Pino transport for OpenTelemetry. Outputs logs in the OpenTelemetry Log Data Model and sends them to an OTLP logs collector.
npm i pino-opentelemetry-transport
can be set to http/protobuf
, grpc
, http
or console
by using
- env var
OTEL_EXPORTER_OTLP_PROTOCOL
- env var
OTEL_EXPORTER_OTLP_LOGS_PROTOCOL
- setting the exporterProtocol option
Settings configured programmatically take precedence over environment variables. Per-signal environment variables take precedence over non-per-signal environment variables. This principle applies to all the configurations in this module.
If no protocol is specified, http/protobuf
is used as a default.
Set either of the following environment variables:
OTEL_EXPORTER_OTLP_LOGS_ENDPOINT
,
OTEL_EXPORTER_OTLP_ENDPOINT
Environment Variable Configuration
If batch log processor is selected (is default), it can be configured using env vars described in the OpenTelemetry specification
When using the transport, the following options can be used to configure the transport programmatically:
loggerName
: name to be used by the OpenTelemetry loggerserviceVersion
: version to be used by the OpenTelemetry loggerseverityNumberMap
: Object mapping Pino log level numbers to OpenTelemetry log severity numbers. This is an override for adding custom log levels and changing default log levels. Undefined default Pino log levels will still be mapped to their default OpenTelemetry log severity. OptionalresourceAttributes
: Object containing resource attributes. OptionallogRecordProcessorOptions
: a single object or an array of objects specifying the LogProcessor and LogExporter types and constructor params. Optional
Make sure you have access to an OTEL collector.
To start quickly, create a minimal configuration for OTEL collector in the otel-collector-config.yaml
file:
receivers:
otlp:
protocols:
grpc:
exporters:
file:
path: ./etc/test-logs/otlp-logs.log
flush_interval: 1
logging:
verbosity: basic
processors:
batch:
service:
pipelines:
logs:
receivers: [otlp]
processors: []
exporters: [logging, file]
The collector can then be ran with:
docker run --volume=$(pwd)/otel-collector-config.yaml:/etc/otel-collector-config.yaml:rw --volume=/tmp/test-logs:/etc/test-logs:rw -p 4317:4317 -d otel/opentelemetry-collector-contrib:latest --config=/etc/otel-collector-config.yaml
Create an index.js file containing
const pino = require('pino')
const transport = pino.transport({
target: 'pino-opentelemetry-transport'
})
const logger = pino(transport)
transport.on('ready', () => {
setInterval(() => {
logger.info('test log')
}, 1000)
})
Install Pino and pino-opentelemetry-transport
npm install pino pino-opentelemetry-transport
Run the service setting the OTEL_EXPORTER_OTLP_LOGS_ENDPOINT
and OTEL_RESOURCE_ATTRIBUTES
env vars
OTEL_EXPORTER_OTLP_LOGS_PROTOCOL='grpc' OTEL_EXPORTER_OTLP_LOGS_ENDPOINT=http://localhost:4317 OTEL_RESOURCE_ATTRIBUTES="service.name=my-service,service.version=1.2.3" node index.js
- Minimalistic
- HTTP Server with trace context propagation
- Sending logs to Grafana Loki
- Using Multiple Record Processors
- TypeScript
Run the OTLP collector in a container
npm run docker-run
Run an example
node examples/minimalistic/minimalistic.js
Observe the logs
tail -f /tmp/test-logs/otlp-logs.log
Note that not all log entries will immediately be written to the otlp-logs.log
file. The collector will flush to the disk eventually. The flush will be forced if the collector receives a kill signal.
This project is kindly sponsored by:
MIT