NextGen UI documentation > Console > Activity log > Logs

Logs¶

A deployment's Logs tab receives logs from models and agentic workflows in the OpenTelemetry (OTel) standard format, centralizing the relevant logging information for deeper analysis, troubleshooting, and understanding of application performance and errors. Additionally, you can filter and view span-specific logs on the Monitoring > Data exploration tab.

The collected logs provide time-period filtering capabilities and the OTel logs API is available to programmatically export logs with similar filtering capabilities. Because the logs are OTel-compliant, they're standardized for export to third-party observability tools like Datadog.

Access and retention

Logs are available for all deployment and target types. Only users with "Owner" and "User" roles on a deployment can view these logs. Logs are stored for a retention period of 30 days, after which they are automatically deleted.

To access the logs for a deployment, on the Deployments tab, locate and click the deployment, click the Activity log tab, and then click Logs. The logging levels available are INFO, DEBUG, WARN, CRITICAL and ERROR.

Control	Description
Range UTC	Select the logging date range Last 15 min, Last hour, Last day, or Custom range.
Level	Select the logging level to view: Debug, Info, Warning, Error, or Critical.
Refresh	Refresh the contents of the Logs tab to load new logs.
Copy logs	Copy the contents of the current Logs tab view.
Search	Search the text contents of the logs tab.

Export OTel logs¶

The code example below uses the OTel logs API to get the OpenTelemetry-compatible logs for a deployment, print a preview and the number of ERROR logs, and then write logs to an output file. Before running the code, configure the entity_id variable with your deployment ID, replacing <DEPLOYMENT_ID> with the deployment ID from the deployment Overview tab or URL. In addition, you can modify the export_logs_to_json function to match your target observability service's expected format.

DataRobot Python Client version

The following script requires datarobot version 3.11.0 or higher installed to support OpenTelemetry logging submodules.

Export OTel logs to JSON
import json
import datarobot as dr
from datarobot.models.otel import OtelLogEntry

# Initialize DataRobot client for authentication
dr_client = dr.Client()

# Provide DataRobot entity information
entity_type: str = "deployment"
entity_id: str = "<DEPLOYMENT_ID>" # From the deployment Overview tab or URL

# Set output filename
output_filename: str = "datarobot_otel_logs.json"

def export_logs_to_json(logs: list[OtelLogEntry], filename: str) -> None:
    """Export logs to a JSON file"""
    log_entries = []
    for log in logs:
        log_payload = dict(log.__dict__)
        log_payload.update(
            {
                "entity_type": entity_type,
                "entity_id": entity_id,
                "timestamp": log.timestamp.isoformat(),
            }
        )
        log_entries.append(log_payload)

    with open(filename, 'w') as f:
        json.dump(log_entries, f, indent=2)

# Get and export OTel logs
try:
    print(f"Getting logs for {entity_type} {entity_id}...")
    logs: list[OtelLogEntry] = OtelLogEntry.list(entity_type, entity_id)
    print(f"Found {len(logs)} logs")

    # Preview first few logs
    for log in logs[:5]:
        print(f"{log}")

    # Example: Filter for ERROR logs only
    error_logs = [log for log in logs if log.level.upper() == 'ERROR']
    print(f"Found {len(error_logs)} ERROR logs")

    # Export to JSON file
    print(f"Writing logs to {output_filename}...")
    export_logs_to_json(logs, output_filename)
    print(f"Exported {len(logs)} logs to {output_filename}")

except dr.errors.ClientError as e:
    print(f"API Error: {e}")
except FileNotFoundError as e:
    print(f"File Error: {e}")
except Exception as e:
    print(f"Unexpected Error: {e}")