Merge pull request 'tools-confluence: sixth green bed (confluence runtime serves 3 tools, no creds)' (#9) from feat/tools-confluence into main

2026-09-06 01:54:54 +02:00
parent 0b8f583f75 e703a94fcd
commit b66b330dba
2 changed files with 283 additions and 0 deletions
@@ -0,0 +1,41 @@
+# One machine that becomes a mesh and is then assigned confluence — a tools-only, outbound-only
+# external-SaaS integration (novox/hq ADR 0039), the same shape gitlab proved. confluence has NO
+# service container, NO listener, NO provisioner — just a broker-bound runtime that serves the
+# module's Confluence tools under a scoped account. It only ever calls out to a Confluence instance.
+#
+# The sharp point this bed proves is the Servarr lesson: the runtime MUST come up and serve its full
+# tool surface even with NO valid Confluence credentials — the lab has no real Confluence, and
+# confluence's own token is a mesh-minted own-secret that points at nothing. confluence's client is
+# built lazily and never throws at registration, so the runtime logs `[mesh-tools] serving 3 tool(s)`
+# and binds its serve queues regardless; a tool would only fail if it were actually invoked.
+#
+#   MESH_LAB_HOST_BINARY=.../mesh-host   MESH_LAB_BUNDLE=.../examples/substrate-first-node.lock
+#   scripts/build-module-runtime.sh confluence builds mesh-runtime-confluence:development into the
+#   local daemon, which this scenario stocks and serves by digest from its own registry. confluence
+#   needs no service image — it is tools-only.
+scenario: tools-confluence
+
+segments:
+  hosting:
+    kind: public
+    cidr: [192.0.2.0/24]
+
+machines:
+  anchor:
+    at: { segment: hosting, address: [192.0.2.10] }
+    inbound: allow
+    memory: 3GiB
+    cpus: 2
+
+images:
+  # The first-node substrate: store, broker, control.
+  - postgres:17-alpine
+  - cloudamqp/lavinmq:latest
+  - mesh-control:development
+  # confluence's runtime, built by scripts/build-module-runtime.sh confluence into the local daemon
+  # and stocked into the scenario's own registry, which is where the host pulls it from. There is no
+  # service image: confluence is tools-only and outbound-only.
+  - mesh-runtime-confluence:development
+
+place:
+  all: [host, runtime]
@@ -0,0 +1,242 @@
+/**
+ * The mesh assigns confluence — a tools-only, outbound-only external-SaaS integration
+ * (novox/hq ADR 0039), the shape gitlab proved — and its runtime comes up and serves the module's
+ * full tool surface, with NO valid Confluence credentials.
+ *
+ * confluence is the thinnest converted shape: no service container, no listener, no provisioner —
+ * only a broker-bound runtime that serves the module's Confluence tools under a scoped account
+ * (ADR 0047). The sharp point (the Servarr lesson) is that the runtime MUST register and serve every
+ * tool even though the lab has no real Confluence and confluence's own token is a mesh-minted
+ * own-secret pointing at nothing. confluence's client is built lazily and never throws at
+ * registration, so:
+ *   - the mesh-runtime-confluence container comes up and STAYS up (no crash on a missing token);
+ *   - it logs `[mesh-tools] serving 3 tool(s)` — the install proof;
+ *   - it binds its serve queues on the broker (e.g. serve.confluence.confluence_search), so the tools
+ *     are actually reachable, not merely loaded;
+ *   - it got its own scoped broker account (anchor-confluence).
+ * A tool would only fail if it were actually invoked without real creds — which this bed does not do,
+ * because the point is exactly that serving does not require them.
+ *
+ *   MESH_LAB_HOST_BINARY=.../mesh-host   MESH_LAB_BUNDLE=.../examples/substrate-first-node.lock
+ *   scripts/build-module-runtime.sh confluence builds mesh-runtime-confluence:development into the
+ *   local daemon; scenarios/tools-confluence.yml stocks it. There is no service image.
