> From: Richard Huxton [mailto:dev@xxxxxxxxxxxx] > Relyea, Mike wrote: > > This seems to be a problem with the ODBC driver? How can I narrow > > that down further? > > Well, first make 100% certain the query being executed is the > one you see being sent from Access. I did by setting log_statement = 'all' and they were. > It will also let you turn statement logging on in that > database (ALTER DATABASE SET ...) and capture everything the > ODBC driver sends. There might be something that leaps out at > you. Take all of those statements and put them into a > text-file and run them using psql -f <file>. That should > cause the same crash. After setting log_statement = 'all', I ran my query using pgAdmin, and then ran the query using Access. I now had all of the commands sent to the DB by each application. > Remember, *something* in the sequence of commands that get > executed from Access must be different than when you execute > them through pgAdmin. That's exactly what it turned out to be. I looked at only the SET statements issued by each application and the differences popped right out at me. Turns out that Access was sending "set geqo to 'OFF'" and pgAdmin wasn't. As soon as I adjusted that setting in my connection string the problem went away. I'm not sure why it was off to begin with anyway. Thanks for your help! Mike ---------------------------(end of broadcast)--------------------------- TIP 9: In versions below 8.0, the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
