For all volume incident charts with row count measurements, we now show the option to view charts in row change, in order to help users better understand the volume update patterns. This is currently not available for byte measurements.
Added a General Information module to the Catalog Summary tab to more quickly access the following for a table:
- Description
- Last updated timestamp
- \# of users
- Latest row count
Fixed a bug where the date filter for the incident feed did not reset after an end date was selected and users came back to the page from other sessions. Now the date filter always resets to include the current date as the latest date.
In the SQL Rule Monitor creation flow, you can now select "Set threshold using: Value returned by query" along with the "Automatic" threshold type. MC will learn the rule's pattern and set automatic breach conditions. Our algorithm learns both trend information and daily & weekly seasonality, making this a good setting for metrics that grow predictably over time or have routine drops overnight / over weekends.
The checks are performed when executed with â-dry-run . It analyses the freshness_threshold defined and the expected threshold for the table. If something is wrong you will see messages like:
Warning - High breaching Freshness SLOs will be created, adjust freshness_threshold - analytics:prod.client_hub: freshness_threshold is 3, the expected minimum threshold is 462 - analytics:prod.client_warehouses: freshness_threshold is 3, the expected minimum threshold is 1410
I want my users to be Viewers of everything, but an Editor within their Domain." Previously, this was complicated to set up for your users, because you could not simultaneously include a user in the default 'Viewer' group and a custom Editor group just for their Domain. But now, this is possible!
Check that data coming from external partners is landing in your warehouse on-time, and in the size and quality that you expect. Read more about it here.
Query Change RCAs are now available for Freshness anomalies. This RCA correlates between changed/new write queries and freshness anomalies in a similar way weâve been doing for volume anomalies. A notification on a Incident with a Freshness will be available when this RCA was detected.
Looking to easily update the Owner or Severity of an Incident without clicking into Incident IQ? You can now update it directly from the Incident card within the Incident Feed.
New filters on the Data Reliability Dashboard help you hone the metrics to just your most important data (Key Assets) or just specific Incident Types.