jochen
eaae0e2b80
Merge remote-tracking branch 'origin/main' into issues/209-a-bundles-own-sdk-copy-registers-nowhere
2026-10-03 13:20:33 +02:00
mesh-admin
2ee6eab7b9
Merge pull request 'Issue 210: the host re-creates the node's runtime on every reconcile, restarting it every ten minutes' ( #317 ) from issues/210-the-host-re-creates-the-runtime-every-cycle into main
2026-10-03 11:20:27 +00:00
jochen
ce5f85f65e
Design 38 WP4 built and proven live on all four machines; issue 209 proven live
2026-10-03 13:20:12 +02:00
jochen
911125148c
Issue 210: the record of the runtime's process carries no digest; the credential and the unit text are ruled out
2026-10-03 13:13:25 +02:00
jochen
d718a917e5
Issue 210: the host re-creates the node's runtime on every reconcile, restarting it every ten minutes
2026-10-03 13:11:17 +02:00
jochen
a572868333
Issue 209 resolved by mesh-tools #32
2026-10-03 13:07:41 +02:00
jochen
55443b67e6
Design 38 WP4: what the review of mesh-catalog #239 found — the credential goes, iptables is declared, escalation is an unchecked machine fact
2026-10-03 13:00:35 +02:00
jochen
89a202f12e
Issue 209: a bundle's own SDK copy registers into a registry the runtime never reads; design 38 WP4 note
...
Found preparing WP4, before the first module's bundle was loaded beside the runtime's own:
proven with a two-copies probe, located in mesh-tools (node-tools), fixed by mesh-tools #32 .
Design 38 WP4 records it and the two things the package left to the module — escalation
through sudo, and the filter file read from the manifest's path rather than a container env.
2026-10-03 12:46:51 +02:00
mesh-admin
b342c9c3da
Merge pull request 'Issues 203, 206 and 207 resolved by mesh-controller #233 ; 207's diagnosis, with its two questions left open' ( #315 ) from issues/203-206-207-resolved into main
2026-10-03 09:50:56 +00:00
jochen
336b8c9b62
Issues 203, 206 and 207 resolved by mesh-controller #233 ; 207's diagnosis, with its two questions left open
2026-10-03 11:50:20 +02:00
mesh-admin
0c82909845
Merge pull request 'Issues 204 and 205 resolved: mesh-controller #232 and mesh-host #79 merged' ( #314 ) from issues/204-205-resolved into main
2026-10-03 09:46:22 +00:00
jochen
d8a0c1b02f
Issues 204 and 205 resolved: mesh-controller #232 and mesh-host #79 merged
2026-10-03 11:45:19 +02:00
mesh-admin
b232b44f4c
Merge pull request 'Design 18 and ADR 0190: shared build work proven live on all four machines, and the five gaps the switch found' ( #313 ) from design/18-shared-build-work-proven-live into main
2026-10-03 09:44:19 +00:00
jochen
c03f2cd4c6
Design 18 and ADR 0190: shared build work proven live on all four machines, and the five gaps the switch found
2026-10-03 11:43:53 +02:00
mesh-admin
73c4d24024
Merge pull request 'Issues 203–206 diagnosed and located' ( #310 ) from issues/203-206-diagnosed into main
2026-10-03 09:43:00 +00:00
mesh-admin
3372d72da0
Merge pull request 'Issue 208: a seat's worker is made only when the controller starts, so a holder assigned later finds none' ( #312 ) from issues/208-a-holder-assigned-later-finds-no-worker into main
2026-10-03 09:33:28 +00:00
jochen
273b932329
Issue 208: a seat's worker is made only when the controller starts, so a holder assigned later finds none
2026-10-03 11:33:08 +02:00
mesh-admin
f91a3efb6b
Merge pull request 'Issue 207: a re-made worker replayed every ask the stream kept, and the mesh re-registered its past' ( #311 ) from issues/207-a-re-made-worker-replayed-history into main
2026-10-03 09:06:17 +00:00
jochen
2f0ce4ce19
Issue 207: a re-made worker replayed every ask the stream kept, and the mesh re-registered its past
2026-10-03 11:05:40 +02:00
jochen
bf3338898e
Issue 204: diagnosed and located — a declaration was numbered after composing, and a dying sender recorded nothing
2026-10-03 04:03:51 +02:00
jochen
4891cdeac5
Issues 203, 205, 206: diagnosed and located
2026-10-03 03:51:41 +02:00
