Observability

See everything. Debug anything.

Track availability and resource usage, inspect Loki-backed logs, and open Kubernetes diagnostics from the same deployment service page.

Deploy Now
Why It Matters

You can't fix what you can't see

Production issues happen. When they do, you need answers:

  • Is the app running?
  • Why is it slow?
  • What's consuming resources?
  • What happened at 3:47am?

LinqProtocol keeps the operational signals for a workload together, so you can move from a chart to logs and diagnostics without rebuilding the context.

The result is a practical view of what is running, what it is consuming, and where to investigate next.

The Stack

Operational data, surfaced clearly

Resource metrics

Time-series resource data is collected from provider clusters and returned through the LinqProtocol dashboard.

What it does

  • Collects deployment and pod resource samples
  • Keeps metrics separated by provider and deployment
  • Supports selectable time ranges and sample intervals
  • Aggregates multiple returned series for a clear workload view

What you get

  • Availability history
  • CPU consumption
  • Working memory usage
  • Ephemeral and persistent storage usage

Dashboard charts

The deployment service page turns supported metrics into readable summaries and charts.

What it does

  • Shows latest, average, and peak values
  • Plots resource use across the selected range
  • Separates availability from resource consumption
  • Makes missing samples explicit

What you get

  • Built-in charts with no setup
  • Seven-day and custom-range views
  • Units suited to each resource
  • Clear empty and partial-data states

Loki-backed logs

Workload logs are collected centrally and made available from the service page.

What it does

  • Collects logs from deployment workloads
  • Queries logs for a bounded time range
  • Scopes results to the selected service or pod
  • Keeps logs beside status and diagnostics

What you get

  • Timestamped service output
  • Time-range controls
  • Copyable logs for debugging or support
  • One place to move between metrics, logs, and diagnostics
Dashboard

Your deployments at a glance

Every deployment in LinqProtocol includes access to:

Availability Overview

Readiness history, latest sample, and coverage for the selected range

Resource Metrics

CPU, memory, ephemeral storage, and persistent storage over time

Service Logs

Timestamped Loki-backed logs for the selected workload

Debug Panel

Kubernetes status, events, and workload diagnostics

Access everything from the LinqProtocol dashboard without additional setup.

Metrics

The metrics available today

Availability

  • Kubernetes readiness samples
  • Average, latest, and lowest readiness
  • Sample coverage across the selected range

Compute

  • CPU cores consumed
  • CPU percentage when limit data exists
  • Working memory in MB
  • Memory percentage when limit data exists

Storage

  • Ephemeral storage in MB
  • Persistent volume usage in MB
  • Persistent volume percentage when capacity data exists
  • Latest, average, and peak values
Logs

Workload logs in context

Logs are collected through Loki and queried from the service page for the workload and time range you select.

Features:

Timestamped workload output
Service and pod-scoped results
Bounded time-range queries
Copy logs for debugging or support

Best practices:

  • • Use structured logging (JSON) for better searchability
  • • Include request IDs for tracing
  • • Don't log sensitive data
Debugging

When things go wrong

Issues Panel

Surfaces critical information about your deployment:

  • Current workload status
  • Failed scheduling or startup events
  • Replica and readiness state
  • Recent Kubernetes warnings

Pod-Level Diagnostics

Drill into specific pods to see:

  • Current status and events
  • Container and image details
  • Replica state
  • Related logs

Namespace-Level View

See all pods in your deployment:

  • Which workloads are healthy
  • Which workloads need attention
  • Recent events
Coming Soon

Get notified before users complain

Coming soon: configure alerts based on metrics thresholds.

Example Alerts:

  • CPU usage > 90% for 5 minutes
  • Error rate > 1% for 10 minutes
  • Pod restarts > 3 in 1 hour
  • Custom metric conditions

Notifications via:

Email
Discord
Slack
Telegram
Webhooks

Deploy and observe