Close the connection properly#37
Conversation
| logger.debug( | ||
| "Exception while closing cursor: {}".format(exc) | ||
| ) | ||
| self._conn.close() |
There was a problem hiding this comment.
I'm wondering whether this is also needed in dbt-spark.
There was a problem hiding this comment.
@jtcohen6 Is it fine in dbt-spark? I suspect it has the same issue without any notice.
There was a problem hiding this comment.
Came here because I was seeing the same unclosed socket error as in #34. Tried the fix in this PR, worked like a charm! Then I saw I was tagged...
This isn't an issue we've noticed with the dbt-spark connector — but that's not so surprising to me, different cursors are liable to behave differently. Even with the change in #35, which will allow dbt-databricks to inherit (and avoid reimplementing) common functionality from dbt-spark, I would expect these methods to be points of departure
There was a problem hiding this comment.
Got it. Thanks for confirming!
|
Thanks! merging. |
resolves databricks#34 ### Description Closes the connection properly.
4 participants
resolves #34
Description
Closes the connection properly.