Slowness and Incorrect Schema Context in SQL Editor 1.1 with Greenplum Database 7.x

Products

• VMware Tanzu Data Suite
• VMware Tanzu Greenplum
• VMware Tanzu Greenplum / Gemfire

Issue/Introduction

Customers may observe the following issues while using GP SQL Editor 1.1 when connected to a Greenplum Database 7.x cluster:

• Slow initial connections: First-time connections or connections after being idle for ~20+ minutes may take significantly longer compared to Greenplum Database 6.x.

• UI responsiveness issues: Database objects (schemas, tables, views) may take a long time to load and in some cases may display as zero objects.

• Incorrect schema context: When switching between databases in the object browser, schema details may be displayed incorrectly (e.g., objects from another database context).

Cause

Auto-Complete Overhead

• By default, the SQL Editor enables auto-complete functionality.

• Auto-complete fetches metadata for Greenplum keywords as well as live data for all columns, tables and views in the connected database.

• For large databases, this results in significant overhead, congesting the connection and slowing down initial UI load times.

Schema Context Handling

• After connecting, the SQL editor maintains the session in the context of the initially selected database.

• When navigating to objects from a different database, the schema resolution may still reference the original database, causing incorrect or misleading results.

Resolution

Workaround:

• Disable the auto-complete option in the SQL Editor.

• This prevents the editor from loading all schema/table/column metadata, significantly improving responsiveness in subsequent sessions.

 

Planned Fix:

• A fix for the incorrect schema context behavior is planned in a future release (v1.2 of the SQL Editor).