Skip to content

summary.md

Latest commit

 

History

History
59 lines (39 loc) · 2.43 KB

summary.md

File metadata and controls

59 lines (39 loc) · 2.43 KB

Summary Module

Aggregates real-time probe data into health metrics for each target.

The Summary module acts as a stateful aggregator. Instead of recalculating statistics from raw logs every time, it maintains a persistent "snapshot" of a target's performance, including uptime percentages, average latency, and current health status.

This module is implemented in the src/modules/summary directory.

API Routes

  • GET /targets/:id/summary: Retrieves the calculated metrics for a specific target (Validated by summarySchema).

Public Service Interface

The summary.service.js provides read access to the aggregated data:

  • getSummary(id): Finds the summary document associated with a target ID. Returns null or a warning if no checks have been performed yet.

Data Model & Schema

  • Collection: summaries
  • Key Metrics:
    • totalCheckCount: Total number of probes executed.
    • successCount / failureCount: Counters used to calculate uptime.
    • uptime: Percentage of successful checks.
    • avgLatency: Rolling average of response times.
    • consecutiveFailureCount: Current streak of failed checks (critical for the Alert module).
    • lastStatus: Current state enum (HEALTHY | DOWN).

Technical Constraints

  • Persistence: Summary data is persistent and represents the entire history of the target.
  • Write Logic: Though managed in this document, the actual updates to this model are performed by the Check service using atomic MongoDB operators like $inc and $set.
  • Validation: Target IDs must be exactly 24 characters (standard MongoDB ObjectID).

Dependencies

  • Inbound: Queried by the Alert module to determine if thresholds are met.
  • Inbound (Writes): Updated by the Check module after every successful or failed probe.
  • Database: Mongoose / MongoDB.

Error Handling

  • Missing Records: If a target exists but hasn't been checked, the service logs a warning rather than throwing an error, allowing the UI to handle an "empty" state.
  • Data Integrity: Relies on the Check module to ensure consecutiveFailureCount never drops below zero using a secondary $set correction.

Usage Example

import { getSummary } from "./summary.service.js";

// Fetch health snapshot for a target
const stats = await getSummary("target-object-id");

if (stats) {
  console.log(`Current Status: ${stats.lastStatus}`);
  console.log(`Uptime: ${stats.uptime}%`);
}