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

Google Cloud

The Google Cloud destination exports telemetry from a Bindplane pipeline to Google Cloud's observability suite. Metrics are sent to Cloud Monitoring, logs to Cloud Logging, and traces to Cloud Trace. Choose which signals to export with the Telemetry Type selector.

Supported Telemetry Types

Metrics
Logs
Traces

Prerequisites

You need a Google Cloud project and credentials authorized to write telemetry to it.

The following Google Cloud APIs must be enabled on the destination project:

  • Cloud Monitoring API (monitoring.googleapis.com)

  • Cloud Logging API (logging.googleapis.com)

  • Cloud Trace API (cloudtrace.googleapis.com)

The collector must be able to reach these endpoints over HTTPS on port 443. When the collector runs on a Google Cloud host, this access is typically available by default.

Credentials and IAM roles

The exporter authenticates using Google Application Default Credentials. When the collector runs on a Google Cloud host (Compute Engine, GKE), it can use the host's attached service account automatically. When the collector runs outside Google Cloud, create a service account and supply its key as JSON or a file path.

Grant the service account the roles matching the signals you export:

Signal
Role
Identifier

Metrics

Monitoring Metric Writer

roles/monitoring.metricWriter

Logs

Logs Writer

roles/logging.logWriter

Traces

Cloud Trace Agent

roles/cloudtrace.agent

Create and download a service account key following the Google Cloud documentation. The downloaded key is used when the Authentication Method is set to json or file.

Configuration

Bindplane docs - Google Cloud - image 1

General

Parameter
Type
Required
Default
Description

Choose Telemetry Type

Telemetry Selector

No

Logs, Metrics, Traces

Which signals to export.

Project ID

String

No

(empty)

The Google Cloud project ID to send telemetry to. Detected from the credentials if left empty.

Authentication Method

Enum: auto, json, file

Yes

auto

How the collector authenticates to Google Cloud.1

Credentials

String (sensitive)

Yes

(empty)

JSON contents of a Google service account credential file. Shown when Authentication Method is json.

Credentials File

String

Yes

(empty)

Path to a Google service account credential file on the collector. The collector's runtime user must be able to read it. Shown when Authentication Method is file.

  1. auto uses the collector's environment (Application Default Credentials), which works on Google Cloud hosts or when GOOGLE_APPLICATION_CREDENTIALS is set. json takes the service account key contents inline. file reads the key from a path on the collector.

Advanced

Parameter
Type
Required
Default
Description

Timeout

Duration

No

5s

Timeout for sending a batch to Google Cloud. Batches that cannot be sent within the timeout are dropped.

Enable Compression

Boolean

No

true

Compress telemetry with gzip before sending.

Metric Write-Ahead-Log

Boolean

No

false

Write metrics to a write-ahead-log before sending, to help avoid loss during network outages or collector restarts.

WAL Max Retry Duration

Integer (minutes)

No

60

Maximum time to retry sending metrics from the write-ahead-log after a network error. Shown when Metric Write-Ahead-Log is enabled.

Default Location

Enum (creatable)

No

us-east1

Region assigned to records (via cloud.region) that lack a region or availability zone. Any Google Cloud region is accepted.

Drop Raw Copy

Boolean

No

true

Drop the raw copy of the log stored in log.record.original.

Sending Queue

Parameter
Type
Required
Default
Description

Enable Sending Queue

Boolean

No

true

Buffer telemetry in memory before sending, to help avoid loss during a temporary network outage.

Number of Consumers

Integer

No

10

Number of consumers that dequeue batches. Shown when the sending queue is enabled.

Queue Size

Integer

No

5000

Maximum number of batches kept in memory before dropping. Shown when the sending queue is enabled.

Enable Persistent Queuing

Boolean

No

true

Buffer telemetry to disk before sending, to help avoid loss during network outages or collector restarts. Shown when the sending queue is enabled.

Persistent Queue Storage

Extension

Yes

File storage extension

Storage backend for the persistent queue. Defaults to a file storage extension at ${OIQ_OTEL_COLLECTOR_HOME}/storage. Shown when both the sending queue and persistent queuing are enabled.

Examples

Send all signals using an inline service account key (JSON)

This destination authenticates with the json method, embedding the service account key directly in the configuration. It exports metrics, logs, and traces to the project bindplane-gcp. Replace the credentials value with your own service account key.

Configuration Tips

  • On Google Cloud hosts, leave Authentication Method on auto and Project ID empty. The exporter resolves both from the attached service account and instance metadata.

  • Set Default Location to the region closest to your collectors so records without an explicit cloud.region are not assigned the us-east1 default.

  • Keep the sending queue and persistent queuing enabled in production so telemetry survives transient network failures and collector restarts.

Troubleshooting

Permission denied or authentication errors

Symptoms: the collector logs PermissionDenied, 403, or credential errors and no telemetry reaches Google Cloud.

Solutions:

  1. Confirm the service account has the role for each exported signal: roles/monitoring.metricWriter for metrics, roles/logging.logWriter for logs, roles/cloudtrace.agent for traces.

  2. Verify the target API (Cloud Monitoring, Cloud Logging, or Cloud Trace) is enabled on the project.

  3. When using file, confirm the collector's runtime user can read the credentials file.

Telemetry sent to the wrong or no project

Symptoms: data does not appear in the expected project.

Solutions:

  1. Set Project ID explicitly rather than relying on detection from credentials.

  2. Confirm the service account belongs to the intended project.

Dropped data during network outages

Symptoms: gaps in metrics or logs after a network interruption or collector restart.

Solutions:

  1. Enable the sending queue and persistent queuing so batches buffer to disk instead of dropping.

  2. For metrics, enable the Metric Write-Ahead-Log to retry sending after the network recovers.

Standalone Destination

Last updated

Was this helpful?