Examples of anaconda-opentelemetry usage¶

These examples serve as an onboarding guide and as further explanations to main documentation.

Configuration¶

Creating Configuration¶

try:
    config = Configuration(default_endpoint='example.com:4317', default_auth_token=token)
    config.set_metrics_export_interval_ms(10000).set_logging_level('warn')
except:
    logging.warning("Telemetry `Configuration` failed to be created")

Specific Endpoint for Signal¶

config.set_metrics_endpoint(
    endpoint='http://localhost:4318',
    auth_token=token,
    cert_ca_file='./ca_cert.cer'
)

Optional Session Entropy¶

entropy_value = time.time()  # current unix time - simple form of entropy
try:
    config = Configuration(default_endpoint='example.com:4317').set_tracing_session_entropy(entropy_value)
except:
    # code here

Disabling Session ID¶

config = Configuration(default_endpoint='example.com:4317').set_disable_session_id(True)

ResourceAttributes¶

service_name, service_version = "service-a", "v1"
try:
    attributes = ResourceAttributes(service_name, service_version)
    attributes.set_attributes(environment="production", platform="aws")
except:
    logging.warning(f"Telemetry `ResourceAttributes` failed to be created")

Adding more attributes¶

try:
    attributes = ResourceAttributes(service_name, service_version)
    attributes.set_attributes(random_attribute_key="test1", random_attribute_key2="test2")
except:
    logging.warning(f"Telemetry `ResourceAttributes` failed to be created")

Initializing Telemetry¶

from  import *

attributes = ResourceAttributes("service-a", "v1")
try:
  initialize_telemetry(
      config=config,
      attributes=attributes
  )
except:
  logging.warning(f"Telemetry failed to be initialized.")

Optional Signal Streams¶

Try except omitted in this example:

initialize_telemetry(
    config=config,
    attributes=attributes,
    signal_types=["tracing", "metrics"]
)

Recording Telemetry¶

Logs¶

Simple Event Logging¶

send_event("Log body", "event_name", attributes=attrs)

Developer logs¶

log = logging.getLogger("your_logger_name_here")
try:
    log.addHandler(get_telemetry_logger_handler())
except:
  log.warning(f"OpenTelemetry logger failed to be initialized.")

¶

Metrics¶

These functions do not need additional error handling. They will all catch exceptions.

Histogram¶

from anaconda_opentelemetry.signals import *

record_histogram("request_duration_ms", value=123.4, attributes={"route": "/home"})

Gauge¶

Records the last value set, for values sampled at a point in time.

from anaconda_opentelemetry.signals import *

set_gauge("request_queue_depth", value=7, attributes={"host": "worker-01"})

Counter (Increment)¶

from anaconda_opentelemetry.signals import *

increment_counter("active_sessions", by=1, attributes={"region": "us-east"})

Counter (Decrement)¶

Restricted to type simple_up_down_counter.

from anaconda_opentelemetry.signals import *

decrement_counter("active_sessions", by=1, attributes={"region": "us-east"})

Traces¶

This function does not need additional error handling. It will all catch exceptions.

from anaconda_opentelemetry.signals import *

with get_trace("process_data", attributes={"job_id": "abc-123"}):
    # Your business logic here
    process_data()

Schema¶

attrs = ResourceAttributes(
  "test_service",  # service_name requires a user-supplied value, not a keyword arg
  "v1",  # service_version requires a user-supplied value, not a keyword arg
  os_type="Darwin",
  os_version="24.2.0",
  python_version="3.13.2",
  hostname="Users-MBP"
)

Dynamic Resource Attributes in the Schema¶

Passing kwargs to the ResourceAttributes set_attributes method

attrs = ResourceAttributes("test-service", "1").set_attributes(foo="test")

Or passing a dictionary

my_attributes = {
  "test1": "one",
  "test2": "two"
}
attrs = ResourceAttributes("test-service", "1").set_attributes(**my_attributes)

Disabling Automatic Attribute Collection¶

You can opt out of auto-collection via ResourceAttributes(auto_collect=False) or exclude specific attributes via exclude_auto_collect=[“hostname”, “python_version”] if you’re collecting this data through other means and it would be redundant (e.g., using anaconda-auth provides a machine token which is simply a better version of hostname). Excluded attributes are completely omitted from the payload.

By default, ResourceAttributes automatically collects os_type, os_version, python_version, and hostname when not explicitly provided:

attrs = ResourceAttributes(
  "test_service",
  "v1",
  auto_collect=False
)

When auto_collect=False, the attributes will completely omitted unless explicitly provided:

attrs = ResourceAttributes(
  "test_service",
  "v1",
  os_type="CustomOS",
  python_version="3.10.0",
  auto_collect=False
)

Excluding Specific Auto-Collected Attributes¶

To exclude only specific attributes from auto-collection while keeping others, use exclude_auto_collect:

attrs = ResourceAttributes(
  "test_service",
  "v1",
  exclude_auto_collect=["hostname", "python_version"]
)

In this example, os_type and os_version will be auto-collected, but hostname and python_version will remain empty strings.

Testing and Visualizing Telemetry Locally¶

To test the telemetry package and view exports locally, the following code can be used. The console exporter exports telemetry payloads to standard output:

from anaconda_opentelemetry.signals import initialize_telemetry, increment_counter
from anaconda_opentelemetry.attributes import ResourceAttributes
from anaconda_opentelemetry.config import Configuration

cfg = Configuration(default_endpoint='http://localhost:4318').set_console_exporter(use_console=True)
att = ResourceAttributes("test_service", "dev-build", environment="test")
initialize_telemetry(config=cfg, attributes=att, signal_types=['metrics'])

increment_counter("test", by=1)