[CT-443] Potentially sensitive data can be written to logfiles on incremental model failure

[dbt-labs]

[CT-443] Potentially sensitive data can be written to logfiles on incremental model failure

[dataders]

@verhey thanks for logging this! My 5 min take is that I think we can handle this scenario by catching the presence of the error, the manipulating the message before logging it with the debugger where at least you could tell the user how many rows are duplicate? in my example below I'm not manipulating the string just giving a new string. hopefully this gives you a start?

dataders's super simple example

if "100090 (42P18): Duplicate row detected during DML action" in msg:
    logger.debug(" 100090 (42P18): Duplicate row detected during DML action!")

Not sure if you've contributed to dbt-core or dbt-snowflake before, but this guide is a great place to get started

similar error handling done previously

dbt-snowflake/dbt/adapters/snowflake/connections.py

Lines 226 to 227 in 184eaeb

	if "Empty SQL statement" in msg:
	logger.debug("got empty sql statement, moving on")

[github-actions]

This issue has been marked as Stale because it has been open for 180 days with no activity. If you would like the issue to remain open, please remove the stale label or comment on the issue, or it will be closed in 7 days.

[verhey]

I have a draft PR open at #271 to resolve this, but it's waiting on review and I imagine it'll need some more iterations.

@dataders let me know how you'd like that to happen, but for now, un-staling this issue.