Record Health Check Log event fields
Look up every field of the Record Health Check Log Platform Event, which reports one technical error that Record Health Check captured during a run, with its message, exception type, and stack trace. Use it only for restricted support monitoring; business automation should use the Result event.
| Property | Value |
|---|---|
| Label | Record Health Check Log |
| API name in your org | rhc_ |
| Event type | High Volume |
| Publish behavior | Publish Immediately |
| Contract Version | 1. |
To save these events in a protected object with Flow or Apex, see Save or route restricted errors.
When it publishes
Section titled “When it publishes”A Log event publishes for each error Record Health Check captures while running a Check Set when:
- Publish Error Log Event is selected on that Check Set. It is cleared by default.
- The user who runs the Checks has the Record Health Check Error Log Publisher permission set.
It does not depend on the card checkboxes or on Event Publication, so it applies to runs from the card, Flow, Apex, and async jobs alike. When Record Health Check cannot identify the Check Set, nothing publishes. Only errors publish, never warnings or informational messages.
Every UNABLE_TO_EVALUATE and ERROR result counts as an error, with Code DIAGNOSTIC_INCIDENT.
When a failure affects a whole group of records, Record Health Check logs it once for the group.
Otherwise, a run in which three Checks return Unable to Check for 200 records can publish up to
600 Log events.
Because the event uses Publish Immediately, Salesforce delivers it even when the transaction that captured the error rolls back. Record Health Check collects the errors during a request and publishes them in groups of up to 100 when the request ends; a transaction that stops at an uncatchable Salesforce limit can end first, so some errors produce no event.
Fields
Section titled “Fields”| Field label | API name in your org | Type | Value |
|---|---|---|---|
| Event ID | rhc_ | Text(80), required | A different value for every event. Save it in a Unique field to detect a repeated event. |
| Run ID | rhc_ | Text(120), required | The run in which the error happened. |
| Occurred At | rhc_ | Date/Time, required | When the error was captured. |
| Level | rhc_ | Text(10), required | ERROR. |
| Code | rhc_ | Text(120) | A short classification of the error. See Codes. |
| Check Set Developer Name | rhc_ | Text(120) | The name of the Check Set that was running. |
| Check Developer Name | rhc_ | Text(120) | The name of the Check that was running, when known. |
| Record ID | rhc_ | Text(18) | The record being checked, when known. |
| User ID | rhc_ | Text(18) | The user who ran the Checks. |
| Exception Type | rhc_ | Text(120) | The Apex exception type, such as System., when an exception occurred. |
| Message | rhc_ | Long Text Area(32768) | The exception message followed by the error details as name=value pairs, up to 4,000 characters. |
| Structured Diagnostic Details | rhc_ | Long Text Area(32768) | The same details as a JSON object, up to 4,000 characters. |
| Stack Trace | rhc_ | Long Text Area(32768) | The Apex stack trace, up to 32,000 characters, when an exception occurred. |
| Contract Version | rhc_ | Text(10), required | 1.. |
| Framework Version | rhc_ | Text(20) | 2.. |
Line breaks and tabs in text values are replaced with spaces.
Code names where the error happened. The codes in 2.0.11 are:
| Code | Where the error happened |
|---|---|
DIAGNOSTIC_ | A Check or a group of records failed, and Record Health Check classified the failure for diagnosis. |
CONTROLLER_ | The record-page card evaluated a Check. |
CONFIG_ | The card loaded the Checks of a Check Set. |
SHELL_ | The card loaded the Check Set’s card settings. |
AVAILABILITY_ | The card checked whether the Check Set applies to the record. |
RUN_ | The card sent the results of a Run or Rerun for publication. |
BATCH_ | A Batch job evaluated one group of records. |
QUEUEABLE_ | A Queueable job ended with an unhandled exception. |
AGENT_ | An Agentforce action ran one Check. |
AGENT_ | An Agentforce action ran a Check Set. |
Codes can be added or changed in any release. Use them to group errors, and send a Code the automation does not recognize to review. For business outcomes, use the Reason codes on the Result event.
For DIAGNOSTIC_INCIDENT, Structured Diagnostic Details contains these keys:
| Key | Value |
|---|---|
diagnosticId | The Diagnostic ID shown in the card’s support report and returned to Flow, Apex, and agents as diagnosticId. Use it to match an event to a reported result |
category | What prevented a verdict: ACCESS, CONFIGURATION, FORMULA, QUERY, LIMIT, APEX_, APEX_, APEX_, APEX_, or FRAMEWORK. See Diagnosis fields |
phase | The step that failed, such as FIELD_ or PLUGIN_, or UNKNOWN |
reasonCode | The result’s Reason Code |
owner | Who usually fixes it: ADMINISTRATOR, SECURITY_, or DEVELOPER |
retryable | true when running again can succeed, as for a LIMIT or FRAMEWORK failure; otherwise false |
scopeImpact | ONE_, or SCOPE_ when one failure affected several records in the request |
fingerprint | A 24-character hash of the category, phase, Reason Code, component, exception type, and failing code location. Events with the same fingerprint describe the same failure, so group on it |
affectedRecordCount | How many records the failure affected |
exceptionType | The Apex exception type, when an exception occurred |
topFrame | The class, method, and line where the exception was raised, when known |
Example event
Section titled “Example event”{ "rhc__EventId__c": "batch-7071x00000AbCdEFGH-BATCH_SCOPE_FAILED-48151623", "rhc__RunId__c": "batch-7071x00000AbCdEFGH", "rhc__OccurredAt__c": "2026-10-07T02:15:00.000Z", "rhc__Level__c": "ERROR", "rhc__Code__c": "BATCH_SCOPE_FAILED", "rhc__CheckSetDeveloperName__c": "Account_Data_Quality", "rhc__CheckDeveloperName__c": null, "rhc__RecordId__c": null, "rhc__UserId__c": "005000000000001AAA", "rhc__ExceptionType__c": "System.DmlException", "rhc__Message__c": "exception=Insert failed. UNABLE_TO_LOCK_ROW", "rhc__DetailsJson__c": "{\"exceptionType\":\"System.DmlException\",\"exceptionMessage\":\"Insert failed. UNABLE_TO_LOCK_ROW\",\"exceptionLine\":42}", "rhc__StackTrace__c": "<Apex stack trace>", "rhc__ContractVersion__c": "1.0", "rhc__FrameworkVersion__c": "2.0.11"}The values are examples. A Batch error has no Record ID or Check because it applies to a whole group of records. Never paste a real event into a ticket, chat channel, or public page: its message and stack trace can reveal record IDs, field names, and code from your org.
Access
Section titled “Access”No runner permission set grants access to this event. The Record Health Check Error Log Publisher permission set grants Create and Read access. Salesforce requires Read with Create, so every assignee can read restricted error details.
Receiving automation needs Read access to the event and access to the protected object it writes to. It must never run Record Health Check, because an error in that run could publish another Log event and start the automation again.