mesh-admin
733cff4c3c
Merge pull request 'Issue 206: a seat's worker changing type strands its holder, and the build that would fix it' ( #309 ) from issues/206-a-worker-changing-type-under-its-holder into main
2026-10-03 01:46:12 +00:00
jochen
561f26fb34
Issue 206: a seat's worker changing type strands its holder, and the build that would fix it
2026-10-03 03:45:33 +02:00
mesh-admin
d8a58dadcf
Merge pull request 'ADR 0190: a seat's work is shared by its holders, and building is the first such role; design 18 says where a build runs' ( #306 ) from decision/0190-a-seats-work-is-shared-by-its-holders into main
2026-10-03 00:44:27 +00:00
mesh-admin
3f4782cfb4
Merge pull request 'Design 38: WP3 proven live on all four machines; the three issues it found' ( #308 ) from design/38-wp3-proven-live into main
2026-10-03 00:41:47 +00:00
jochen
d076647b5d
Design 38: WP3 proven live on all four machines; the three issues it found
2026-10-03 02:41:24 +02:00
mesh-admin
ae83c5e09b
Merge pull request 'Issues 203–205: what the first node-tools rollout found' ( #307 ) from issues/203-205-what-the-first-node-tools-rollout-found into main
2026-10-03 00:41:07 +00:00
jochen
feeea127e2
Issues 203–205: what the first node-tools rollout found — an unissued credential pushed, a stale declaration during a controller handover, a package resource against a stale database
2026-10-02 23:47:12 +02:00
jochen
76b563e0ce
ADR 0190: the ack-wait race is issue 175
2026-10-02 22:29:53 +02:00
jochen
f56686d1e5
ADR 0190: a seat's work is shared by its holders, and building is the first such role; design 18 says where a build runs
...
The build role was a mesh seat with one holder and its worker a push consumer with one delivery in
flight, so thirty-five images rebuilt serially on one machine while three others idled. The bus was
drawn for the alternative — a seat's accept subjects on a queue group of holders (design 25) — and
0190 uses it as one pattern for every role: a node seat with accepts, every holder pulling one ask
when idle, the asker addressing the role. Building is the first use: node-build-agent, held by the
build-agent module on every machine with a container runtime. Notes in 0121 and 0162 where their
facts went stale.
2026-10-02 22:29:08 +02:00
mesh-admin
5938d40dee
Merge pull request 'ADRs 0164–0166 and issue 190: the container runtime gets a module, a seat and declared settings' ( #271 ) from decision/docker-module into main
2026-10-02 20:14:58 +00:00
mesh-admin
f062672f83
Merge pull request 'Issues 192 and 193: the console reaches a person only by hand; the store's read-only query is not' ( #272 ) from issues/192-193-console-registration-and-store-query into main
2026-10-02 20:14:46 +00:00
jschoubben
e4a0c73e2b
Merge remote-tracking branch 'origin/main' into issues/192-193-console-registration-and-store-query
2026-10-02 22:14:24 +02:00
jschoubben
d1aeee42a4
Regenerate the decision index after merging main
2026-10-02 22:14:21 +02:00
mesh-admin
81d780f973
Merge pull request 'Design 38: WP3 built — node-tools beside mesh-tools, the three things the plan did not say, and the gate refuses spreading not standing' ( #304 ) from design/38-wp3-built-and-the-gate-softened into main
2026-10-02 19:57:16 +00:00
jochen
3e30846e0f
Design 38: point at ADR 0069's real file name
2026-10-02 21:46:34 +02:00
jochen
b3f18c54c6
Design 38: WP3 built — node-tools beside mesh-tools, the three things the plan did not say, and the gate refuses spreading not standing
2026-10-02 21:46:01 +02:00
mesh-admin
e11bf320c9
Merge pull request 'ADR 0188: a module's own code is bundles in any language, and a tools bundle speaks MCP to the runtime; design 38 gains WP1b' ( #300 ) from decision/0187-a-modules-own-code-is-bundles-in-any-language into main
2026-10-02 19:27:39 +00:00
jschoubben
da8b4b4ee4
Renumber to ADR 0188: 0187 landed on main first, as the dead-tracker record
...
