Monitor Deployments#

Apperate logs your workspace activities. The logs capture basic things such as when items are created, updated, and deleted and process milestones associated with dataset validation, data ingestion, and more. Go to Logs); it’s listed in your user menu.

The Logs page appears.


Log Stream#

The Log Stream lists activity messages in descending chronological order, most recent messages are at the top. Ingestion jobs include a JobId.

For example, the messages for the data ingestion job in the image above include the 769ab8* job ID. These messages capture the following data ingestion milestones:

  1. Creating the data ingestion job (i.e., job 769ab8*)

  2. Data upload

  3. Data validation

  4. Data processing (i.e., populating the table)

  5. Ingestion completion

The Log Stream shows log messages from the past hour. It’s a great resource for tracking process milestones and concurrent process timing.


To access older log messages, use the API. See the GET /logs API reference.

To see a log stream message’s details, double-click the message. The message details appear.


Log message timestamps are in UNIX epoch time.

Ingestion Logs#

The Ingestion Logs page lists each data ingestion job and provides status, record ingestion/validation count, processing times, matched record handling type, and data source type. You can use the Columns and Filters drop downs to show the columns you want and filter on the jobs.


If ingestion fails, the Last Good Status column indicates how far the job progressed (e.g., Validating). Click on Current Status to see the offending error message.


To get additional context for an ingestion, search the Log Stream for messages mentioning the job ID (shown in the Datasets column under the dataset name).

If ingestion completes, the Current Status is Ingested.

Invalid records are excluded from ingestion. A document icon in the Invalid Records column links to the ingestion job’s invalid record list.


Click the Invalid Records icon to download a CSV file of the invalid lines.


You can re-run a successful job by clicking the retry job icon (curved arrow) next to the dataset name.