Observability

Full visibility.
Zero gaps.

Built-in metrics, heartbeats, logs, and audit trails for every deployment, out of the box. Redirect to your own OpenTelemetry stack when you need it. Alerts fire and resolve automatically.

Start FreeRead the Docs

3

telemetry types

4

alert types

365d

max retention

OTEL

native export

Architecture

Warden ships it. You consume it.

The Warden agent on each node continuously collects metrics, heartbeats, and logs, then pushes them to the LaserData control plane. Available immediately in the Console, API, or forwarded to your own OTEL-compatible endpoint.

Your Deployment Node

Iggy Server
Connectors
Stream UI
Warden Agent

Warden ships telemetry

LaserData Control Plane

Metrics Store
Log Ingestor
Heartbeat Tracker
Alert Engine
Console · API · OTEL redirect

Console UI

Dashboard · Metrics · Logs

REST API

Supervisor API endpoints

Your OTEL Stack

Loki · Datadog · Elastic…

Metrics

Per-node. Per-runtime. Continuous.

Metrics are tagged by node and runtime (Iggy, Connectors) and available with time-range filtering from the Console or API.

What's Collected

SystemCPU usage, total CPU, memory usage, available memory
ProcessPID, run time, start time
Disk I/OBytes read, bytes written
IggyMessages, streams, topics, partitions, segments
ClientsConnected clients, consumer groups
laserdata.cloud
Console: Metrics tab — CPU, memory, disk, load average, file descriptors, network I/O for a deployment

How to View

  1. 1Navigate to your deployment in the Console
  2. 2Open the Metrics tab
  3. 3Select a node and runtime
  4. 4Adjust time range to zoom in

API Example

# Get all metrics for a deployment
curl {supervisor_url}/deployments/{deployment_id}/metrics \
  -H "ld-api-key: YOUR_API_KEY"

# Get per-node Iggy metrics with time range
curl "{supervisor_url}/deployments/{deployment_id}/nodes/{node_id}/metrics/iggy?page=1&results=10&to=2026-02-01T00:00:00Z" \
  -H "ld-api-key: YOUR_API_KEY"

Heartbeats & Logs

Always-on health and log streams.

Heartbeats

Warden sends periodic heartbeats for each managed runtime (Iggy, Connectors). Heartbeats confirm the runtime is alive and responsive. Missing heartbeats beyond 120 seconds trigger a node_unreachable alert.

  • Viewable per node and per runtime
  • Time-range filtering for historical analysis
  • Auto-resolves when node comes back online
curl {supervisor_url}/deployments/{deployment_id}/heartbeats \
  -H "ld-api-key: YOUR_API_KEY"

Logs

Centralized log collection from all runtimes on all nodes. Ingested by Warden and stored for querying with rich filters.

NodeSpecific node
RuntimeIggy · Connectors · Warden
Leveldebug · info · warn · error
MessagePattern match
ScopeComponent filter
Time rangeStart + end timestamps
curl "{supervisor_url}/deployments/{deployment_id}/logs/iggy?page=1&results=10&level=error&message=* \
  -H "ld-api-key: YOUR_API_KEY"
laserdata.cloud
Console: Logs Viewer — node + runtime filter, deployment log lines streamed from Iggy / Warden / Connectors

Alerts

Fire fast. Resolve automatically.

LaserData automatically monitors your deployments and fires alerts when thresholds are breached. Each alert has a matching resolution event. Alerts are deduplicated: no repeated noise while a condition persists.

high_cpu_usageCPU ≥ 80% (5-min window)

Sustained high CPU usage on a node

cpu_usage_resolved
high_memory_usageMemory ≥ 90%

Memory usage approaching capacity

memory_usage_resolved
high_disk_usageDisk ≥ 80%

Disk usage approaching capacity

disk_usage_resolved
node_unreachableNo heartbeat for 120s

Node stopped reporting heartbeats

node_reachable

Connect alerts to your channels

Alerts are delivered via your configured notification channels via Slack, webhook, or email. Subscribe to specific event types and filter by deployment, environment, or division.

OpenTelemetry

Bring your own observability stack.

By default, all telemetry goes to LaserData Cloud. When you need data in your own systems (for compliance, privacy, or integration), redirect logs and traces to any OTEL-compatible endpoint.

