Release 8.0.0¶

Breaking changes¶

Python 3.7 and 3.8 are no longer supported¶

As of version 8.0.0, Python 3.7 and 3.8 are no longer supported. The minimum required Python version is now 3.9.

If your add-on runs on Python 3.7 or 3.8, you must upgrade your environment before upgrading solnlib to 8.0.0.

New required dependencies¶

The observability module introduces the following new dependencies:

opentelemetry-api >=1.39.1,<2
opentelemetry-sdk >=1.39.1,<2
opentelemetry-exporter-otlp-proto-grpc >=1.39.1,<2
grpcio >=1.74.0

These are installed automatically when installing solnlib via pip.

New features¶

ObservabilityService¶

A new solnlib.observability module has been added. It provides OpenTelemetry metric instrumentation for Splunk modular inputs, with automatic integration with the Splunk Spotlight collector.

The module exposes two public components:

ObservabilityService — initializes a MeterProvider with two built-in event counters and attempts to connect to the local Spotlight OTLP collector. Falls back silently when the collector is not available, so observability failures never break the add-on.
LoggerMetricExporter — an OpenTelemetry MetricExporter that writes data points to a Python logger. Useful for local development and debugging.

Built-in counters:

Attribute	Metric name	Unit
`event_count_counter`	`splunk.addon.events`	`1`
`event_bytes_counter`	`splunk.addon.events.bytes`	`By`

Example usage:

import logging
from solnlib.observability import ObservabilityService, ATTR_MODINPUT_NAME

logger = logging.getLogger(__name__)

obs = ObservabilityService(
    modinput_type="my-input",
    logger=logger,
    ta_name="my_ta",
    ta_version="1.0.0",
)

# In your event collection loop:
if obs.event_count_counter:
    obs.event_count_counter.add(len(events), {ATTR_MODINPUT_NAME: stanza_name})
if obs.event_bytes_counter:
    obs.event_bytes_counter.add(total_bytes, {ATTR_MODINPUT_NAME: stanza_name})

When ta_name and ta_version are not provided, they are read automatically from app.conf via get_conf_stanzas. Pass them explicitly when running outside a full Splunk environment.

Additional instruments can be registered using ObservabilityService.register_instrument.

For full API reference see observability.py.