Skip to main content
Feedback

Viewing the Activity Log

Updated 

Use the Activity Log tab to review every ingestion run against any Source in a Knowledge Base, newest first. Use this tab to confirm that a Knowledge Base ingests data as expected and to find the cause of a failed run without opening each Source.

Before you begin

You need the Knowledge Hub Admin, Developer, Viewer, or Executor role to view the Activity Log tab. The Retriever role has no UI access. Refer to Access and permissions overview.

Open the Activity Log tab

  1. Open Knowledge Hub and select the repository.
  2. Select the Knowledge Base.
  3. Select the Activity Log tab.

If the Knowledge Base has no ingestion activity, the tab shows an empty state instead of a table.

Filter and sort the activity list

The controls above the table let you narrow the list:

  • Status: Select a status from the drop-down list. The default value is All.
  • Date: Enter a Start date and an End date to show only runs in that range.

Select a column header to sort the table by that column. Select the header again to reverse the sort order.

The Last Updated timestamp at the top right of the table shows when the list was last refreshed. Select the refresh icon to reload the list.

Read the activity list

Each row represents one ingestion activity. The table shows these columns:

ColumnWhat it shows
Execution DateThe date and time the run ran, with the time zone and offset from Coordinated Universal Time (UTC).
StatusThe run outcome. Refer to Ingestion activity statuses.
DurationHow long the run took, for example 8s or 1m 22s.
SourceThe name of the Source the run ingested from. The Source ID appears below the name. Select the Source name to open its detail page in Knowledge Hub.
Total RecordsThe number of records in the run. For unstructured Sources, such as files in an S3 bucket or a Confluence space, one file counts as one record. For structured Sources, such as database tables or Salesforce objects, one row counts as one record.
Processed RecordsThe number of records that processed successfully.
Failed RecordsThe number of records that failed to process.
ErrorA brief description of the error for a failed run. A dash (-) appears when the run has no error.

The three record columns appear under the Records group heading. Select the chevron beside Records to collapse the group.

The Error message can be cut off in the table. Open the activity details to read the full message.

The table shows 20 rows per page. Use the pagination controls below the table to view older or newer activity.

Ingestion activity statuses

StatusMeaning
PendingThe run is queued and waiting to start.
RunningThe run is in progress.
SuccessThe run completed and all records processed successfully.
Partial SuccessSome records failed and others loaded. A failed record keeps its previous version.
FailedThe run did not complete. A failed record keeps its previous version, and a Failed run can still have updated some records. Running ingestion again is safe.

Status thresholds

A run's status depends on how many records and files fail:

  • If no records fail, the run is Success.
  • If some records fail and no threshold is crossed, the run is Partial Success.
  • If a threshold is crossed, the run is Failed.
FailureThresholdMinimum sampleWhat happens when the threshold is crossed
A field value cannot be converted to its data type (for example, an invalid date)More than 5% of the records in one file20 records per fileAll records in that file fail. The other files in the run still load.
Files that cannot be read20% of the files in the runFive files
Records or files that fail to load20% of the run20 records or files
One part (chunk) of a file or record failsAnyNoneThe whole file or record fails and keeps its previous version.

For a data type failure, if more than 5% of a file's records fail, all records in that file fail, and the other files in the run still load. If 5% or less fail, a new record loads with that field empty, and an existing record keeps its previous version.

If a run has fewer records or files than the minimum sample, it is Failed only if everything fails.

If a mapped field is empty in every record (20 or more records), the run shows a warning and the status stays Success.

View activity details

Select the more options icon (⋮) at the end of a row to open the Ingestion activity details for that row.

FieldDescription
Execution DateThe date and time the run ran.
Activity IDThe unique identifier for this ingestion activity.
StatusThe run outcome. Refer to Ingestion activity statuses.
DurationHow long the run took.
SourceThe Source the run ingested from.
Source IDThe unique identifier for the Source.
ProviderThe data provider that ran the ingestion, for example Boomi Data Integration. Select the link to open the run in the provider and review connector-level diagnostics.
Total RecordsThe total number of records in the run. A record is one file for unstructured Sources or one row for structured Sources.
Processed RecordsThe number of records that processed successfully.
Failed RecordsThe number of records that failed to process.
SizeThe volume of data ingested during the run.
ErrorThe full error message for a failed run.

Select Download Log to save the run's log file.

Investigate a failed or partially successful run

  1. Filter the list by Status to show Failed or Partial Success runs, or find the row in the table.
  2. For a failed run, read the Error column for a brief description of the problem.
  3. Check the Failed column to see how many records failed. For example, a run with 31 total records, 0 processed, and 31 failed records returns the error No records processed successfully. Open the activity details to confirm which Source was involved.
  4. Select the Provider link to open the run in the data provider, for example Data Integration, and review the execution log for connector-level error details.

For a partially successful run, the Error column shows a dash (-). Open the activity details to see how many records failed.

You must have Data Integration access to follow the Provider link.

On this Page