fix(services): считать failover по IP Health а не по строке ноды
quality / commitlint (push) Skipped
quality / changes (push) Successful in 9s
quality / docker-check (push) Skipped
CD / update-wiki (push) Successful in 4s
quality / web (push) Successful in 52s
quality / api (push) Successful in 42s
CD / quality (push) Successful in 1m50s
CD / publish (push) Successful in 1m43s

Панель брала service_nodes, которую затирал group apply.
Теперь тот же binding ip_health, что таблица активов; group не пишет в ноду.

Co-authored-by: Cursor <[email protected]>
This commit is contained in:
Denozordec
2026-08-20 15:08:10 +07:00
co-authored by Cursor
parent 994e79e118
commit 7938d2f707
7 changed files with 161 additions and 75 deletions
@@ -386,7 +386,8 @@ function applyAggregatedStatus(
{ colo, provider: statusProvider },
);
const matchedNode = repos.findNodeByIp(db, target.ip);
if (matchedNode && matchedNode.enabled) {
// Binding-scope only: group apply must not clobber node with its own fetch failed.
if (matchedNode && matchedNode.enabled && target.scope === "binding") {
repos.updateNode(db, matchedNode.id, {
health_status: node,
consecutive_failures: failures,
+71
View File
@@ -291,6 +291,77 @@ describe("health-check state derivation via runAllChecks", () => {
expect(targets[0]?.ip).toBe("2.59.161.102");
expect(targets[0]?.hostname).toBe("s.rkns.top");
});
it("does not mark node unhealthy when binding majority is OK and group local fails", async () => {
const { createMemoryDb, repos, runMigrations } = await import("@cfdm/db");
const { db, sqlite } = createMemoryDb();
runMigrations(sqlite);
const tcp = await startTcpServer();
try {
const domain = repos.createDomain(db, null, "example.com", "zone-id");
const group = repos.createServiceGroup(
db,
"VPN",
"vpn",
null,
"vpn.example.com",
{
health_check_enabled: true,
health_check_type: "http",
health_check_port: 1,
health_check_timeout_ms: 200,
health_check_path: "/",
},
);
const service = repos.createService(db, "Svc", "svc");
repos.setServiceGroup(db, service.id, group.id);
repos.setServiceEnabled(db, service.id, true);
const binding = repos.insertBinding(db, domain.id, service.id, "@", null);
repos.updateBindingLbConfig(db, binding.id, {
health_check_enabled: true,
health_check_type: "tcp",
health_check_port: tcp.port,
health_check_timeout_ms: 500,
});
repos.replaceBindingIpsWithMeta(db, binding.id, [
{ ip: "127.0.0.1", weight: 1, priority: 1 },
]);
const node = repos.findNodeByIp(db, "127.0.0.1");
expect(node).not.toBeNull();
await healthCheckService.runAllChecks(db, {
probeGapMs: 0,
thresholds: {
degradedFailures: 1,
downFailures: 1,
latencyWarnMs: 1000,
},
});
const bindingHealth = repos.getIpHealthStatusRow(
db,
"binding",
binding.id,
"127.0.0.1",
);
const groupHealth = repos.getIpHealthStatusRow(
db,
"group",
group.id,
"127.0.0.1",
);
const after = repos.getNode(db, node!.id);
expect(bindingHealth?.status).toBe("up");
expect(groupHealth?.status).toBe("down");
expect(after.health_status).toBe("healthy");
expect(after.consecutive_failures).toBe(0);
expect(after.last_failure_reason).toBeNull();
} finally {
await new Promise<void>((resolve) => tcp.server.close(() => resolve()));
}
});
});
describe("CNAME health mapped onto service IPs", () => {