feat(server): add mongodb command monitoring metrics

[mihir-dixit2k27]

What this PR does

This PR introduces MongoDB observability to the GraphQL server by implementing a Prometheus-based event.CommandMonitor. It enables real-time tracking of database latency, throughput, and concurrency without modifying repository or business logic.

Key Changes

New Telemetry Package

• Added pkg/telemetry/mongo_monitor.go to intercept MongoDB Started, Succeeded, and Failed events.

New Metrics

• litmus_mongo_command_duration_seconds (Histogram) — latency (5ms–10s buckets)
• litmus_mongo_command_total (Counter) — total commands by status (success, failed)
• litmus_mongo_in_flight_commands (Gauge) — concurrent operations

Integration

• Injected monitor in pkg/database/mongodb/init.go
• Exposed metrics via /metrics

Technical Details

• Cardinality Protection: Commands are normalized. High-value commands (find, insert, update, delete, aggregate) are labeled explicitly; others are grouped as "other" to prevent metric explosion.
• Safety: Uses sync.Once for safe metric registration. The gauge is decremented in both Succeeded and Failed handlers to prevent stuck metrics.
• Performance: Relies on MongoDB driver event hooks with negligible overhead and does not wrap or alter repository logic.

Verification

Verified locally using MongoDB 4.4 (Docker).

Steps

go run server.go
curl http://localhost:8080/metrics | grep litmus_mongo

After triggering database operations via the UI or GraphQL queries.

Example Output

litmus_mongo_command_total{command="find",status="success"} 2
litmus_mongo_in_flight_commands{command="find"} 0

Note: Metric values vary based on traffic.

Impact

• Additive change
• No breaking changes
• Negligible performance overhead

[Copilot]

Pull request overview

Adds MongoDB command-level observability to the GraphQL server by introducing a Prometheus-backed MongoDB event.CommandMonitor, enabling visibility into command latency, throughput, and concurrency with minimal impact to existing repository/business logic.

Changes:

• Added a new telemetry package that defines and registers Prometheus metrics and a MongoDB CommandMonitor.
• Wired the command monitor into the MongoDB client initialization (SetMonitor(...)).
• Exposed a Prometheus scrape endpoint at GET /metrics on the main Gin server.

Reviewed changes

Copilot reviewed 4 out of 5 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
chaoscenter/graphql/server/server.go	Registers telemetry metrics at startup and exposes /metrics via promhttp.Handler().
chaoscenter/graphql/server/pkg/telemetry/mongo_monitor.go	Implements MongoDB command monitoring + Prometheus histogram/counter/gauge with command normalization.
chaoscenter/graphql/server/pkg/database/mongodb/init.go	Enables MongoDB driver command monitoring by setting a client monitor.
chaoscenter/graphql/server/go.mod	Adds Prometheus client dependency (and related indirects).
chaoscenter/graphql/server/go.sum	Adds checksums for newly introduced Prometheus dependencies.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.