Exporting 5 million records (1 reply)

I have a table with 5 million rows of data.
I need to do a export without adding the PK column. mysqldump cant select columns.

sql = """SELECT `col2`,`col3`,`col4`,`col5`,`col6`,`col7` FROM `tbl1`;"""
cursor.execute(sql)

while True:
    r = cursor.fetchone()
    if r is None: break
    print "INSERT IGNORE INTO `tbl2` VALUES " , ("",str(r['col2']),r['col3'],str(r['col4']),str(r['col5']),str(r['col6']),str(r['col7'])),";"

python export.py > insert.sql

But this is taking forever.
I understand if fetchall() first gets all the data into the buffer, but I though fetchone would take one row at a time in buffer.

freemem :
             total       used       free     shared    buffers     cached
Mem:       8061084    8014704      46380          0       1228      40300
-/+ buffers/cache:    7973176      87908
Swap:      2000084    1673828     326256

So how I do select all rows and print/update each one ?