Two records shared 0187 (issue 155's collision); the branch landing last
renumbers, and this is it. Only the number changes.
2026-10-02 21:01:02 +02:00
jschoubben
c026d5221e
Merge remote-tracking branch 'origin/main' into renumber-0187
2026-10-02 21:00:45 +02:00
mesh-admin
983fd412c6
Merge pull request 'ADRs 0180, 0186 and 0187: their live rows, done' ( #302 ) from docs/the-four-fixes-proven-live into main
2026-10-02 18:46:49 +00:00
jschoubben
9ac2493e2c
ADRs 0180, 0186 and 0187: their live rows, done — all four machines filtered by the mesh alone, both front ends removed, no machine wrong or behind
2026-10-02 20:46:35 +02:00
mesh-admin
560f25c2c7
Merge pull request 'ADR 0187: a dead tracker is not the machine's failure' ( #301 ) from fix/a-dead-tracker-is-not-the-meshs-failure into main
2026-10-02 18:42:01 +00:00
jschoubben
9e0288128b
ADR 0187: a dead tracker is not the machine's failure; design 32
2026-10-02 20:41:30 +02:00
jochen
709240ec1f
ADR 0187: a module's own code is bundles in any language, and a tools bundle speaks MCP to the runtime; design 38 gains WP1b
...
The operator's direction, absent from every record until now: the SDK must not limit who writes a
module; tools and services may be written in any language; one module may ship several bundles
(tools, a seat's implementation, a daemon); skeleton first, a full implementation when the work
requires it. ADR 0175 had the runtime import a bundle, which only JavaScript can be.
0187 makes a tools bundle a process the node's runtime launches and speaks MCP over stdio to —
the vocabulary the runtime already speaks outward — so any language with an MCP library can write
one today and the mesh's SDK per language is thin; the transport stays in the runtime (0039's
refusal, kept). Importing a TypeScript bundle is the shortcut, not the contract. Notes in 0175,
0039 and 0150 say where their mechanism moved; design 38 records WP1 as built and adds WP1b (the
launcher and the skeleton SDKs); the glossary's bundle widens.
2026-10-02 18:56:57 +02:00
mesh-admin
d57289e049
Merge pull request 'ADR 0186: a ban list never holds a neighbour, and the mesh's own bans are its own wherever they hang' ( #299 ) from fix/a-ban-list-never-holds-a-neighbour into main
2026-10-02 16:43:32 +00:00
jschoubben
d4a2f99ab5
ADR 0186: a ban list never holds a neighbour, and the mesh's own bans are its own wherever they hang; design 31
2026-10-02 18:42:16 +02:00
mesh-admin
a9f91fdd0c
Merge pull request 'ADRs 0184 and 0185: a service is still running a moment later; a control plane behind its row serves what it can' ( #298 ) from fix/a-service-asked-to-run-is-still-running into main
2026-10-02 16:25:44 +00:00
jschoubben
131a5e4714
Issue 201: what was established about the race while closing the outage half
2026-10-02 18:19:12 +02:00
jschoubben
329a24fdae
ADRs 0184 and 0185: a service is still running a moment later; a control plane behind its row serves what it can; issue 201 half closed
2026-10-02 18:16:24 +02:00
jochen
mesh-admin
jschoubben