Accel-PPP Prometheus Exporter

A Prometheus exporter for Accel-PPP that exposes metrics from the accel-cmd show stat command.

Some small story behind this exporter on my blog.

Minimum Requirements

• Go 1.24 or higher (that what I'm using, most likely is compatible down to Go 1.20)
• Accel-PPP 1.12 or compatible version
• Permission to execute accel-cmd show stat command

Note

This exporter has been tested with accel-ppp:

• 1.12

Features

• Exposes all metrics from accel-cmd show stat as Prometheus metrics (have to have access rights to issue command)
• Configurable listen address (default: :9101)
• Configurable metrics path (default: /metrics)
• Configurable path to accel-cmd binary
• Ready-to-use Grafana dashboard

Installation

Prebuilt Binaries

Download the latest release from the Releases page.

Docker

docker run -d --name accel-exporter -p 9101:9101 taihen/accel-exporter:latest

From Source

go install github.com/taihen/accel-exporter/cmd/accel-exporter@latest

Usage

Usage of accel-exporter:
  -accel-cmd.path string
        Path to accel-cmd binary (default "accel-cmd")
  -log.level string
        Log level (debug, info, warn, error) (default "info")
  -web.listen-address string
        Address to listen on (default ":9101")
  -web.metrics-path string
        Path to expose metrics (default "/metrics")

You can also configure the exporter using environment variables:

• ACCEL_EXPORTER_PORT: The port to listen on (overrides -web.listen-address)

Prometheus Configuration

Add a scrape configuration to your prometheus.yml:

scrape_configs:
  - job_name: 'accel-ppp'
    static_configs:
      - targets: ['localhost:9101']

Grafana Dashboard

A Grafana dashboard is included in the dashboards directory. You can import it into your Grafana instance.

Building from Source

# Clone the repository
git clone https://github.com/taihen/accel-exporter.git
cd accel-exporter

# Build
go build -o accel-exporter ./cmd/accel-exporter

# Run
./accel-exporter

Docker Build

docker build -t accel-exporter .
docker run -p 9101:9101 accel-exporter

Metrics

The exporter exposes the following metrics:

General:

• accel_exporter_build_info{version, commit, date}: Metric with constant '1' value
  labeled with version, commit, and build date
• accel_up: Was the last accel-cmd scrape successful (1 = yes, 0 = no).
• accel_scrape_failures_total: Number of errors while scraping accel-cmd.
• accel_uptime_seconds: Uptime of accel-ppp in seconds.
• accel_cpu_usage_percent: CPU usage percentage.
• accel_memory_rss_bytes: RSS memory usage in bytes.
• accel_memory_virtual_bytes: Virtual memory usage in bytes.

Core:

• accel_core_mempool_allocated_bytes: Allocated memory pool size
• accel_core_mempool_available_bytes: Available memory pool size
• accel_core_thread_count: Number of core threads
• accel_core_thread_active: Number of active core threads
• accel_core_context_count: Number of core contexts
• accel_core_context_sleeping: Number of sleeping core contexts
• accel_core_context_pending: Number of pending core contexts
• accel_core_md_handler_count: Number of MD handlers
• accel_core_md_handler_pending: Pending MD handlers count
• accel_core_timer_count: Number of core timers
• accel_core_timer_pending: Number of pending core timers

Sessions:

• accel_sessions_starting: Number of sessions starting
• accel_sessions_active: Number of active sessions
• accel_sessions_finishing: Number of sessions finishing

PPPoE:

• accel_pppoe_starting: Number of PPPoE sessions starting
• accel_pppoe_active: Number of active PPPoE sessions
• accel_pppoe_delayed_pado: Number of delayed PADO packets
• accel_pppoe_recv_padi_total: Total received PADI packets
• accel_pppoe_drop_padi_total: Total dropped PADI packets
• accel_pppoe_sent_pado_total: Total sent PADO packets
• accel_pppoe_recv_padr_total: Total received PADR packets
• accel_pppoe_recv_padr_dup_total: Total received duplicate PADR packets
• accel_pppoe_sent_pads_total: Total sent PADS packets
• accel_pppoe_filtered_total: Total filtered PPPoE packets

RADIUS (Labels: server_id, server_ip):

• accel_radius_state: State of RADIUS server (1 = active, 0 = inactive)
• accel_radius_fail_count_total: Total RADIUS server fail count
• accel_radius_request_count: Current RADIUS server request count
• accel_radius_queue_length: Current RADIUS server queue length
• accel_radius_auth_sent_total: Total RADIUS auth packets sent
• accel_radius_auth_lost_total: Total RADIUS auth packets lost
• accel_radius_auth_lost_5m: RADIUS auth packets lost (5m window)
• accel_radius_auth_lost_1m: RADIUS auth packets lost (1m window)
• accel_radius_auth_avg_time_5m_seconds: Average auth response time (5m)
• accel_radius_auth_avg_time_1m_seconds: Average auth response time (1m)
• accel_radius_acct_sent_total: Total accounting packets sent
• accel_radius_acct_lost_total: Total accounting packets lost
• accel_radius_acct_lost_5m: Accounting packets lost (5m window)
• accel_radius_acct_lost_1m: Accounting packets lost (1m window)
• accel_radius_acct_avg_time_5m_seconds: Average accounting response (5m)
• accel_radius_acct_avg_time_1m_seconds: Average accounting response (1m)
• accel_radius_interim_sent_total: Total interim accounting packets sent
• accel_radius_interim_lost_total: Total interim accounting packets lost
• accel_radius_interim_lost_5m: Interim packets lost (5m window)
• accel_radius_interim_lost_1m: Interim packets lost (1m window)
• accel_radius_interim_avg_time_5m_seconds: Avg interim response (5m)
• accel_radius_interim_avg_time_1m_seconds: Avg interim response (1m)

License

MIT