Prometheus

The Prometheus destination exposes the metrics flowing through a Bindplane pipeline on an HTTP endpoint in the Prometheus exposition format. Unlike push-based destinations, it does not send data anywhere: it opens a listen address and port, and a Prometheus server scrapes that endpoint on its own interval. This destination handles metrics only.

Supported Telemetry

Metrics	Logs	Traces
✓

Prerequisites

You need a Prometheus server configured to scrape the collector's exposed endpoint. Add a job to the server's scrape_configs whose target matches the Listen Address and Listen Port set on this destination. The endpoint serves metrics on the /metrics path.

Example scrape job pointing at a collector exposing metrics on port 9000:

scrape_configs:
  - job_name: bindplane-collector
    static_configs:
      - targets: ['collector-host:9000']

The listen address must be reachable from the Prometheus server. The default 127.0.0.1 only accepts connections from the same host, so set it to 0.0.0.0 (or a specific routable address) when Prometheus runs elsewhere. See the Prometheus scrape configuration reference for the full set of scrape options.

Configuration

Metrics

Parameter	Type	Required	Default	Description
Listen Address	String	No	127.0.0.1	The IP address the Prometheus exporter listens on, to be scraped by a Prometheus server. Use 0.0.0.0 to accept connections from other hosts.
Listen Port	Integer	No	9000	The TCP port the Prometheus exporter listens on. Valid range is 1 to 65535.

Advanced

Parameter	Type	Required	Default	Description
Namespace	String	No	(empty)	When set, exports metrics under the provided namespace, prefixing each metric name.
Add Unit and Type Suffixes	Boolean	No	false	When enabled, appends unit and type suffixes to metric names to follow Prometheus naming conventions.
Metric Expiration	Duration	No	5m	The time after which a metric is considered stale and is no longer exported.
Convert Resource Attributes to Metric Attributes	Boolean	No	true	When enabled, converts all resource attributes into metric labels.

TLS

Parameter	Type	Required	Default	Description
Enable TLS	Boolean	No	false	Serve the scrape endpoint over TLS. Reveals the certificate and key fields below.
TLS Server Certificate File	String	No	(empty)	Path to the TLS certificate file. Shown when Enable TLS is on.
TLS Server Private Key File	String	No	(empty)	Path to the TLS private key file. Shown when Enable TLS is on.
Mutual TLS	Boolean	No	false	Require and verify client certificates (mTLS). Shown when Enable TLS is on.
TLS Certificate Authority File for mTLS	String	No	(empty)	Path to the certificate authority file used to validate client certificates. Shown when both Enable TLS and Mutual TLS are on.

Examples

Expose metrics for a remote Prometheus server with mTLS

This destination listens on all interfaces so a Prometheus server on another host can reach it, and requires client certificates so only that authorized server can scrape. The Prometheus scrape job must present a matching client certificate signed by the configured CA.

apiVersion: bindplane.observiq.com/v1
kind: Destination
metadata:
  name: prometheus
spec:
  type: prometheus
  parameters:
    - name: listen_address
      value: 0.0.0.0
    - name: listen_port
      value: 9000
    - name: configure_tls
      value: true
    - name: cert_file
      value: /etc/otel/certs/server.crt
    - name: key_file
      value: /etc/otel/certs/server.key
    - name: mutual_tls
      value: true
    - name: client_ca_file
      value: /etc/otel/certs/ca.crt

Configuration Tips

• Leave Listen Address at 127.0.0.1 only when the Prometheus server runs on the same host as the collector. For a remote server, set 0.0.0.0 or a routable address, and confirm the port is open through any firewall.
• Metric Expiration controls how long a metric stays scrapable after the collector last received it. Raise it if your scrape interval is longer than the default 5 minutes so metrics are not dropped between scrapes.
• Enable Add Unit and Type Suffixes when you want metric names that match Prometheus naming conventions (for example a _seconds or _total suffix). Leave it off to keep names closer to their OpenTelemetry originals.

Troubleshooting

Prometheus shows the target as down

Symptoms: the scrape target reports a connection refused or timeout in the Prometheus targets page.

Solutions:

1. Confirm the scrape target host and port in scrape_configs match this destination's Listen Address and Listen Port.
2. If Prometheus runs on a different host, change Listen Address from 127.0.0.1 to 0.0.0.0 or a routable address. 127.0.0.1 only accepts local connections.
3. Verify no firewall or security group is blocking the listen port between Prometheus and the collector.

Metrics disappear between scrapes

Symptoms: series appear and then go stale before Prometheus scrapes them again.

Solutions:

1. Increase Metric Expiration so it exceeds the Prometheus scrape interval.
2. Confirm the source pipeline is still producing the metrics; the exporter only serves metrics it received within the expiration window.

TLS handshake or client certificate errors

Symptoms: Prometheus logs a TLS handshake failure or a certificate verification error when scraping.

Solutions:

1. Confirm the certificate and key paths exist on the collector host and are readable by the collector process.
2. When Mutual TLS is enabled, ensure the Prometheus scrape config presents a client certificate signed by the CA set in TLS Certificate Authority File for mTLS.

Standalone Destination

apiVersion: bindplane.observiq.com/v1
kind: Destination
metadata:
  name: prometheus
spec:
  type: prometheus
  parameters:
    - name: listen_address
      value: 0.0.0.0
    - name: listen_port
      value: 9000

Related Resources

• OpenTelemetry Prometheus Exporter
• Prometheus scrape configuration reference