jschoubben
9dc57b4712
Continuing the sweep. Both were answered by records that did not cite them, which is the same pattern 003 showed -- an effort stays active because the decision that resolved it was reached from another direction. 005 graduates. Three of its four questions are answered: provider modules do not group (0044), the ~50 modules that co-change with nothing stay as they are, and 'group or leave' was never the right pair -- 0054 reframes it as authority versus package. Worth noting the debt runs the other way too: this effort's measurement, that reachability is the ONLY place modules genuinely co-change, is what 0054 rests on and why connectivity is a context while nothing else needed one. Its fourth question moves rather than closes. Whether applications leave the monorepo before or after they group is a sequencing question, so it belongs to 009-migration. 008 stays active, with its central question marked answered: the coordinator converges nodes on a declaration rather than dispatching stages (0058). The three-silo split survives with the third redefined. What 0058 explicitly does NOT answer is how a change becomes a pipeline reliably -- detection is upstream of everything it changed and remains the fragile input.
4.4 KiB
status, initiated, touches, became
| status | initiated | touches | became | ||||
|---|---|---|---|---|---|---|---|
| active | 2026-08-23 |
|
008 — The coordinator: a change checked in becomes a deployed state
What is being investigated
The mesh's own continuous delivery: a change is committed, and the mesh ends up in the state that change describes — across every node the change touches, with a verdict that says whether it worked.
The coordinator is what orchestrates that, and it is the mesh's most consequential machinery: everything reaches every node through it.
Why now
The as-is record (03-DESIGN/00-as-is/04-delivery.md)
names problems that are structural rather than incidental:
- A green pipeline proves transport, not effect. The stages report that a message was dispatched and accepted, which is not the same as the thing running, correct, or present. This is the mesh's single most consistent failure shape.
- Detection is the most fragile input. A merge that creates no pipeline, with nothing saying so, is the characteristic bad outcome — and it has happened for reasons unrelated to the change.
- The fan-out point is asymmetric. The build node has already passed two silos when work fans out, and code that knew only about the first parked it forever while every other node deployed cleanly.
- There is no end-to-end coverage. The harness has not built since 2026-06-04
(
04-ISSUES/005).
Research 006 adds a requirement the current design does not have: the coordinator must work before the mesh is self-hosting, when source and artifacts come from outside, and keep working across the transition to self-hosted providers.
Answered since
Does the coordinator dispatch stages, or converge nodes on a declaration? — Converge. ADR 0058: a pipeline ends when the declaration is updated, and deploy stops being once-per-node. The host applies it and reads back, so the reporter is the applier — which is what this effort was circling when it asked what a deployed state is.
Does the three-silo split survive? — Yes, with the third redefined. Build and publish are unchanged; deploy becomes one write rather than a fan-out.
What is a deployed state? — Partly. "The declaration is updated, and here is which nodes have applied it" is the answer 0058 gives, and it makes outstanding a first-class result rather than a stall. What it does not answer is what a node reports and how, which waits on the link.
Explicitly NOT answered, and 0058 says so: how a change becomes a pipeline, reliably. Detection remains the fragile input — a merge that created no pipeline, and nothing said so is upstream of everything 0058 changed and is untouched by it.
The questions
| Question | Why it matters |
|---|---|
| What is a deployed state, and how does the mesh know it is in one? | Everything follows from this. If a stage reports transport, "deployed" is a claim nobody checked. A desired-state model with reconciliation gives a different answer from a job-completion model. |
| Does the coordinator dispatch stages, or converge nodes on a declaration? | The current model is a state machine over stages. The alternative is that a node is told what should be true and reports what is. The second makes drift visible; the first cannot see it. |
| How does a change become a pipeline, reliably? | Detection has failed for reasons unrelated to the change, silently. |
| What produces a verdict, and what is it a verdict about? | Ties to the lab (ADR 0016) and to a module carrying its own assertions. |
| How does delivery work before self-hosting, and across the transition? | From research 006: source and artifacts start external and are re-bound to internal providers. The coordinator has to be indifferent to which. |
| Does the three-silo split survive the artifact/part split? | ADR 0014 is cardinality-driven, and research 006 renames the thing the cardinality is about. |