For the complete documentation index, see llms.txt. This page is also available as Markdown.

OpenTelemetry (OTLP)

The OpenTelemetry (OTLP) source receives metrics, logs, and traces sent in OpenTelemetry Protocol (OTLP) format. The collector runs an OTLP server that accepts data over gRPC (default port 4317) and HTTP (default port 4318). Any OTLP exporter, SDK, or other OpenTelemetry collector can send to it.

Supported Telemetry Types

Platform
Metrics
Logs
Traces

Linux

macOS

Windows

Kubernetes Node (DaemonSet)

Kubernetes Gateway

OpenShift Node (DaemonSet)

OpenShift Gateway

Azure Container Apps (Gateway)

AIX

Prerequisites

This is a receive-only source, so the only requirement is a reachable listen address and open ports on the collector host.

  • The collector must be able to bind the configured listen address and gRPC/HTTP ports. The defaults are 0.0.0.0:4317 (gRPC) and 0.0.0.0:4318 (HTTP).

  • The gRPC port and HTTP port must be different values.

  • Any firewall, security group, or Kubernetes Service in front of the collector must allow the senders to reach those ports.

  • If you enable TLS, you need a server certificate and private key on the collector host. For mutual TLS you also need a certificate authority file to validate client certificates.

Configuration

Bindplane docs - OpenTelemetry (OTLP) - image 1

General

Parameter
Type
Required
Default
Description

Choose Telemetry Type

telemetrySelector

No

["Logs", "Metrics", "Traces"]

Which signals the source receives. Valid values are Logs, Metrics, and Traces.

Listen Address

string

No

0.0.0.0

The IP address to listen on.

GRPC Port

int

No

4317

TCP port to receive OTLP telemetry using the gRPC protocol. The port used must not be the same as the HTTP port. Set to 0 to disable. Range 0-65535.

HTTP Port

int

No

4318

TCP port to receive OTLP telemetry using the HTTP protocol. The port used must not be the same as the gRPC port. Set to 0 to disable. Range 0-65535.

Advanced

Parameter
Type
Required
Default
Description

Maximum Message Size

int

No

20

Sets the maximum size (in MiB) of messages accepted by the server. Visible when GRPC Port is not 0.

Include Metadata

bool

No

false

Propagates the incoming connection's metadata to downstream consumers.

TLS

Parameter
Type
Required
Default
Description

Enable TLS

bool

No

false

Whether or not to use TLS.

Server Certificate File

string

Yes

(empty)

A path to the server certificate to be used for TLS. Required and visible when Enable TLS is true.

Server Private Key

string

Yes

(empty)

A path to the server private key to be used for TLS. Required and visible when Enable TLS is true.

Mutual TLS

bool

No

false

Whether or not to require client TLS authentication (mTLS). Visible when Enable TLS is true.

TLS Certificate Authority File

string

Yes

(empty)

A path to the certificate authority to use for authenticating client certificates. Required and visible when Enable TLS and Mutual TLS are both true.

GRPC Timeout

Parameter
Type
Required
Default
Description

Enable GRPC Timeout

bool

No

true

Enable to configure advanced GRPC timeout behavior. Visible when GRPC Port is not 0.

Max Idle Time

int

No

60

Maximum amount of time (seconds) after which an idle connection would be closed by sending a GoAway. Visible when GRPC Port is not 0 and Enable GRPC Timeout is true.

Max Connection Age

int

No

60

Maximum amount of time (seconds) a connection may exist before it will be closed by sending a GoAway. Visible when GRPC Port is not 0 and Enable GRPC Timeout is true.

Max Connection Age Grace

int

No

300

Additive period after Max Connection Age after which the connection will be forcibly closed. Visible when GRPC Port is not 0 and Enable GRPC Timeout is true.

CORS

Parameter
Type
Required
Default
Description

Configure CORS

bool

No

false

Enable Cross-Origin Resource Sharing support for HTTP requests. Visible when HTTP Port is not 0.

CORS Allowed Origins

strings

No

(empty)

A list of origins a cross-domain request can be executed from. Visible when Configure CORS is true and HTTP Port is not 0.

