quality / commitlint (push) Skipped
CD / update-wiki (push) Failing after 8s
quality / changes (push) Successful in 5s
quality / docker-check (push) Skipped
quality / web (push) Successful in 50s
quality / api (push) Successful in 41s
CD / quality (push) Successful in 1m40s
CD / publish (push) Successful in 1m33s
- Introduced origin health check routes and integrated them into the application. - Updated health check configuration to include success recovery thresholds. - Expanded error handling with new error codes for health check failures. - Added new service routes for managing health checks, including creation and listing. - Improved health check service logic to track consecutive successes and failures. This commit enhances the health check capabilities, providing better monitoring and management of service health.
76 lines
2.0 KiB
TypeScript
76 lines
2.0 KiB
TypeScript
import type { IpHealthState, NodeHealthState } from "@cfdm/shared";
|
|
|
|
export interface HealthThresholds {
|
|
degradedFailures: number;
|
|
downFailures: number;
|
|
successRecoveries: number;
|
|
latencyWarnMs: number;
|
|
}
|
|
|
|
export interface HealthCounters {
|
|
status: string;
|
|
consecutive_failures: number;
|
|
consecutive_successes?: number;
|
|
}
|
|
|
|
export interface NextHealth {
|
|
legacy: IpHealthState;
|
|
node: NodeHealthState;
|
|
failures: number;
|
|
successes: number;
|
|
}
|
|
|
|
function wasHealthy(status: string | undefined): boolean {
|
|
return status === "up" || status === "healthy";
|
|
}
|
|
|
|
function wasUnhealthy(status: string | undefined): boolean {
|
|
return (
|
|
status === "down" ||
|
|
status === "unhealthy" ||
|
|
status === "checking" ||
|
|
status === "degraded"
|
|
);
|
|
}
|
|
|
|
export function nextHealthState(
|
|
ok: boolean,
|
|
latencyMs: number,
|
|
prev: HealthCounters | null,
|
|
thresholds: HealthThresholds,
|
|
): NextHealth {
|
|
if (!ok) {
|
|
const failures = (prev?.consecutive_failures ?? 0) + 1;
|
|
if (failures >= thresholds.downFailures) {
|
|
return { legacy: "down", node: "unhealthy", failures, successes: 0 };
|
|
}
|
|
return { legacy: "degraded", node: "degraded", failures, successes: 0 };
|
|
}
|
|
|
|
if (latencyMs > thresholds.latencyWarnMs) {
|
|
return { legacy: "degraded", node: "degraded", failures: 0, successes: 0 };
|
|
}
|
|
|
|
if (!prev || wasHealthy(prev.status) || !wasUnhealthy(prev.status)) {
|
|
return {
|
|
legacy: "up",
|
|
node: "healthy",
|
|
failures: 0,
|
|
successes: (prev?.consecutive_successes ?? 0) + 1,
|
|
};
|
|
}
|
|
|
|
const successes = (prev.consecutive_successes ?? 0) + 1;
|
|
if (successes >= thresholds.successRecoveries) {
|
|
return { legacy: "up", node: "healthy", failures: 0, successes };
|
|
}
|
|
return { legacy: "unknown", node: "checking", failures: 0, successes };
|
|
}
|
|
|
|
export function toLegacyHealth(status: NodeHealthState | IpHealthState): IpHealthState {
|
|
if (status === "healthy" || status === "up") return "up";
|
|
if (status === "unhealthy" || status === "down") return "down";
|
|
if (status === "degraded") return "degraded";
|
|
return "unknown";
|
|
}
|