Zach Musgrave
Zach Musgrave
I'm seeing the same behavior. This is a bug in testify, whether it's caused by testify code or indirectly by the dependency on spew. I've filed an issue with spew,...
At first glance it would seem that the server is reusing the memory allocated to handle bindvars in another request. We'll see if we can reproduce this on our end.
From a brief read through the vitess code that controls read and writing packets, it looks like this code is just wrong. It uses a buffer pool, and immediately after...
Sorry to hear you couldn't get it to work! Keeping this open since the bug appears to be real.
We believe this is fixed on main, caused by inappropriate reuse of MySQL connection buffer slices. Please open a new issue if you run into it again.
This is a bug: unique indexes other than primary keys are not enforced in the in-memory implementation.
@VinaiRachakonda did you fix this with your recent work in this space?
This needs an engine test -- we don't have tests of inserting duplicate indexes for the most part
Thanks for the feature request, @suessflorian Until we get to this, please use the full type description: `BIGINT UNSIGNED NOT NULL AUTO_INCREMENT UNIQUE`
This is now fixed.