We are seeing some odd behavior when connecting to Snowflake from a task using the snowflake connector. This works:

for doc in query:

, but it appears to load the entire resultset into memory before enumerating the rows. The better practice is to use

doc = query.fetchone()
while doc: 
  doc = query.fetchone()

However, this hangs immediately when run from within a task. Has anyone else seen this behavior? Any idea why it might be behaving this way?

Hey Adam 👋 I've let the team know you've got a question outstanding. Can you share a minimum reproducible example with us?

Hey Dylan, this is the same issue as https://prefect-community.slack.com/archives/CL09KU1K7/p1670860398321709 so let's disregard this thread. Thank you

Gotcha, thank you 🙏