+ */
+
+import { test, before, after } from "node:test";
+import assert from "node:assert/strict";
+import { existsSync, readFileSync } from "node:fs";
+import { loadScenario } from "../../src/declaration/parse.ts";
+import { raise } from "../../src/lifecycle/raise.ts";
+import { destroy, exec } from "../../src/lifecycle/operate.ts";
+import { hostBinaryPath, HOST_PATH } from "../../src/lifecycle/place.ts";
+import { labIsUsable, destroyAll } from "./harness.ts";
+
+const capability = await labIsUsable();
+const binary = hostBinaryPath();
+const bundle = process.env["MESH_LAB_BUNDLE"] ?? "";
+
+const skip = !capability.usable
+  ? `lab not usable: ${capability.why}`
+  : !binary || !existsSync(binary)
+    ? "MESH_LAB_HOST_BINARY is not set to a built mesh-host"
+    : !bundle || !existsSync(bundle)
+      ? "MESH_LAB_BUNDLE is not set to a substrate bundle (mesh-host examples/)"
+      : false;
+
+const SCENARIO = "tools-confluence";
+const MACHINE = "anchor";
+
+let instanceId = "";
+/** What the scenario's registry serves, by digest. */
+let stocked: string[] = [];
+
+function quote(s: string): string {
+  return `'${s.replaceAll("'", `'\\''`)}'`;
+}
+
+async function on(command: string, timeoutMs?: number): Promise<{ out: string; ok: boolean }> {
+  const { stdout } = await exec(instanceId, MACHINE, [
+    "sh", "-c", `exec 2>&1\n${command}\necho "__exit=$?"`,
+  ], timeoutMs);
+  const marker = stdout.lastIndexOf("__exit=");
+  if (marker < 0) return { out: stdout, ok: false };
+  return { out: stdout.slice(0, marker), ok: stdout.slice(marker + 7).trim() === "0" };
+}
+
+async function must(command: string, timeoutMs?: number): Promise<string> {
+  const { out, ok } = await on(command, timeoutMs);
+  if (!ok) throw new Error(`${MACHINE}: ${command}\n${out}`);
+  return out;
+}
+
+/** The control plane, a container on the node. */
+async function mesh(command: string, timeoutMs?: number): Promise<string> {
+  return must(`docker exec mesh-control /mesh-control ${command}`, timeoutMs);
+}
+
+/** The pinned reference for one of the scenario's images, by repository. */
+function pinned(repository: string): string {
+  const found = stocked.find((r) => r.slice(r.indexOf("/") + 1, r.indexOf("@")) === repository);
+  assert.ok(found, `the scenario stocks no ${repository}; it serves ${stocked.join(", ")}`);
+  return found;
+}
+
+/** The substrate bundle, its image references pointed at this scenario's own registry. */
+function bundleFor(images: string[]): string {
+  let text = readFileSync(bundle, "utf8");
+  for (const ref of images) {
+    const repository = ref.slice(ref.indexOf("/") + 1, ref.indexOf("@"));
+    const escaped = repository.replaceAll("/", "\\/").replaceAll(".", "\\.");
+    text = text.replaceAll(new RegExp(`[A-Za-z0-9_.:-]+\\/${escaped}@sha256:[0-9a-f]+`, "g"), ref);
+  }
+  return text;
+}
+
+function tokenFrom(said: string): string {
+  const found = said.split("\n").map((l) => l.trim()).find((l) => l.length > 100 && !l.includes(" "));
+  assert.ok(found, `no token in:\n${said}`);
+  return found;
+}
+
+async function settled(withinMs = 600_000): Promise<void> {
+  const until = Date.now() + withinMs;
+  let last = "";
+  while (Date.now() < until) {
+    const asked = await on(`docker exec mesh-control /mesh-control status --json`);
+    if (asked.ok) {
+      try {
+        const state = JSON.parse(asked.out) as {
+          wrong: { node: string; outcome: string }[];
+          waiting: { node: string }[];
+          reported: { node: string; outcome: string; current: boolean }[];
+        };
+        const bad = state.wrong.find((w) => w.node === MACHINE);
+        if (bad) throw new Error(`${MACHINE} did not apply what it was sent: ${bad.outcome}\n${asked.out}`);
+        const word = state.reported.find((r) => r.node === MACHINE);
+        if (!state.waiting.some((w) => w.node === MACHINE) && word?.outcome === "applied" && word.current) return;
+        last = asked.out;
+      } catch (err) {
+        if (err instanceof Error && err.message.includes("did not apply")) throw err;
+        last = asked.out;
+      }
+    }
+    await new Promise((r) => setTimeout(r, 5000));
+  }
+  throw new Error(`${MACHINE} never caught up within ${Math.round(withinMs / 1000)}s. Last:\n${last}`);
+}
+
+before(async () => {
+  if (skip) return;
+
+  const raised = await raise(loadScenario(`scenarios/${SCENARIO}.yml`), {
+    onProgress: (m) => console.log(`raise: ${m}`),
+  });
+  instanceId = raised.instanceId;
+  stocked = raised.images;
+
+  // Raise the substrate — store, broker, control — from the bundle.
+  await must(`cat > /tmp/substrate.lock <<'MESHBUNDLE'\n${bundleFor(raised.images)}\nMESHBUNDLE`);
+  await must(`${HOST_PATH} apply /tmp/substrate.lock`, 600_000);
+  const up = await must(`docker ps --format '{{.Names}}'`);
+  for (const c of ["mesh-store", "mesh-broker", "mesh-control"]) {
+    assert.match(up, new RegExp(c), `the substrate did not raise ${c}:\n${up}`);
+  }
+
+  // The node joins its own mesh, so it is a node the mesh can assign to, and start the host so it
+  // applies what it is pushed.
