Bug #19021 | Server crashes while resolving index merge | ||
---|---|---|---|
Submitted: | 11 Apr 2006 20:24 | Modified: | 14 Apr 2006 14:13 |
Reporter: | Kolbe Kegel | Email Updates: | |
Status: | Closed | Impact on me: | |
Category: | MySQL Server: Optimizer | Severity: | S1 (Critical) |
Version: | 5.0.20 | OS: | Linux (Linux) |
Assigned to: | Sergey Petrunya | CPU Architecture: | Any |
[11 Apr 2006 20:24]
Kolbe Kegel
[12 Apr 2006 23:29]
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/4891
[13 Apr 2006 8:07]
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/4905
[13 Apr 2006 8:21]
Sergey Petrunya
The fix has been pushed into 5.0.21 tree. Note for the changelog: The bug was affected when evaluating possibility to use Index Merge/Intersection+"Using index" variant of index_merge. In order to hit a bug, all usable indexes together should cover all fields accessed by the query, and table statistics must have some particular values (the precise criteria is very complicated). The effect of the bug was either a crash or choice of non-optimal subset of indexes to be used in Index Merge/intersection.
[14 Apr 2006 14:13]
Paul DuBois
Noted in 5.0.21 changelog. The optimizer could cause a server crash or use a non-optimal subset of indexes when evaluating whether to use <literal>Index Merge/Intersection</literal> variant of <literal>index_merge</literal> optimization. (Bug #19021)