Aaron Stone
Aaron Stone
Now the builds are broken due to a memory issue. Also updated .travis.yml to build Ruby 2.0 and 2.1.
I remembered that `ALLOC_N` should be paired with `xfree` and not ordinary `free, and set`bzf->buf = NULL` after each free. Now it just doesn't work instead of crashing. Progress!
tl;dr - look for circumstances by which a connection is being returned to the pool while still in an in-use state. Could be an exception handler, could be improper logic,...
Yes, this can be caused by a race condition. Your code is responsible for managing the connection pool and for handling how threads check out a connection and return it...
Expected to work fine on Debian 11. Looks like going through the standard motions on debugging will be helpful here. Can you get a backtrace from the Ruby process?
That looks like an interesting addition if you would consider opening a PR?
Great! Next step is to click this button Open Pull Request -> 
Thank you for the thorough bug report and your initial efforts at debugging this! What a weird crash. I'm not sure off the cuff what action to take to avoid...
I didn't get this one into the 0.5.3 release, but I want to plan for whether to include it in the next release or not. If you'd be so kind...
I am confused why MySQL doesn't return NULL as the value type in this case, and is instead giving back an all-zeroes date value. It seems like the fix is...