+  await mesh(`node add ${MACHINE}`);
+  const token = tokenFrom(await mesh(`token issue --node ${MACHINE}`));
+  await must(`${HOST_PATH} enrol --token ${quote(token)}`);
+  await must(`nohup ${HOST_PATH} run > /var/log/mesh-host.log 2>&1 & sleep 3`);
+}, { timeout: 1_800_000 });
+
+after(async () => {
+  if (instanceId) await destroy(instanceId);
+  await destroyAll(`${SCENARIO}-`);
+}, { timeout: 600_000 });
+
+test("the mesh assigns confluence: its tools-only runtime comes up and serves the full tool surface with no valid token", {
+  skip, timeout: 1_500_000,
+}, async () => {
+  // confluence is tools-only and outbound-only: no service, no listener, no provisioner — just a
+  // broker-bound runtime serving the module's tools. Its own token is a mesh-minted own-secret; the
+  // lab has no real Confluence, so the token points at nothing — and that is the case under test: the
+  // runtime must serve every tool regardless. The runtime container name and shape mirror the
+  // committed manifest, with the image pinned to what this scenario serves by digest.
+  const manifest = JSON.stringify({
+    module: "confluence",
+    version: "1",
+    "own-secrets": {
+      token: "/var/lib/confluence/token",
+      broker: "/var/lib/mesh/confluence/broker",
+    },
+    resources: [
+      { id: "mesh-state", type: "directory", path: "/var/lib/mesh/confluence", mode: "0700" },
+      { id: "state", type: "directory", path: "/var/lib/confluence", mode: "0700" },
+      { id: "config", type: "file", path: "/var/lib/confluence/config.json", merge: "json", content: "{}", mode: "0600" },
+      {
+        id: "runtime", type: "container", name: "mesh-runtime-confluence",
+        image: pinned("mesh-runtime-confluence"), network: "host",
+        volumes: [
+          "/var/lib/confluence/config.json:/run/config/config.json:ro",
+          "/var/lib/confluence/token:/run/secrets/token:ro",
+          "/var/lib/mesh/confluence/broker:/run/secrets/broker:ro",
+        ],
+        env: {
+          MESH_CONFLUENCE_TOKEN_FILE: "/run/secrets/token",
+          MESH_CONFLUENCE_CONFIG_FILE: "/run/config/config.json",
+          MESH_BROKER_FILE: "/run/secrets/broker",
+        },
+      },
+    ],
+  });
+
+  await must(`printf %s ${quote(manifest)} > /tmp/confluence.json && docker cp /tmp/confluence.json mesh-control:/confluence.json`);
+  await mesh("module add /confluence.json");
+  // confluence serves tools, so it is issued a scoped broker account (it emits/consumes nothing else).
+  const issued = await mesh(`module issue confluence --node ${MACHINE}`);
+  assert.match(issued, /scoped to what it emits and consumes/, issued);
+  await mesh(`assign ${MACHINE} confluence`);
+
+  // ONE push, ONE convergence.
+  await mesh(`push ${MACHINE}`);
+  await settled();
+
+  // --- the runtime is up and STABLE — it did not crash on a missing/invalid token -----------------
+  const running = await must(`docker ps --format '{{.Names}}'`);
+  assert.match(running, /(^|\n)mesh-runtime-confluence(\n|$)/,
+    `mesh-runtime-confluence is not running after the push:\n${running}\n---host log---\n${(await on(`tail -40 /var/log/mesh-host.log`)).out}`);
+  await new Promise((r) => setTimeout(r, 5000));
+  const state = (await must(`docker inspect -f '{{.State.Running}} {{.RestartCount}}' mesh-runtime-confluence`)).trim();
+  assert.equal(state, "true 0",
+    `the confluence runtime is not up and stable — it should serve tools even with no valid token:\n${state}\n` +
+      `${(await on(`docker logs mesh-runtime-confluence 2>&1 | tail -30`)).out}`);
+
+  // --- the install proof: the runtime logs it is serving the full tool surface --------------------
+  let log = "";
+  const untilLogged = Date.now() + 60_000;
+  while (Date.now() < untilLogged) {
+    log = (await on(`docker logs mesh-runtime-confluence 2>&1`)).out;
+    if (/\[mesh-tools\] serving \d+ tool\(s\)/.test(log)) break;
+    await new Promise((r) => setTimeout(r, 3000));
+  }
+  const served = log.match(/\[mesh-tools\] serving (\d+) tool\(s\)/);
+  assert.ok(served, `the confluence runtime never logged it is serving tools:\n${log}`);
+  assert.equal(Number(served[1]), 3,
+    `the confluence runtime served ${served?.[1]} tools, expected the full surface of 3:\n${log}`);
+
+  // --- the tools are actually reachable: the runtime bound its serve queues on the broker ---------
+  let served2 = "";
+  const untilServing = Date.now() + 90_000;
+  while (Date.now() < untilServing) {
+    served2 = await must(`docker exec mesh-broker lavinmqctl list_queues name 2>&1 || true`);
+    if (/serve\.confluence\.confluence_search/.test(served2)) break;
+    await new Promise((r) => setTimeout(r, 3000));
+  }
+  assert.match(served2, /serve\.confluence\.confluence_search/,
+    `confluence's runtime never bound its serve queue:\n${(await on(`docker logs mesh-runtime-confluence 2>&1 | tail -20`)).out}\n---\n${served2}`);
+
+  // --- confluence got its own scoped broker account -----------------------------------------------
+  const users = await must(`docker exec mesh-broker lavinmqctl list_users 2>&1`);
+  assert.match(users, /anchor-confluence/, `the scoped account anchor-confluence is not on the broker:\n${users}`);
+});