Python SDK Error Handling

This page explains how the Demandbase Python SDK reports API errors, retries supported failures, and exposes error details to your application.

Error Types

The main SDK error type is demandbase.DemandBaseAPIError.

The SDK raises DemandBaseAPIError when:

• The Demandbase API returns an HTTP error response.
• The SDK prevents a request because required request values are missing.
• A response has an unexpected shape for the requested SDK method.

Network-level errors, such as DNS failures, connection failures, and request timeouts, can be raised by the underlying HTTP client. Catch those separately if your application needs custom network retry or alerting behavior.

Basic Handling Pattern

import demandbase

try:
    with demandbase.DBClient(timeout=60.0, retry_count=2) as client:
        response = client.b2b_api.list_subscriptions(page=1, per_page=10)
        print(response.model_dump())
except demandbase.DemandBaseAPIError as error:
    print(f"Demandbase API error: {error.http_status_code} - {error.error_message}")
except Exception:
    # Handle network errors, timeouts, or unexpected application errors here.
    raise

DemandBaseAPIError Attributes

Example:

except demandbase.DemandBaseAPIError as error:
    print(error.request_path)
    print(error.http_status_code)
    print(error.error_message)
    print(error.response_headers)

Retries

DBClient accepts a retry_count argument.

import demandbase

with demandbase.DBClient(retry_count=2) as client:
    response = client.b2b_api.list_subscriptions()

The SDK retries these cases:

For 429 and supported 5xx responses, retry backoff starts at 30 seconds and doubles up to a maximum of 600 seconds.

If retries are exhausted, the SDK raises DemandBaseAPIError.

Rate Limits

If a rate-limit response is returned after configured retries are exhausted, catch DemandBaseAPIError and inspect:

• error.http_status_code
• error.response_headers
• error.error_message

Example:

except demandbase.DemandBaseAPIError as error:
    if error.http_status_code == 429:
        print("Rate limit reached. Retry later.")
        print(error.response_headers)
    else:
        raise

Rate-limit header names can vary by API response. Use the returned headers to decide whether your application should retry later, queue the request, or reduce request volume.

Authentication Errors

The SDK reads credentials from environment variables:

• DEMANDBASE_CLIENT_ID
• DEMANDBASE_CLIENT_SECRET

If either value is missing, client initialization raises RuntimeError.

If credentials are present but invalid, or the credentials do not have access to the requested API, the SDK raises DemandBaseAPIError.

Checklist:

• Confirm both environment variables are set in the runtime environment.
• Confirm the client ID and client secret are valid.
• Confirm the API Key Set includes the API permissions required by your integration.
• Confirm your integration is calling the production Demandbase API.

Validation Errors

The SDK validates some inputs before sending requests.

Example:

try:
    with demandbase.DBClient() as client:
        client.b2b_api.list_subscriptions(page=0)
except ValueError as error:
    print(f"Invalid request value: {error}")

Logging And Sensitive Data

SDK log output redacts bearer access tokens from authorization headers.

Request payloads, query parameters, and response bodies can still contain sensitive business data. Avoid logging full request or error context in production unless your application has appropriate access controls and retention policies.

Enable SDK logs when troubleshooting:

import demandbase

demandbase.enable_logging("INFO")

Use DEBUG only during focused troubleshooting because it can produce verbose request diagnostics.

Troubleshooting Guide

Related Pages

• Python SDK API Reference