ADR 0083 and issues 057/058/060/063/064 — the P1 sweep #54

2026-09-18T00:53:57Z

jschoubben commented

2026-09-18 00:53:57 +00:00

The design records for a sweep of P1:

ADR 0083 — one push leaves the mesh consistent (proposed, for your acceptance): a named push flushes every machine now behind, so a cross-node consumer needs one push, not two.
057 resolved — the cascade (mesh-controller PR 31).
058 resolved — the patient runtime (mesh-tools PR 10).
060 resolved — 44 of 70 modules now build (mesh-catalog PR 27); the mechanical majority, with the structural remainder split out.
063 resolved — the foundation's ports are forwarded (a new bug the no-fake bed surfaced; mesh-controller PR 31).
064 opened and located — a mesh build cannot fetch a module's external npm packages or vendor binaries; the reason three modules are deferred from 060.

All proven by a green fresh run of the built-store-cross-node bed, which now enforces each. Checks: records, index, cycle pass (186 documents).

Note: ADR 0083 is proposed — it awaits your acceptance, since it changes what push <node> does.

The design records for a sweep of P1: - **ADR 0083 — one push leaves the mesh consistent** (proposed, for your acceptance): a named push flushes every machine now behind, so a cross-node consumer needs one push, not two. - **057** resolved — the cascade (mesh-controller PR 31). - **058** resolved — the patient runtime (mesh-tools PR 10). - **060** resolved — 44 of 70 modules now build (mesh-catalog PR 27); the mechanical majority, with the structural remainder split out. - **063** resolved — the foundation's ports are forwarded (a new bug the no-fake bed surfaced; mesh-controller PR 31). - **064** opened and located — a mesh build cannot fetch a module's external npm packages or vendor binaries; the reason three modules are deferred from 060. All proven by a green fresh run of the built-store-cross-node bed, which now enforces each. Checks: records, index, cycle pass (186 documents). Note: ADR 0083 is `proposed` — it awaits your acceptance, since it changes what `push <node>` does.

jschoubben added 4 commits 2026-09-18 00:53:57 +00:00

ADR 0083 proposed; 057/058 diagnosed and located 9fd7e6c458

One push leaves the mesh consistent (the 057 decision, proposed for
acceptance); the shared runtime waits for its broker (058). Fixes on
mesh-control fix/one-push-is-enough and mesh-tools
fix/the-runtime-waits-for-its-broker; the built-store-cross-node bed
enforces both.

Issue 063 — the foundation's ports are forwarded, resolved 1572d74a18

The broker's port was accepted on input but not forwarded, so a joined
node reached a DNAT'd broker only until it restarted. Fixed in
mesh-controller (25e42b3), proven by the built-store-cross-node bed
adopting the foundation's broker (which restarts it) and the joined
node still receiving declarations.

ADR 0083: the cascade compares against what was last sent, not a snapshot 0d5693cc2c

Issues 060 resolved, 064 opened; 057/058 on this branch too 98e9e62fce

060: 44 of 70 modules now mesh-buildable (was 8) — the mechanical
majority, proven by direct builds and the bed. The structural
remainders: external-dependency fetch (new issue 064) and route-proxy's
cross-repo build context. 064 records the build environment's isolation
from public npm and Docker Hub.

jschoubben added 1 commit 2026-09-20 10:52:58 +00:00

Accept ADR 0083 — one push leaves the mesh consistent 05857f71f0

jschoubben merged commit 31c5aee72d into main

2026-09-20 10:53:25 +00:00

jschoubben referenced this issue from a commit

2026-09-24 14:21:55 +00:00

ADR 0107: persistent data is a directory bind, never a named volume

jschoubben referenced this pull request

2026-09-24 14:22:01 +00:00

ADR 0107: persistent data is a directory bind, never a named volume #106

Sign in to join this conversation.