056414b; proved again there.	2026-10-08 16:22:30 +02:00
jochen	c02a8ec4c8	Register R325, so an assignment made while its build runs stays answered ... hq issue 325: assign said no module of that name a minute after the merge that asked for the module's build. Proved: fails on a759ac6, passes on b93dea8.	2026-10-08 15:50:18 +02:00
mesh-admin	983fd839eb	Merge pull request 'Register R318: a wait for a person's new login passes the gate (hq issue 318)' (#67) from fix/318-a-wait-for-a-person-is-not-a-failure into main	2026-10-08 13:03:08 +00:00
jochen	1455d9560e	Prove R318 on the fix's head, which gives the new condition its plain words (hq issue 318)	2026-10-08 14:37:35 +02:00
jochen	65ae6a9f89	Register R318: a wait for a person's new login passes the gate and stalls no walk (hq issue 318)	2026-10-08 13:42:33 +02:00
mesh-admin	596f0b4532	Merge pull request 'Register R314: an answer larger than one message of the bus is paged, not lost (hq issue 314)' (#66) from replays/314-a-large-answer into main	2026-10-08 11:09:58 +00:00
jochen	d452a47be0	Register R314: an answer larger than one message of the bus is paged, not lost (hq issue 314) ... Proved by the prover: fails on the controller's main before the fix (the caller times out), passes on mesh-controller 175b28e.	2026-10-08 13:07:43 +02:00
mesh-admin	99923bed12	Merge pull request 'Register the replay of issue 309 (hq ADR 0237)' (#64) from replays/309-an-order-rule-yields-to-a-stronger-one into main	2026-10-08 09:27:28 +00:00
jochen	434bac93ce	Register the replay of issue 309 (hq ADR 0237) ... Proved with cmd/prove: TestReplay309 fails on the controller before c5a2edf (the group refused as a cycle) and passes on it.	2026-10-08 11:11:54 +02:00
mesh-admin	1e371cd829	Merge pull request 'Register R310: a pull request is judged by the base branch's merge-check.sh (hq issue 310)' (#65) from replays/310-the-base-branchs-check-judges into main	2026-10-08 09:10:24 +00:00
jochen	157bb18fa8	Register R310: a pull request is judged by the base branch's merge-check.sh ... Proved: it fails on the commit before the controller's fix and passes on the fix (novox/hq issue 310).	2026-10-08 10:59:10 +02:00
mesh-admin	6fd0683425	Merge pull request 'The lab runs on the laptop again: the walk's builder on the bus, runtime images on node-tools, enumeration tests independent of the host's incus (hq issues 307, 308)' (#63) from fix/the-lab-runs-on-the-laptop into main	2026-10-08 08:34:39 +00:00
mesh-admin	193e669bf1	Merge pull request 'Register the replay of issue 305 (hq ADR 0237)' (#62) from replays/305-a-recheck-left-the-old-verdict-standing into main	2026-10-08 08:34:27 +00:00
mesh-admin	20bdf96b31	Merge pull request 'Prove a machine joins through the tunnel with the bus closed to it (hq ADR 0169)' (#61) from feat/a-machine-joins-through-the-tunnel into main	2026-10-08 08:24:17 +00:00
jochen	8368516253	The two-node walk's builder holds its seat over the bus, on a credential the mesh delivered ... It dialled the old broker's guest account, which the builder no longer reads and the mesh no longer runs. And the walk stops stocking a packet-filter runtime no container of the module names.	2026-10-08 10:15:59 +02:00
jochen	c76c64a4b5	Build a module's runtime image on node-tools, the runtime that replaced mesh-tools' npm package ... The tool runtime is a Go binary that launches each bundle through the launcher the builder writes; the repository root has no package.json any more, so the old script failed at its first step.	2026-10-08 10:15:59 +02:00
jochen	179f73c18f	Inject the incus command, so the enumeration tests fail incus on every machine ... The tests set MESH_LAB_INCUS in their body, which runs after the client module has read it, so they asked the real incus: green where none is installed, red on the workstation that runs the lab.	2026-10-08 10:15:59 +02:00
jochen	505db85539	Merge remote-tracking branch 'origin/main' into fix/the-lab-runs-on-the-laptop	2026-10-08 10:15:59 +02:00
