This is the Dash0 OpenTelemetry distribution for Node.js. It is primarily intended to be used by the Dash0 Kubernetes operator to instrument Node.js workloads with OpenTelemetry.
If no service name has been set, a service name and a service version are automatically derived by reading the main package.json file
(if it is present):
- The
service.nameresource attribute will be set to the value of thenameattribute found in thepackage.jsonfile. - The
service.versionresource attribute will be set to the value of theversionattribute found in thepackage.jsonfile.
This behavior can be disabled either
- by setting an explicit service name using the
OTEL_SERVICE_NAMEenvironment variable, - by setting an explicit service name using the
OTEL_RESOURCE_ATTRIBUTESenvironment variable (using theservice.nameattribute key), or - by setting the environment variable
DASH0_AUTOMATIC_SERVICE_NAME=false.
If set to a non-empty string, the distribution will create a span immediately at startup with the span name set to the value of DASH0_BOOTSTRAP_SPAN.
Additional debug logs can be enabled by setting DASH0_DEBUG=true.
If DASH0_DEBUG_PRINT_SPANS=true is set, all spans are printed to stdout via the
ConsoleSpanExporter.
Disables the Dash0 Node.js distribution entirely.
By default, the instrumentation plug-in @opentelemetry/instrumentation-fs is disabled. Set DASH0_ENABLE_FS_INSTRUMENTATION=true to enable spans for file system access.
If DASH0_FLUSH_ON_SIGTERM_SIGINT=true is set, the Dash0 Node.js distribution will install a handler for SIGTERM and
SIGINT that will shutdown the OpenTelemetry SDK gracefully when one of these signals is received.
The SDK shutdown is timeboxed to 500 milliseconds.
The signal handler will call process.exit(0) after the SDK's shutdown has completed, or after the 500 millisecond
timeout, whichever happens sooner.
This option can be helpful if you care about telemetry that is being produced shortly before the process terminates.
This option must not be used if the application under monitoring has its own handler for SIGTERM or SIGINT, because
Dash0's handler (and in particular the necessary process.exit(0) call) might interfere with the application's own
signal handler.
By default, the Dash0 Node.js distribution will install a hook that will shutdown the OpenTelemetry SDK gracefully when
the Node.js runtime is about to exit because the event loop is empty.
This can be disabled by setting DASH0_FLUSH_ON_EMPTY_EVENT_LOOP=false.
The SDK shutdown is timeboxed to 500 milliseconds.
This hook can be helpful if you care about telemetry that is being produced shortly before the process
exits.
Disabling it can be useful if you care about the process terminating as quickly as possible when the event loop is
empty.
In contrast to the handlers for SIGTERM/SIGINT (see above), this hook will not call process.exit (since the Node.js
runtime will exit on its own anyway).
The base URL of the OpenTelemetry collector that the distribution will send data to.
It defaults to http://dash0-operator-opentelemetry-collector.dash0-operator-system.svc.cluster.local:4318.
By default, all
supported instrumentations
are enabled (with the exception of @opentelemetry/instrumentation-fs), but you can use the environment variable
OTEL_NODE_ENABLED_INSTRUMENTATIONS to enable only certain instrumentations by providing a comma-separated list of the
instrumentation package names without the @opentelemetry/instrumentation- prefix.
For example, to enable only
@opentelemetry/instrumentation-http
and @opentelemetry/instrumentation-nestjs-core
instrumentations, set OTEL_NODE_ENABLED_INSTRUMENTATIONS="http,nestjs-core".
See https://github.com/open-telemetry/opentelemetry-js-contrib/blob/main/metapackages/auto-instrumentations-node/README.md#usage-auto-instrumentation for more information.