--- sourceDocument: Australia IT Operations Management sourceDocumentLink: https://www.servicenow.com/docs/r/australia/it-operations-management Release : - australia ft:locale : - en-US ft:publication_title : - Australia IT Operations Management ft:clusterId : - itom bundleId : - itom workflow : - Technology --- # Source type structure adjustment # Source type structure adjustment in Health Log Analytics {#ariaid-title1} Release version: Australia Updated March 12, 2026 ![](https://www.servicenow.com/docs/portal-asset/ico-clock) 2 minutes to read Summarize ![AI sparkle icon](https://servicenow.com/docs/portal-asset/ai-sparkle-icon) Summarized using AI This content was generated using new OpenAI-powered functionality. Results are provided on an as is basis and are not guaranteed to be accurate or complete. ## Summary of Source type structure adjustment in Health Log Analytics Health Log Analytics (HLA) in ServiceNow helps you accurately analyze log data by enabling reclassification of auto-classified log properties and modification of auto-mapped labels. This ensures the HLA engine interprets log fields correctly, enhancing anomaly detection and log analysis. Show full answer Show less HLA automatically separates transport headers from inner log messages and processes only the inner message for property extraction and classification. It organizes log data by source type, not by data input, allowing precise structuring across multiple source types within a single input. ## Key Features * **Automatic Classification and Labeling:** HLA classifies fields into categories such as METER, GAUGE, ARCONLY, HISTOGRAM, and INVALID, and maps labels like MESSAGE, HOST, TIMESTAMP, SEVERITY, and EVENT-ID to log properties. * **AI-assisted Parsing with ServiceNow Otto:** Integration with ServiceNow Otto provides AI-driven recommendations for optimal classifications and labels, reducing manual configuration effort and improving accuracy. Activating AI-powered parsing enables automatic analysis and suggestions with visible AI status indicators. * **Manual Adjustments and Overrides:** Customers can review and override AI or automatic classifications and labels to tailor log data interpretation. JavaScript can be used to refine extracted properties, such as extracting meaningful message parts or reclassifying insignificant fields as INVALID. * **Learning Mode and Relearning:** The HLA engine's learning mode can be reactivated to update parsing, extraction, classification, and labeling when initial learning becomes outdated, ensuring ongoing accuracy. ## Practical Application for ServiceNow Customers By customizing the source type structure, customers can fine-tune how HLA reads inner log messages, improving anomaly detection and log data quality. Using AI assistance simplifies setup and maintenance, while manual overrides and JavaScript provide flexibility to handle unique log formats or requirements. This capability is vital for ensuring that log analytics accurately reflect the underlying data, enabling better operational insights and faster troubleshooting. ## Next Steps * Review auto-classified properties and labels in the source type structure to confirm accuracy. * Activate AI-assisted parsing with ServiceNow Otto for enhanced classification recommendations. * Use JavaScript within the source type structure to customize property extraction or reclassification as needed. * Reactivate the learning mode when log formats evolve or initial learning data becomes outdated. Health Log Analytics (HLA) enables you to reclassify auto-classified log properties and change auto-mapped labels. These adjustments help HLA machine learning analyze your data accurately. ## What HLA does automatically {#hla-source-type-structure-adjustment__section_own_3dj_c3c} HLA automatically separates the transport header from the inner log message and sends only the inner log message to the source type structure. The system extracts properties from incoming log messages and auto-maps labels to source type fields. * Classification determines how the HLA engine analyzes and processes each field. Available classifications are: METER, GAUGE, ARC_ONLY, HISTOGRAM, INVALID. * Labels tell HLA what role a field plays in the log structure. Available labels are: MESSAGE, HOST, TIMESTAMP, SEVERITY, EVENT-ID. {#hla-source-type-structure-adjustment__ul_o5x_ldj_c3c} Reviewing and, if necessary, modifying automatically mapped labels and auto-classified properties in the source type structure verifies the HLA engine interprets log fields correctly. Note: Because a single data input can contain multiple source types, HLA structures log data by source type rather than by data input. ## AI-assisted log parsing with ServiceNow Otto {#hla-source-type-structure-adjustment__section_h1f_gqf_bkc} AI-assisted log parsing using ServiceNow Otto provides recommendations for optimal classifications and labels for parsed log properties based on your log patterns. This approach reduces manual effort and improves accuracy without requiring custom JavaScript functions. To enable this capability, activate the integration with AI-powered capabilities by selecting the Activate with AI option. This option enables AI-powered automatic parsing and mapping of log data. When the integration is activated, ServiceNow Otto collects and analyzes log data and displays suggested classifications and labels for your source type properties. An AI icon (![]()) indicates that AI is active. Verify that ServiceNow Otto correctly classified and labeled your log properties. You can override the suggestions by selecting a different classification or label from each list, if needed. While ServiceNow Otto generates a parser automatically for a source type, a notification banner on the Source Type Structure page shows the current status of the generation process. ## Example {#hla-source-type-structure-adjustment__section_d4h_rnd_yhc} The following example shows how to modify values of extracted properties in the source type structure with JavaScript. Consider the following log: { "TenantId": "abc-01-02-03-04-05050708091011121314", "@timestamp": "2020-08-28T08:29:23.967Z", "Computer": "john Doe_computer", "EventType_s": "LogMessage", "Job_s": "johnDoe_cell", "IP_s": "1.00.00.00", "message": "This is the extracted message. This part of the message includes superfluous content and values", "MessageType_s": "OUT", "Timestamp_d": 1598603359017850000, "Type": "my_LogMessage_is", "_ResourceId": "" } The sample code contains `"key":"value"` pairs: The key is the property name. The value is the property value. The key "message" in the sample has the following value: `"This is the extracted message. This part of the message includes superfluous content and values"`. If you wanted your logs to contain only the meaningful part of that message, you would add JavaScript code instructing the system to extract only that part: //Added JavaScript to extract only the first sentence in the message! if (output['message'] != null){ output['message'] = output['message'].slice(0, output['message'].indexOf("\.")); } (edited) You could use the same logic to reclassify a value. For example, if the key "Computer" is insignificant, you could set its value to "Invalid." For more information about the source type structure, see the [Source Type Structure -- Labels \& Classifications \[KB0863562\]](https://support.servicenow.com/kb_view.do?sysparm_article=KB0863562) article in the Now Support Knowledge Base. Note: You can reactivate the learning mode for a source type structure to use the HLA engine's learning functionality again after the initial learning period becomes outdated. On the Source Type Structure form, select Relearn to reset parsing, extraction, classification, and labeling for the source type structure and restart the learning process. * **[Refine the source type structure in Health Log Analytics](https://www.servicenow.com/docs/ha4j~5dtevLR_LAFEJMnhw)** Fine-tune how Health Log Analytics reads your inner log messages and detects anomalies by customizing the extracted properties in the source type structure.