Commonly used Prometheus exporters and their recommended ports: Difference between revisions

From Jwiki
← Older edit
No edit summary
No edit summary
 
Line 1: Line 1:
= Prometheus Exporter Port Allocation Standard =
= Prometheus Exporter Port Allocation Standard =


This document defines the '''official internal standard''' for Prometheus exporter port assignments within our infrastructure. The primary goal is to create a cohesive, consistent, and maintainable port scheme that prevents conflicts, is easy to understand, and is simple to extend.
This document defines the '''official internal standard''' for Prometheus exporter port assignments within our infrastructure. The primary goal is to create a single, cohesive, and maintainable port scheme that prevents conflicts and is easy to understand and extend.


== Guiding Principles ==
== Guiding Principles ==
* '''Internal Consistency is Priority''': This list is our single source of truth. It overrides external community standards if they conflict with our internal scheme.
* '''One List, One Standard''': This page is the single source of truth. All services that expose a Prometheus metrics endpoint '''must''' be included in this list with a unique port.
* '''Uniqueness is Mandatory''': Every exporter in this list is assigned a unique port to prevent overlaps.
* '''Internal Consistency is Priority''': This list overrides external community standards or vendor defaults if they conflict with our internal, sequential scheme.
* '''Logical Grouping''': Ports are allocated in logical, contiguous blocks, making the scheme predictable and easy to manage.
* '''Logical Grouping''': Ports are allocated in logical, contiguous blocks, making the scheme predictable and easy to manage.
* '''Documentation First''': Before deploying a new exporter, it '''must''' be added to this list with an assigned port.
* '''Documentation First''': Before deploying a new exporter, it '''must''' be added to this list with an assigned port.
Line 11: Line 11:
----
----


== Special Case: Grafana Alloy ==
== Observability & Core Infrastructure (9100-9119) ==
Grafana Alloy is our primary telemetry collector and is treated as a special case due to its unique role and well-known default port.
This block covers the foundational layer of our stack, from the observability agent itself to the hosts and orchestration platforms.
 
* '''Role''': Alloy is configured to scrape other exporters on their assigned ports (defined below). It also receives OpenTelemetry data and tails log files.
* '''Management UI Port''': The Alloy web UI for debugging and graph visualization will always run on its default port, '''`12345`''' [11].
* '''Internal Metrics''': If Alloy is configured to expose its own internal metrics, it will be assigned a port from the standard block below, just like any other exporter.
 
----
 
== Core Infrastructure & Orchestration (9100-9109) ==
This block covers the foundational layer of our stack, from the hosts to the orchestration platforms.


{| class="wikitable"
{| class="wikitable"
Line 29: Line 20:
! Assigned Port
! Assigned Port
! Notes
! Notes
|-
| '''Grafana Alloy'''
| `9100`
| For exposing Alloy's '''own internal metrics''' to be scraped by Prometheus. Its separate management UI runs on port `12345`.
|-
|-
| '''Node Exporter'''
| '''Node Exporter'''
| `9100`
| `9101`
| Linux/Unix host metrics. The foundational Prometheus exporter.
| Linux/Unix host metrics.
|-
|-
| '''Windows Exporter'''
| '''Windows Exporter'''
| `9101`
| `9102`
| Windows host metrics.
| Windows host metrics.
|-
|-
| '''Blackbox Exporter'''
| '''Blackbox Exporter'''
| `9102`
| `9103`
| Probing endpoints (ICMP/HTTP/TCP) for availability.
| Probing endpoints (ICMP/HTTP/TCP) for availability.
|-
|-
| '''cAdvisor'''
| '''cAdvisor'''
| `9103`
| `9104`
| Container-level resource usage metrics.
| Container-level resource usage metrics.
|-
|-
| '''kube-state-metrics'''
| '''kube-state-metrics'''
| `9104`
| `9105`
| Kubernetes API object state metrics.
| Kubernetes API object state metrics.
|-
|-
| '''Proxmox VE Exporter'''
| '''Proxmox VE Exporter'''
| `9105`
| `9106`
| Proxmox hypervisor, VM, and container metrics.
| Proxmox hypervisor, VM, and container metrics.
|-
|-
| '''eBPF Exporters'''
| '''eBPF Exporters'''
| `9106`
| `9107`
| For advanced Linux kernel telemetry.
| For advanced Linux kernel telemetry.
|-
|-
| '''Process Exporter'''
| '''Process Exporter'''
| `9107`
| `9108`
| For detailed metrics on specific Linux processes.
| For detailed metrics on specific Linux processes.
|}
|}

Latest revision as of 16:22, 27 August 2025

Prometheus Exporter Port Allocation Standard

This document defines the official internal standard for Prometheus exporter port assignments within our infrastructure. The primary goal is to create a single, cohesive, and maintainable port scheme that prevents conflicts and is easy to understand and extend.

