Bug #32839 | Inconsistent behavior of LOW_PRIORITY in 5.0 and 5.1 | ||
---|---|---|---|
Submitted: | 29 Nov 2007 8:07 | Modified: | 4 Dec 2007 14:47 |
Reporter: | Sveta Smirnova | Email Updates: | |
Status: | Can't repeat | Impact on me: | |
Category: | MySQL Server | Severity: | S3 (Non-critical) |
Version: | 5.1 BK | OS: | Any (Linux) |
Assigned to: | Davi Arnaut | CPU Architecture: | Any |
[29 Nov 2007 8:07]
Sveta Smirnova
[4 Dec 2007 14:47]
Davi Arnaut
It seems this is merely a timing issue, somehow on your machine the select thread got schedule first and hit the server before the update, probably due to "cosmic entropy" :-) 5.1-runtime: update t2 set b=1; update low_priority t2 set b=2; select * from t2 limit 3; a b 1 2 2 2 3 2
[4 Dec 2007 14:53]
Davi Arnaut
Expanding the comment: the update properly tries to get a low priority write lock and succeeds, the select has not yet hit the server. If the sleep is commented, the select gets to run first.
[4 Dec 2007 19:37]
Sveta Smirnova
In main tree bug is not repeatable as well.