CORS Allowed Headers

strings

No

(empty)

The allowed headers sets what headers will be allowed in CORS requests. Visible when Configure CORS is true and HTTP Port is not 0.

CORS Max Age

int

No

0

Indicates how long (in seconds) the results of a preflight request can be cached. Visible when Configure CORS is true and HTTP Port is not 0.

Examples

Receive all three signals with defaults

The OTLP source has no required fields. With the defaults it listens on 0.0.0.0:4317 (gRPC) and 0.0.0.0:4318 (HTTP) for logs, metrics, and traces, without TLS.

Kubernetes: DaemonSet vs. Gateway

The OTLP source works on Node (DaemonSet) and Gateway collectors. Applications in the cluster send telemetry to the collectors through their clusterIP Services. The source needs no extra configuration; attach it to the collector type you want to receive on.

Use these endpoints to send to the managed Node (DaemonSet) collectors:

Protocol
Service
Endpoint

gRPC

clusterIP

bindplane-node-agent.bindplane-agent.svc.cluster.local:4317

gRPC

headless clusterIP

bindplane-node-agent-headless.bindplane-agent.svc.cluster.local:4317

HTTP

clusterIP

http://bindplane-node-agent.bindplane-agent.svc.cluster.local:4318

Use these endpoints to send to the managed Gateway collectors:

Protocol
Service
Endpoint

gRPC

clusterIP

bindplane-gateway-agent.bindplane-agent.svc.cluster.local:4317

gRPC

headless clusterIP

bindplane-gateway-agent-headless.bindplane-agent.svc.cluster.local:4317

HTTP

clusterIP

http://bindplane-gateway-agent.bindplane-agent.svc.cluster.local:4318

A DaemonSet collector runs one pod per node, so an application sends to the collector on its own node (typically through the node-local Service). This keeps the network hop local and tags telemetry with node-level context, but it scales with the number of nodes and competes for node resources with your workloads.

A Gateway collector runs as a separate scalable Deployment that all applications send to over the cluster Service. It scales independently of cluster size and centralizes processing, at the cost of an extra network hop and no automatic node-level association. Prefer the Gateway when DaemonSet resource consumption per node is a concern or when you want centralized, independently scaled ingestion. A common pattern is both: DaemonSet collectors enrich and forward node-local data to a Gateway tier.

Receive over gRPC only with mutual TLS

Disable the HTTP listener by setting its port to 0, then enable TLS and mutual TLS so only clients presenting a valid certificate are accepted.

Configuration Tips

  • The gRPC and HTTP ports must differ. Set a protocol's port to 0 to disable that protocol entirely (for example HTTP Port 0 for a gRPC-only receiver).

  • Enabling Mutual TLS requires Enable TLS to be on, and the CA file is then required so the server can validate client certificates. Enable mTLS on both this source and the sending exporter together, or connections will be rejected.

  • CORS settings only apply to the HTTP listener. They have no effect when the HTTP port is disabled.

Troubleshooting

Collector fails to start or bind the port

Symptoms: The collector logs an "address already in use" or bind error, or no data arrives.

Solutions:

  1. Confirm the gRPC and HTTP ports are different. Setting both to the same value is invalid.

  2. Check that no other process or another OTLP source is already bound to the same listen address and port.

  3. Verify the listen address is reachable. 0.0.0.0 binds all interfaces; a specific IP must exist on the host.

Senders cannot connect or are rejected

Symptoms: Exporters report connection refused, TLS handshake failures, or certificate errors.

Solutions:

  1. Ensure firewalls, security groups, and Kubernetes Services forward the gRPC/HTTP ports to the collector.

  2. When TLS is enabled, confirm the sender trusts the server certificate and is using the correct scheme (https/secure gRPC).

  3. When Mutual TLS is enabled, confirm the sender presents a client certificate signed by the configured CA file.

Large payloads are dropped over gRPC

Symptoms: Big batches fail over gRPC while smaller ones succeed, often with a message-size error.

Solutions:

  1. Raise Maximum Message Size (MiB) on this source to accommodate the largest expected batch.

  2. Reduce batch size on the sending exporter so individual messages stay under the limit.

Standalone Source

Last updated

Was this helpful?