README.md

TelemetryFlow Agent Integrations

This document provides an overview of all third-party integrations supported by the TelemetryFlow Agent.

Integration Architecture

flowchart TB
    subgraph "TelemetryFlow Agent"
        IM[Integration Manager]

        subgraph "Cloud Providers"
            GCP[GCP Exporter]
            AZURE[Azure Exporter]
            ALIBABA[Alibaba Exporter]
        end

        subgraph "Infrastructure"
            PROXMOX[Proxmox Exporter]
            VMWARE[VMware Exporter]
            NUTANIX[Nutanix Exporter]
            AZUREARC[Azure Arc Exporter]
        end

        subgraph "Network & IoT"
            CISCO[Cisco Exporter]
            SNMP[SNMP Exporter]
            MQTT[MQTT Exporter]
        end

        subgraph "Kernel/System"
            EBPF[eBPF Exporter]
        end

        subgraph "APM Platforms"
            DD[Datadog]
            NR[New Relic]
            DT[Dynatrace]
            INST[IBM Instana]
        end

        subgraph "Observability Backends"
            PROM[Prometheus]
            SPLUNK[Splunk]
            ES[Elasticsearch]
            INFLUX[InfluxDB]
            KAFKA[Kafka]
            LOKI[Loki]
            JAEGER[Jaeger]
            ME[ManageEngine]
        end
    end

    IM --> GCP & AZURE & ALIBABA
    IM --> PROXMOX & VMWARE & NUTANIX & AZUREARC
    IM --> CISCO & SNMP & MQTT
    IM --> EBPF
    IM --> DD & NR & DT & INST
    IM --> PROM & SPLUNK & ES & INFLUX & KAFKA & LOKI & JAEGER & ME

Loading

Integration Categories

Category	Integrations	Description
Cloud Providers	GCP, Azure, Alibaba	Major cloud platform integrations
Infrastructure	Proxmox, VMware, Nutanix, Azure Arc	Virtualization and hybrid cloud
Network & IoT	Cisco, SNMP, MQTT	Network devices and IoT messaging
Kernel/System	eBPF	Linux kernel-level observability
Observability	Prometheus, Datadog, Dynatrace, Instana, ManageEngine, Splunk, etc.	APM & monitoring backends
Prometheus Exporter	Prometheus /metrics endpoint	Prometheus-compatible scrape endpoint
Kubernetes Collector	Kubernetes resource state	Drop-in kube-state-metrics replacement
Node Exporter	System metrics (CPU, Memory, Disk, Network, etc.)	Drop-in node_exporter replacement

Data Flow

sequenceDiagram
    participant S as Data Source
    participant E as Exporter
    participant M as Manager
    participant B as Backend

    S->>E: Raw Data
    E->>E: Transform
    E->>E: Validate
    E->>M: Telemetry Data
    M->>M: Batch & Buffer
    M->>B: Export
    B-->>M: Ack/Error
    M-->>E: Result

Loading

Integration Interface

All integrations implement the Exporter interface:

type Exporter interface {
    Name() string
    Type() string
    IsEnabled() bool
    SupportedDataTypes() []DataType
    Init(ctx context.Context) error
    Validate() error
    Export(ctx context.Context, data *TelemetryData) (*ExportResult, error)
    ExportMetrics(ctx context.Context, metrics []Metric) (*ExportResult, error)
    ExportTraces(ctx context.Context, traces []Trace) (*ExportResult, error)
    ExportLogs(ctx context.Context, logs []LogEntry) (*ExportResult, error)
    Health(ctx context.Context) (*HealthStatus, error)
    Close(ctx context.Context) error
}

Supported Data Types

graph LR
    subgraph "Telemetry Data Types"
        M[Metrics]
        L[Logs]
        T[Traces]
    end

    subgraph "Exporters"
        E1[Full Support<br/>Metrics + Logs + Traces]
        E2[Metrics Only]
        E3[Logs Only]
    end

    M --> E1 & E2
    L --> E1 & E3
    T --> E1

Loading

Integration	Metrics	Logs	Traces
GCP	✅	✅	✅
Azure	✅	✅	✅
Alibaba	✅	✅	✅
Proxmox	✅	❌	❌
VMware	✅	❌	❌
Nutanix	✅	❌	❌
Azure Arc	✅	❌	❌
Cisco	✅	❌	❌
SNMP	✅	❌	❌
MQTT	✅	✅	✅
eBPF	✅	❌	❌
Dynatrace	✅	✅	✅
IBM Instana	✅	✅	✅
ManageEngine	✅	✅	❌

Configuration

All integrations are configured in the integrations section of tfo-agent.yaml:

integrations:
  # Cloud Providers
  gcp:
    enabled: false
    project_id: "${GCP_PROJECT_ID}"

  azure:
    enabled: false
    subscription_id: "${AZURE_SUBSCRIPTION_ID}"

  # Infrastructure
  proxmox:
    enabled: false
    api_url: "https://proxmox:8006"

  vmware:
    enabled: false
    vcenter_url: "https://vcenter.example.com"

  # Network
  cisco:
    enabled: false
    api_type: dnac

  snmp:
    enabled: false
    version: v2c

  mqtt:
    enabled: false
    broker: "tcp://mqtt:1883"

  # System
  ebpf:
    enabled: false

Health Monitoring

Each integration reports health status:

stateDiagram-v2
    [*] --> Disabled
    [*] --> Initializing
    Initializing --> Healthy: Init Success
    Initializing --> Unhealthy: Init Failed
    Healthy --> Unhealthy: Connection Lost
    Unhealthy --> Healthy: Reconnected
    Healthy --> [*]: Close
    Unhealthy --> [*]: Close

Loading

Quick Links

• Cloud Providers - GCP, Azure, Alibaba Cloud
• Infrastructure - Proxmox, VMware, Nutanix, Azure Arc
• Network & IoT - Cisco, SNMP, MQTT
• Kernel/System - eBPF observability
• Observability Backends - Prometheus, Datadog, etc.
• Prometheus Exporter - Prometheus /metrics scrape endpoint
• Kubernetes Collector - Kubernetes resource state metrics
• Node Exporter - Drop-in replacement for prometheus/node_exporter

---

Copyright (c) 2024-2026 Telemetri Data Indonesia. All rights reserved.