How it works

When log redirection is enabled, Warden sends logs using the standard OTEL ExportLogsServiceRequest protocol. Traces use ExportTracesServiceRequest.

Grafana Loki
Datadog
Elastic
Any OTEL-compatible collector

Configure via

Console deployment settings or by contacting support. Your OTEL endpoint can be internal or external, as long as it accepts standard OTEL protocol traffic.

OTEL Protocols Supported

ExportLogsServiceRequest

Type: Logs · gRPC: yes

ExportTracesServiceRequest

Type: Traces · gRPC: yes

Notifications

Alert where your team lives.

Multi-channel delivery with per-subscription event filtering. Scope notifications to specific deployments, environments, or divisions. Experimental: APIs may change.

💬
slack

Formatted messages via incoming webhook

🔗
webhook

JSON POST to any HTTPS endpoint

✉️
email

HTML email with subject and body

📱
sms

Phone delivery (coming soon)

How channels and subscriptions work

  1. 1Create a channel: define where (Slack URL, email, webhook endpoint)
  2. 2Add subscriptions: define which events trigger it, scoped to deployments/environments
  3. 3Test the channel: send a test notification to verify delivery
  4. 4Alerts fire automatically as events match your subscriptions

Plan limits

ResourceBasicProEnterprise
Channels1320
Subscriptions31020
Create channel
curl -X POST https://api.laserdata.cloud/tenants/{tenant_id}/channels \
  -H "ld-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "channel": "slack",
    "name": "production-alerts",
    "destination": "https://hooks.slack.com/services/T00/B00/xxx",
    "settings": {
      "slack": { "channel": "#alerts", "username": "LaserData" }
    }
  }'
Create subscription
curl -X POST https://api.laserdata.cloud/tenants/{tenant_id}/channels/{channel_id}/subscriptions \
  -H "ld-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "message_types": [
      "node_unreachable",
      "high_cpu_usage",
      "high_memory_usage"
    ],
    "scope_deployment_ids": [42]
  }'
laserdata.cloud
Console: Create Notification Channel — Slack, Webhook, Email options with channel settings
Full notifications API reference

Audit & Compliance

Immutable. Encrypted. Tenant-isolated.

Every state-changing operation is recorded in an immutable audit log. Records cannot be modified or deleted. All PII is encrypted at rest.

01

Audit Logging

Every create, update, and delete across all resources. Includes actor info, resource context (tenant → division → environment → deployment), previous values on updates, and precise timestamps.

Read the docs
02

Access Logs

Who read sensitive data: user profiles, member listings, invitations, audit log reads, data exports. Includes request ID, actor, action, IP address, and User-Agent. Answers "who looked at this?"

Read the docs
03

Security Events

Permission denied attempts and authentication failures captured with failure reason and request metadata. Designed for anomaly detection and incident response.

Read the docs

GDPR & Data Protection

  • PII encrypted at restEmails, names, IDP external IDs, invitation emails, all encrypted
  • Data exportFull personal data export in JSON: profile, sessions, memberships, invitations
  • Right to erasureFull account deletion: identities, memberships, invitations removed
  • Audit retentionRetained under legitimate interest as permitted by GDPR
Export account data
curl https://api.laserdata.cloud/account/export \
  -H "ld-api-key: YOUR_API_KEY"

Audit Log Retention

Basic7 days
Pro30 days
Enterprise365 days
Query audit logs
curl "https://audit.laserdata.cloud/audit/tenants/{tenant_id}?page=1&results=10" \
  -H "ld-api-key: YOUR_API_KEY"

# Response includes:
# type, name, author {id, name},
# division, environment, deployment,
# timestamp
laserdata.cloud
Console: Audit Log — environment, deployment, API key events with actor identity and correlation IDs

Retention

7 days free. Up to 365 days.

Every plan includes 7 days of telemetry retention at no extra cost. Extend up to 365 days; additional days beyond 7 are billed as a telemetry cost component. Configurable per deployment.

Data TypeConfigurableRange
MetricsPer deployment7–365 days
HeartbeatsPer deployment7–365 days
LogsPer deployment7–365 days

Required permission: deployment:telemetry:read (view) · deployment:telemetry:manage (configure retention)

Observability from day one.

Metrics, logs, heartbeats, alerts, and audit trails: all built in, no configuration required. Spin up a Free-tier deployment and see your infrastructure in full.

Join PreviewObservability Docs