maria-developers team mailing list archive

[Paul McCullagh <paul.mccullagh@xxxxxxxxx>]

I suspect that this is not due to rows being lost, but rather to the  
rows not being found for some timing reason.

But, of course, only debugging will tell...

We have scheduled time to look at this error.

On Feb 25, 2011, at 3:58 PM, Michael Widenius wrote:

> Hi!
>
> This is a bug that has been around a long time but not yet fixed (at
> least not in 1.0.11 which is in the MariaDB 5.1 tree).
>
> Once in a while when running the pbxt.insert test, pbxt fails to find
> all rows in the table:
>
> pbxt.insert                              [ fail ]
>        Test ended at 2011-02-25 16:44:03
>
> CURRENT_TEST: pbxt.insert
> --- /home/my/maria-5.1-merge/mysql-test/suite/pbxt/r/ 
> insert.result      2011-02-10 22:57:21.000000000 +0200
> +++ /home/my/maria-5.1-merge/mysql-test/suite/pbxt/r/ 
> insert.reject      2011-02-25 16:44:03.030856755 +0200
> @@ -298,7 +298,7 @@
> insert into  t2 select t1.* from t1, t2 t, t3 where  t1.id1 = t.id2  
> and t.id2 = t3.id3;
> select count(*) from t2;
> count(*)
> -25500
> +25495
> drop table t1,t2,t3;
> create table t1 (a int, b int);
> insert into t1 (a,b) values (a,b);
>
> Rerunning the test usually works
>
> It would be really good to get this fixes ASAP as it's not good if a
> storage engines looses rows randomly...
>
> The above happend in my latest 5.1 tree on Linux.
>
> Regards,
> Monty