> For my personal Django based project, I am getting random test failures > on Travis, e.g.: > > https://travis-ci.org/brianmay/spud/jobs/231462150 > > OperationalError: (2006, 'MySQL server has gone away') > > The usual culprit for the message - that the server closed its > connection due to idle time out and we didn't reconnect as we were > suppose to - doesn't seem to be applicable for a sequence of automated > tests - most if not all requiring db access. > > If I click restart job, it almost certainly will succeed. > > That message can be triggered by more than just an idle timeout.
Whenever I've encountered this it's almost always because the SQL query was too big (typically inserting/updating a large blob, or a very large 'WHERE x IN (1,2,3,..)' clause) https://dev.mysql.com/doc/refman/5.7/en/packet-too-large.html (Although without knowing the test case in detail I would have expected it would fail consistently) -- Kind Regards, -------------------------------------------------------- Alliance Software Ph: 03 9955 7000 Fax: 03 9894 2106 Email: [email protected] Web: www.alliancesoftware.com.au
_______________________________________________ melbourne-pug mailing list [email protected] https://mail.python.org/mailman/listinfo/melbourne-pug
