Change Details

During an update, we encountered a 'Duplicate entry' error while trying to insert an IP. The previous import was done on 10 Nov. The update attempt was 16 Nov. No other imports had been done in between: ``` SELECT feed_file_today, COUNT(*) FROM import_status GROUP BY feed_file_today; +-----------------------+----------+ | feed_file_today | COUNT(*) | +-----------------------+----------+ | /tmp/20231110.json.gz | 3286 | | /tmp/20231116.json.gz | 2003 | +-----------------------+----------+ 2 rows in set (0.007 sec) ``` This leaves the database in a corrupted state, where it contains data from two differnt days. This is a problem because the next update will assume that the database contents are the same as the old data file, for diffing purposes. T344941 is trying to address this.

During an update, we encountered a 'Duplicate entry' error while trying to insert an IP. The previous import was done on 10 Nov. The update attempt was 16 Nov. No other imports had been done in between: ``` SELECT feed_file_yesterday, feed_file_today, COUNT(*) FROM import_status GROUP BY feed_file_today; +-----------------------+-----------------------+----------+ | feed_file_yesterday | feed_file_today | COUNT(*) | +-----------------------+-----------------------+----------+ | | /tmp/20231110.json.gz | 3286 | | /tmp/20231110.json.gz | /tmp/20231116.json.gz | 2003 | +-----------------------+-----------------------+----------+ 2 rows in set (0.041 sec) ``` This leaves the database in a corrupted state, where it contains data from two differnt days. This is a problem because the next update will assume that the database contents are the same as the old data file, for diffing purposes. T344941 is trying to address this.

During an update, we encountered a 'Duplicate entry' error while trying to insert an IP. The previous import was done on 10 Nov. The update attempt was 16 Nov. No other imports had been done in between: ``` SELECT feed_file_yesterday, feed_file_today, COUNT(*) FROM import_status GROUP BY feed_file_today; +-----------------------+-----------------------+----------+ | feed_file_yesterday | feed_file_today | COUNT(*) | +-----------------------+-----------------------+----------+ | | /tmp/20231110.json.gz | 3286 | | /tmp/20231110.json.gz | /tmp/20231116.json.gz | 2003 | +-----------------------+-----------------------+----------+ 2 rows in set (0.007041 sec) ``` This leaves the database in a corrupted state, where it contains data from two differnt days. This is a problem because the next update will assume that the database contents are the same as the old data file, for diffing purposes. T344941 is trying to address this.