Code: Select all
Copy myTable From ‘/myDirectory/*’ On Any Node
Rejected Data ‘/rejects_directory/’ On node_1, ‘/rejects_directory/’ on node_2
Exceptions ‘/exceptions_directory/’ On node_1, ‘/exceptions_directory/’ on node_2
If I change the query to specify a rejection and exception file like so:Error 2016: Copy: Could not open rejected data file for writing: Is a directory
Code: Select all
Copy myTable From ‘/myDirectory/*’ On Any Node
Rejected Data ‘/rejects_directory/rejects.txt’ On node_1, ‘/rejects_directory/rejects.txt’ on node_2
Exceptions ‘/exceptions_directory/exceptions.txt’ On node_2, ‘/exceptions_directory/exceptions.txt’ on node_2
If I just comment out the rejected data and exceptions lines, the query works fine, but then I’m stuck searching through each node of my cluster for my reject files. What makes this even stranger is that the table and source directory are passed through as parameter values in a shell script and the query works fine in most cases, but throws the error in others. Any ideas why this is happening?Error 2016: Copy: Could not open rejected data file for writing: Is not a directory