Bug #30751 rpl_ndb_multi missing row in output

Description:
Once in a rare while, the rpl_ndb_multi test fails with this difference:

-------------------------------------------------------
*** r/rpl_ndb_multi.result	Thu Mar 16 18:08:55 2006
--- r/rpl_ndb_multi.reject	Fri Mar 17 05:05:59 2006
***************
*** 51,55 ****
  c1	c2	c3
  row3	C	3
  row4	D	4
- row5	E	5
  STOP SLAVE;
--- 51,54 ----

How to repeat:
The first occurrance I can find is in March 2006, at:

https://intranet.mysql.com/secure/pushbuild/showpush.pl?dir=mysql-5.1&order=721

It's currently showing up in 5.1 and 5.2, e.g.:

https://intranet.mysql.com/secure/pushbuild/showpush.pl?dir=mysql-5.2&order=82

Currently, it seems to happen almost always on powermacg5 n_mix, but see the following for a failure on rh-x86-32 ps_row:

https://intranet.mysql.com/secure/pushbuild/showpush.pl?dir=mysql-5.1-new-maint&order=703

It does still happen.
Disabling the test case. Please re-enable after fix.

For a current failure see pushbuild, for example https://intranet.mysql.com/secure/pushbuild/showdir.pl?dir=mysql-5.1-engines
log https://intranet.mysql.com/secure/pushbuild/getlog.pl?dir=mysql-5.1-engines&entry=rkalimul...

Pushed into 6.0.4-alpha

Pushed into 5.1.23-rc

Leaving my finding -

ChangeSet 1.2512.5.10 2008/02/12 14:50:36 tomas@poseidon.ndb.mysql.com
  corrected really strange test
mysql-test/suite/rpl_ndb/t/rpl_ndb_multi.test 1.9 2008/02/12 14:50:34 tomas@poseidon.ndb.mysql.com
  corrected really strange test

that definetly fixed the issue in this test.

I think the bug can be closed.

cheers,

Andrei