Guiding Principles

  • One List, One Standard: This page is the single source of truth. All services that expose a Prometheus metrics endpoint must be included in this list with a unique port.
  • Internal Consistency is Priority: This list overrides external community standards or vendor defaults if they conflict with our internal, sequential scheme.
  • Logical Grouping: Ports are allocated in logical, contiguous blocks, making the scheme predictable and easy to manage.
  • Documentation First: Before deploying a new exporter, it must be added to this list with an assigned port.

Observability & Core Infrastructure (9100-9119)

This block covers the foundational layer of our stack, from the observability agent itself to the hosts and orchestration platforms.

Core Infrastructure
Exporter Assigned Port Notes
Grafana Alloy `9100` For exposing Alloy's own internal metrics to be scraped by Prometheus. Its separate management UI runs on port `12345`.
Node Exporter `9101` Linux/Unix host metrics.
Windows Exporter `9102` Windows host metrics.
Blackbox Exporter `9103` Probing endpoints (ICMP/HTTP/TCP) for availability.
cAdvisor `9104` Container-level resource usage metrics.
kube-state-metrics `9105` Kubernetes API object state metrics.
Proxmox VE Exporter `9106` Proxmox hypervisor, VM, and container metrics.
eBPF Exporters `9107` For advanced Linux kernel telemetry.
Process Exporter `9108` For detailed metrics on specific Linux processes.

Databases & Caching (9120-9139)

Data stores, caches, message queues, and related proxies.

Databases & Caching
Exporter Assigned Port Notes
PostgreSQL Exporter `9120` For PostgreSQL databases.
MySQL Exporter `9121` For MySQL and MariaDB databases.
Redis Exporter `9122` For Redis in-memory databases.
MongoDB Exporter `9123` For MongoDB document databases.
Elasticsearch Exporter `9124` For Elasticsearch / OpenSearch clusters.
RabbitMQ Exporter `9125` For RabbitMQ message brokers.
Kafka Exporter `9126` For Apache Kafka cluster metrics.
Memcached Exporter `9127` For Memcached key-value stores.
Oracle DB Exporter `9128` For Oracle Database instances.
SQL Exporter `9129` For running generic SQL queries against various databases.
ClickHouse Exporter `9130` For ClickHouse columnar database.

Applications & Web Services (9140-9159)

Web servers, application servers, and specific software platforms.

Applications & Web Services
Exporter Assigned Port Notes
JMX Exporter `9140` For any JVM-based application (e.g., Tomcat, custom Java apps).
Apache Exporter `9141` For Apache HTTPD servers.
NGINX Exporter `9142` For NGINX web servers and reverse proxies.
HAProxy Exporter `9143` For HAProxy load balancers.
PHP-FPM Exporter `9144` For PHP FastCGI Process Manager.
GitLab `9145` For the main GitLab application metrics.
GitLab Runner `9146` For GitLab Runner agents.
Consul Exporter `9147` For HashiCorp Consul service mesh.
Nomad Exporter `9148` For HashiCorp Nomad cluster orchestrator.

Networking, Security & Miscellaneous (9160-9189)

Exporters for network devices, protocols, security tools, and other utilities.

Networking, Security & Miscellaneous
Exporter Assigned Port Notes
SNMP Exporter `9160` For network gear (switches, routers) and other hardware via SNMP.
Script Exporter `9161` For wrapping arbitrary shell scripts as a metrics endpoint.
CloudWatch Exporter `9162` For fetching metrics from AWS CloudWatch.
Azure Exporter `9163` For fetching metrics from Azure Monitor.
Stackdriver Exporter `9164` For fetching metrics from Google Cloud Platform.
BIND Exporter `9165` For BIND DNS server metrics.
WireGuard Exporter `9166` For WireGuard VPN tunnel metrics.
OpenVPN Exporter `9167` For OpenVPN server metrics.
DHCPD Leases Exporter `9168` For monitoring DHCP server leases.
Retrieved from "https://wiki.jandzsogyorgy.hu/index.php?title=Commonly_used_Prometheus_exporters_and_their_recommended_ports&oldid=290"