Hello everyone, I'll try to be very pointed with my question for easy understanding. So, guys, the thing is that I'm preparing my exam for HDPCD so I'm playing a lot with sqoop. In this case, I have a sample txt file to load data in Mysql. I have done the first load and then I would like to try the properties --update-key and --update-mode allowinsert to join and update my table. I performed a first load data to mysql table and that looks like: so after the first load, I have added new rows and change one row for my column called season in id <10> to <winter> so my file looks now like I executed: sqoop export --connect jdbc:mysql://master/poc --username root --table cours --export-dir /user/hdfs/mysql/export.txt --update-key id --update-mode allowinsert --fields-terminated-by '\t' -m 1 My new table output looks like: That's correct cause the new courses were included and the season change on row id 10 also was updated but instead of update the same row sqoop has added a new column so i think that it should better just really update and not duplicate my id cause what if it's a primary key so guys. do you know how perform the update without duplicate non necessary data?
... View more
