I have a QlikSense application (May 2024 version) sourced from Snowflake. For the past year, my query has been running successfully in production, with daily data loads completing without issues. However, today, an error occurred—the query took more than two hours and failed with a "warehouse timed out" error (Error Code: 630).
Our database administrator mentioned that the failure was due to an unmasked role on one of the tables in the SQL query. Since Snowflake’s query plan generation is dynamic, it sometimes produces inefficient execution plans, impacting only specific queries while others work fine. Is this true?
Additionally, in Snowflake, the query owner or developer has no direct control over how the query executes once submitted. Instead, Snowflake’s cost-based optimizer (CBO) dynamically determines the execution plan. My query has been running successfully in 10 minutes for the past year, but suddenly, yesterday, it took an unusually long time and failed with a timeout error.
Can you confirm whether a developer has any control over a Snowflake SQL query’s execution? Also, is there any resolution available in QlikSense for this issue?
Great dreamer's dreams never fulfilled, they are always transcended. Please appreciate our Qlik community members by giving Kudos for sharing their time for your query. If your query is answered, please mark the topic as resolved 🙂
