Code review comment for lp:~stewart/percona-server/bug1012488

On 09.07.12 9:28, Stewart Smith wrote:
> Alexey Kopytov <email address hidden> writes:
>> All of that is because of 2 rudementary tests pam_negative_test.py and
>> pam_basic_test.py. There's absolutely no reason to base them on
>> kewpie. Apparently there was also an attempt to use the XtraBackup
>> test suite, because there's also
>> plugin/percona-pam-for-mysql/test/bootstrap.sh which is a verbatim
>> copy from the XtraBackup tree and is unused in PAM plugin tests.
>
> Kewpie still needs a server to run, and bootstrap.sh is used to fetch it
> (unless you're building a PS tree, in which case you can use that).

Well, in this case we do build in a PS tree.

Besides, bootstrap.sh is a verbatim copy from the XB tree, so I wonder
how it's arguments (innodb51, xtradb51, etc.) apply here.