5.2 Release Notes

Another Move Toward Infinite Scalability

We've increased the number of possible Cinchy IDs that can be generated. This in turn allows the creation of more records within one table, so that you can create and manage larger data sets.

Previous Limit: 2,147,483,647 (2^31-1) Cinchy IDs per table

Updated Limit: 9,223,372,036,854,775,807 (2^63-1) Cinchy IDs per table

For backward compatibility with your database, you will need to manually run the below script against your TSQL or PGSQL databases. For instructions on how to run this upgrade, see the documentation here.

Connections Enhancements

• We've expanded our Connections capabilities to support binary file types as a data source. Review the documentation here for more information.

• We've improved the Connections experience by now making it optional to input a username or password when starting a batch sync job. This field can now be left blank if you want to run the job as the currently logged in user.

• We've added new optimizations and quicker processing of Kafka messages for real time data syncs in Connections.

• For added security, any logged password or sensitive parameters from the request details of a SOAP connector data sync is now redacted in the logs.

• Dead messages in the event listener are now written out to the execution errors table for easy collection and querying.

• The Connections experience now supports sourcing file based data sources from Azure Blob Storage and Amazon S3.

General Enhancements

• You now have the option to free up database space by using S3 compatible or Azure Blob Storage for file storage. This is configured in your deployment.json for Kubernetes installations and the appsettings.json in an IIS deployment.

  • If you are upgrading from an earlier v5 version, you can update your previous configuration to take advantage of this. For further instructions, review the documentation here.

• You are now able to run an erasure maintenance job on records that link to the Files table to delete the underlying referenced file.

• We've made improvements to the Files API to avoid cache build up and optimize the API.

• General security fixes and updates.

GraphQL (Beta)

• We've added anonymous API access to GraphQL, no token required.

• We've added write operations to our GraphQL beta, meaning that you can now insert and modify data.

Bug Fixes

• We've fixed a UI instability bug that resulted in the inability to resize view panes in the query designer and difficulty in selecting any cell in the first row of a table. This bug was affecting Chromium users (Google Chrome, Brave, Microsoft Edge, etc.) who had recently updated their internet browsers.

• We've fixed a bug that caused the “Created” column to incorrectly display as the last approved date of the column instead of the column created date.

• You can now add/remove a column from a table that has a columnar index without needing to remove said index entirely. We've also fixed a bug that prevented users from reapplying their columnar index to a table once it had been removed.

• We've fixed an issue where an “Unsupported Function Call” error was raised in certain situations when using the REPLACE function in conditional calculated columns in Connections.

• We've fixed a bug that caused unnecessary updates to the Users table when a user’s Language and Region wasn't set.

• We've fixed a bug that was causing some GUID calculated columns to appear as blank, such as in the Integrated Clients table. If you are experiencing this bug, a manual update on the affected rows, either through the UI or through an UPDATE query, will resolve it.

• We've fixed an issue where the '&' in links was sometimes showing up as 'amp&' in the table view instead. This fix will only appear for customers on Postgres/Microsoft SQL servers 2017 or higher.

• We've fixed an issue where certain Update statements on multi-select link columns were failing to properly update with the link values specified. This bug was affecting statement with long strings done via the API.

• We've fixed a bug that was causing the Event Listener to pick up and process messages from deleted configs from the LIstener Configs table.

• We've fixed a bug that was causing an InvalidOperationException when executing a POST request to a Saved Query API.

• We've fixed a bug that was throwing errors on reconciliation when Data Syncs compared Text Conditional Calculated Columns to Links (PGSQL).

• FOR JSON PATH now works as expected in PostGres deployments.