Comment 6 for bug 1264164

Further analysis on this issue.

1. Issue is confirmed sporadic
2014-09-03 23:31:48 [Init] Looking for this string: 'm_can_overwrite_status' in mysqld error log output...
[...]
2014-09-03 23:33:21 [*] [Stage V] [MULTI] Only 3 out of 20 threads reproduced the issue: this issue is sporadic
(and this is after already 10 threads were tried just before it)
2. Issue can likely be reproduced without any specific SQL. I could reproduce it using the one-line default from reducer.sh "DROP DATABASE transforms;CREATE DATABASE transforms;DROP DATABASE test;CREATE DATABASE test;USE test;" - but it is likely that not even this SQL (or all of it) is really required to reproduce the issue. It may be a general outside-of-sql-parsing issue.
3. Issue may require fast memory storage (reducer.sh uses /dev/shm by default) to reproduce - thinking race condition or something.
4. Issue does not require ANY specific mysqld parameters to be set - works on vanilla PS.
5. Issue is annoying and causes RQG runs to fail and reducer.sh runs to be cut short/be misled (due to hitting this in general-crash simplification method) - warrants adding qablock

Will now add a number of occurences for core analysis.