Bug #22393 | Optimizer choose non-optimal index | ||
---|---|---|---|
Submitted: | 15 Sep 2006 14:37 | Modified: | 4 Oct 2006 20:25 |
Reporter: | Victoria Reznichenko | Email Updates: | |
Status: | Closed | Impact on me: | |
Category: | MySQL Server: Optimizer | Severity: | S3 (Non-critical) |
Version: | 4.1,5.0,5.1 | OS: | Any (any) |
Assigned to: | Sergey Petrunya | CPU Architecture: | Any |
[15 Sep 2006 14:37]
Victoria Reznichenko
[16 Sep 2006 12:27]
Sergey Petrunya
This bug will be fixed by this patch: http://lists.mysql.com/commits/396. I'll need to apply it and add testcase. The bugfix will go into 5.1 (following the "no query plan changes in GA versions" rule)
[18 Sep 2006 10:45]
Bugs System
A patch for this bug has been committed. After review, it may be pushed to the relevant source trees for release in the next version. You can access the patch from: http://lists.mysql.com/commits/12121 ChangeSet@1.2289, 2006-09-18 14:49:54+04:00, sergefp@mysql.com +3 -0 BUG#22393: Very wrong E(#rows(ref(const)) for key with skewed distribution - Check if we have E(#rows) for 'range' access on the smaller interval on the same index. If yes, adjust the estimate.
[1 Oct 2006 9:02]
Georgi Kodinov
Pushed in 5.1.12
[4 Oct 2006 20:25]
Paul DuBois
Noted in 5.1.12 changelog. The optimizer could make an incorrect index choice for indexes with a skewed key distribution.
[19 Aug 2007 18:54]
Igor Babaev
Bug #20994 is marked a duplicate of this bug.