Research 032: a module says how it is healthy — evidence, options, recommendation #158

2026-10-06T23:18:11Z

mesh-admin commented

2026-10-06 23:18:11 +00:00

Research 032's findings, measured read-only on the live mesh (2026-10-07). Status stays active; graduation is playbook 02's.

01-evidence.md — 125 catalogue modules; 68 run something long-lived (49 a container, 19 a service). No manifest can declare a check; the node-engine reports no container or unit state; the gate and no probe read one. 19 of 73 long-running catalogue containers have an image HEALTHCHECK (7 modules), 2 of them wrong in the mesh's configuration until fixed. All 94 running containers report restart count 0 (a recreate loses it); the runtime's event history is ~1 minute, all health-check execs. Seven incidents classified: liveness would have caught the crash loop, readiness the 11-hour silent web app, only a module's own tool the refused identity-provider admin.
02-options.md — who runs checks (runtime / node-engine / module tool / mix), where results go (report / event / condition / verb), liveness vs readiness, dependency-aware health, the field's shape, migration.
03-recommendation.md — a proposed decision text (8 rules), how each is checked, and the migration of the 68 modules.

Checks: records, index, cycle pass.

Research 032's findings, measured read-only on the live mesh (2026-10-07). Status stays `active`; graduation is playbook 02's. - **01-evidence.md** — 125 catalogue modules; 68 run something long-lived (49 a container, 19 a service). No manifest can declare a check; the node-engine reports no container or unit state; the gate and no probe read one. 19 of 73 long-running catalogue containers have an image HEALTHCHECK (7 modules), 2 of them wrong in the mesh's configuration until fixed. All 94 running containers report restart count 0 (a recreate loses it); the runtime's event history is ~1 minute, all health-check execs. Seven incidents classified: liveness would have caught the crash loop, readiness the 11-hour silent web app, only a module's own tool the refused identity-provider admin. - **02-options.md** — who runs checks (runtime / node-engine / module tool / mix), where results go (report / event / condition / verb), liveness vs readiness, dependency-aware health, the field's shape, migration. - **03-recommendation.md** — a proposed decision text (8 rules), how each is checked, and the migration of the 68 modules. Checks: records, index, cycle pass.

mesh-admin added 1 commit 2026-10-06 23:18:11 +00:00

Research 032: measure module health on the live mesh and propose a decision

mesh/merge-gate pass: the change touches no module of the mesh's graph

Details

mesh/repo-check pass: its merge-check.sh passed

Details

mesh/delivery delivered

Details

ef6ab814b0

Liveness is judged nowhere and readiness cannot be declared; the evidence,
options and a graduation-ready recommendation say how a module states it.

mesh-admin merged commit 2e8b7321c7 into main

2026-10-06 23:18:40 +00:00

mesh-admin deleted branch research/032-a-module-says-how-it-is-healthy-findings

2026-10-06 23:18:40 +00:00

mesh-admin commented

2026-10-06 23:18:41 +00:00

Delivery novox/hq@ef6ab814b053 — delivered since 2026-10-09T07:31:48Z

Delivery plan — builds nothing: the change touches no module of the mesh's graph

Transitions

2026-10-06 23:18 (new) → proposed (announced): novox/hq#158's head announced
2026-10-06 23:19 proposed → held (merged-unchecked): merged without a passing check — a verdict known with it is taken first: only a person decides that it goes on
2026-10-09 07:31 held → delivered (done): merged, and nothing for a walk to move: builds nothing: the change touches no module of the mesh's graph

The commit's note under refs/notes/mesh-plan keeps every transition: git log --notes=mesh-plan.

**Delivery** `novox/hq@ef6ab814b053` — **delivered** since 2026-10-09T07:31:48Z **Delivery plan** — builds nothing: the change touches no module of the mesh's graph **Transitions** - 2026-10-06 23:18 (new) → proposed (announced): novox/hq#158's head announced - 2026-10-06 23:19 proposed → held (merged-unchecked): merged without a passing check — a verdict known with it is taken first: only a person decides that it goes on - 2026-10-09 07:31 held → delivered (done): merged, and nothing for a walk to move: builds nothing: the change touches no module of the mesh's graph The commit's note under `refs/notes/mesh-plan` keeps every transition: `git log --notes=mesh-plan`.

mesh-admin referenced this pull request

2026-10-06 23:33:18 +00:00

ADR 0240 and to-be 48: a module says how it is healthy, and the node-engine judges it #159

mesh-admin referenced this pull request

2026-10-08 15:28:24 +00:00

Issue 332: the service manager could neither clear a failed unit nor say what wants it #222

jschoubben referenced this issue from a commit

2026-10-09 10:40:28 +00:00

Issue 346: user-dirs.dirs is trusted again for now (mesh-catalog #158), the stop-gap before the fix

Sign in to join this conversation.