randgen team mailing list archive

["John H. Embretsen" <655740@xxxxxxxxxxxxxxxxxx>]

I reverted the original change (revid
philips@eve-20101001113351-hr7p09x3ukmfvf63) with revid
john.embretsen@xxxxxxxxxx-20101007071323-wnoa7jarq802zedg.

-- 
Information schema tests end prematurely due to RQG regression
https://bugs.launchpad.net/bugs/655740
You received this bug notification because you are a member of Random
Query Generator Team, which is subscribed to Random Query Generator.

Status in SQL Generator for testing SQL servers (MySQL, JavaDB, PostgreSQL): New

Bug description:
It seems that a change pushed October 1st has caused some kind of regression in the RQG. It is visible when using the information_schema.yy grammar, and the symptom is that the test run ends prematurely with the error 

failure status STATUS_PERL_FAILURE (255)

Steps to reproduce:

export CODE=/path/to/mysql

perl runall.pl \
--grammar=conf/runtime/information_schema.yy \
--threads=10 \
--duration=120 \
--reporters=Deadlock,ErrorLog,Backtrace,Shutdown \
--basedir=$CODE \
--queries=1000


With the revision prior to this push, the test runs as expected.

The revision info for the commit that caused the change is:

------------------------------------------------------------
revno: 542.1.8
revision-id: philips@eve-20101001113351-hr7p09x3ukmfvf63
parent: philips@eve-20101001074412-z53uxgzkrafcyyy4
committer: eve <philips@eve>
branch nick: randgen-bugfixing
timestamp: Fri 2010-10-01 14:33:51 +0300
message:
  Slightly improved the I_S metadata query so that tests with --threads=>10 have at least some chance to run before the end of the universe
modified:
  lib/GenTest/Executor/MySQL.pm  mysql.pm-20090724104844-fe42pf4t7xmrd94m-175


Diff currently at at http://bazaar.launchpad.net/~randgen/randgen/rqg2/revision/542.1.8