Gameye Infrastructure Observability

TL;DR: Key Takeaways

Gameye includes built-in observability for session-based game server fleets: real-time logs and performance metrics, dashboards for active sessions (including regional breakdown), per-server CPU/memory/bandwidth with alerting, and custom game KPIs graphed from JSON logged inside your containers.

 

  • Operators can filter/sort/group logs, stream raw data live, and (optionally) ship logs and metrics to existing observability stacks.
  • An Infrastructure View provides a live bird’s-eye view of running workloads and a “go back in time” snapshot mode for incident review.

Observability built for how game sessions actually run/fail

Dedicated game servers aren’t long-lived web services. Sessions spin up fast, churn constantly, and the impact of issues shows up as “players can’t join,” “matches crash,” or “a region melts.” Gameye’s observability is designed around that reality: session-level visibility, per-server resource truth, and fast drill-down from fleet → region → server → logs.

Whether you’re planning a launch, operating live events, or scaling across regions, Gameye gives you the operational clarity to answer two questions quickly:

Dashboards that match game server operations

Active Sessions Dashboard

Understand demand at a glance. Gameye shows how many sessions are currently active and supports breakdowns like by region so you can spot hotspots, imbalances, and sudden spikes quickly.

Use it for:

Container Metrics Dashboard (per game server)

See the metrics that actually matter for game servers: CPU, memory, and bandwidth at the server/container level, with support for alerting when you approach thresholds that put gameplay at risk.

Use it for:

Custom Metrics Dashboard (your game KPIs)

Gameye can graph your own game-specific metrics (e.g., tick rate, match duration, player counts, disconnect rate, queue times) by ingesting structured JSON logs emitted from inside your container. That gives your team a practical path to seeing game KPIs alongside infrastructure signals—without standing up a separate pipeline first.

Use it for:

Logs: fast isolation, live streaming, and optional export

When something breaks, speed wins. Gameye’s logs workflow is built to reduce time-to-answer:

This makes Gameye usable out of the box for operators, while still supporting studios with established monitoring standards.

Infrastructure View: see what’s running where (and rewind time)

Dashboards show trends; incidents require context. Gameye’s Infrastructure View provides a live, bird’s-eye snapshot of workloads running across your servers, with direct paths into per-server metrics and logs.

Two capabilities are especially useful for real-world operations:

Alerting & telemetry integrations

Gameye supports alerting and aligns with common observability patterns, including Prometheus-style metrics and integrations frequently used by engineering organizations. If your studio already has a preferred stack (e.g., Datadog or Elastic), Gameye can complement it: use Gameye’s session-centric dashboards for operations, and export telemetry where you need it for centralized visibility and retention.

Operational Scenarios

Scenario

What if one region spikes 5× in 10 minutes?
  • Confirm demand concentration in the Active Sessions dashboard (by region).
  • Check CPU/memory/bandwidth on servers handling the region to identify the real bottleneck.
  • Use alerting to catch threshold breaches before they cascade into player-impacting failures.

Scenario

What if a new build causes crashes, but only for one game mode?
  • Graph custom metrics (or structured JSON signals) for build/version, game mode, crash reason, etc.
  • Use logs filtering/grouping to isolate the failing slice quickly.
  • Use time-based snapshots to compare the “before vs. after rollout” state of running workloads.

Scenario

What if sessions start fine, then degrade after 20–30 minutes?
  • Watch memory growth and long-running CPU trends per server to detect leaks and creeping contention.
  • Stream logs live to catch the first onset of degradation and correlate it with resource behavior.

Scenario

What if my central observability team wants everything in one place?
  • Keep Gameye’s dashboards for session-centric operations and on-call triage.
  • Ship logs and metrics to your existing platform for standardized retention and cross-service correlation.

Scenario

What if I need more log storage space (or longer retention)?

Need longer than the default 3-day log retention window? Gameye can provide additional log storage/retention beyond the default at a competitive rate, typically provisioned within 24 hours (plan- and terms-dependent).

Scenario

What if I need PagerDuty alerts?

If your team uses PagerDuty, Gameye can work with you to connect your PagerDuty account so alerts from the dashboard/alerting system can trigger PagerDuty incidents in your PD account (configured per customer based on routing and escalation requirements).

Summary

Gameye Observability is built for the operational reality of dedicated game servers:

Want to see it in action? Talk to the Gameye team about a sandbox environment and a walkthrough of the dashboards for your game.

FAQ

Does Gameye include dashboards out of the box?

Yes. Gameye includes dashboards for active sessions, container-level CPU/memory/bandwidth, alerting, and custom metrics derived from structured JSON logs generated inside your containers.

Can we keep our existing observability stack?

Yes. You can use Gameye’s built-in dashboards for session-focused operations and optionally ship logs and metrics into your existing platform for centralized retention and correlation.

Can we investigate incidents after the fact?

Yes. The Infrastructure View supports “go back in time” snapshots so you can review what was running where at a specific point in time—useful for postmortems and release validation.

Can we retain logs longer than the default window?

Yes. Gameye can extend log retention beyond the default 3 days, typically within 24 hours, at a competitive rate (plan-dependent).

Can Gameye trigger PagerDuty incidents in our account?

Yes. Gameye can work with you to connect your PagerDuty account so alerts can trigger incidents within your PD services/escalation policies.