jochen	09d347a045	Register the replay of issue 305, run in the catalogue module it lives in (hq ADR 0237) ... A catalogue module is a Go module of its own, so the prover runs a replay in the directory its register entry names. R305 fails on the commit before the fix and passes on it.	2026-10-08 10:11:57 +02:00
jochen	e80a4b1642	Prove a machine joins through the tunnel in a bed of its own ... Two machines: the anchor raises the foundation, joins over its own loopback and becomes the hub; the joiner makes its tunnel key, is issued a token for it and enrols with the bus's port closed to its own address, so the enrolment can arrive only over the tunnel (novox/hq ADR 0169). The check that had been added to the two-node walk moves here, closing the port the bundle publishes the bus on rather than the bus's own. And the uplink's range may be named: behind a VPN client that routes every private range, incus had none left to pick and no scenario could be raised.	2026-10-08 01:52:58 +02:00
mesh-admin	35babebbd1	Merge pull request 'Register the replays of issues 292 and 298 (hq ADR 0237)' (#60) from replays/292-298 into main	2026-10-07 23:41:44 +00:00
jschoubben	4b7ad9a387	The two-node bed stocks the packet filter's seat runtime ... The filter module now serves its verbs from a runtime the mesh builds (novox/hq ADR 0170), and a bed registered its raw manifest, which the mesh refuses as unbuilt. The bed stocks mesh-runtime-nftables and the filter helper registers the module through the stocked image.	2026-10-08 01:39:02 +02:00
jschoubben	8fdbf9ca90	The two-node bed waits for the anchor's first apply before its filter	2026-10-08 01:39:02 +02:00
jschoubben	db0069f262	The two-node bed joins its second machine through the tunnel ... A token carries the bus's address, and at genesis that is the anchor's loopback, which no other machine reaches. The anchor joins locally and becomes the hub; the laptop makes its tunnel key, is issued a token for it and joins over the tunnel (novox/hq ADR 0169, issue 146).	2026-10-08 01:39:02 +02:00
jschoubben	960539066e	The two-node bed places the bus's users as genesis must ... This bed raises genesis by hand, so it places the composed user list after each token and each enrolment, as the trust bed does (novox/hq issue 146); without it the first join was refused.	2026-10-08 01:39:02 +02:00
jschoubben	d2c6e8fc5a	The two-node bed proves a machine joins through the tunnel with the bus closed to it ... A third machine makes its tunnel key, is issued a token for it, and enrols while the anchor drops its packets to the bus at the first hook; it can only have arrived over the tunnel (novox/hq ADR 0169).	2026-10-08 01:39:02 +02:00
jochen	d0b0c7b02a	Register the replays of issues 292 and 298, each proved to fail before its fix and pass on it (hq ADR 0237)	2026-10-08 01:27:37 +02:00
mesh-admin	b75a17e51b	Merge pull request 'Check the Go replays in the Go toolchain, and register R301 and R302 (hq issues 301, 302)' (#59) from fix/the-replays-are-checked-in-go into main	2026-10-07 22:26:26 +00:00
jochen	c4f57432e1	Check the Go replays in the Go toolchain, and register R301 and R302 (hq issues 301, 302) ... The lab's check ran in the TypeScript toolchain, which holds no Go compiler, so the replays register was never compiled before a merge. merge-check.sh now declares replays/merge-check.sh as its Go part, which the build seat runs in the Go toolchain: gofmt, go vet and the register's own tests. Both new replays are proved: each fails on the commit before its fix and passes on it.	2026-10-08 00:13:46 +02:00
mesh-admin	a9c3bd8e6d	Merge pull request 'Register the replays of issues 296, 299 and 300 (hq ADR 0237)' (#58) from replays/296-299-300 into main	2026-10-07 21:55:20 +00:00
jochen	74c59661b1	Register the replays of issues 296, 299 and 300, each proved to fail before its fix and pass on it (hq ADR 0237)	2026-10-07 23:43:56 +02:00
mesh-admin	a7903562b4	Merge pull request 'Name the merges R145 and R-crashloop replay at, not branches that are gone' (#57) from fix/r145-names-its-merges into main	2026-10-07 16:54:27 +00:00
jochen	14246bcf01	Name the merges R145 and R-crashloop replay at, not branches that are gone	2026-10-07 18:49:58 +02:00