unresolved MDL deadlock with limited innodb thread concurrency
Bug #861212 reported by
Seppo Jaakola
This bug affects 1 person
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
MySQL patches by Codership |
Fix Released
|
High
|
Seppo Jaakola | ||
5.5 |
Fix Released
|
High
|
Seppo Jaakola |
Bug Description
Hanging happens in 1.0beta evaluation with following use case:
* innodb_
* ~50 client connections running short autocommit queries
(UPDATE,
* one UPDATE with INNER JOIN, updating 'queue' table
* one DDL running: CREATE TABLE IF NOT EXISTS queue(...)
The DDL stays hanging in state: "Waiting for table metadata lock", killing this query resolves the deadlock.
Related branches
description: | updated |
summary: |
- unresolved MDL conflict with limited innodb thread concurrency + unresolved MDL deadlock with limited innodb thread concurrency |
To post a comment you must log in.
development head has a number of fixes for DDL processing with respect to MDL usage. These fixes mostly dealt with issues found with FTWRL..
Change set 3567 adds one more fix for this: now BF thread can use wsrep_abort_thd() to kill local state MDL victim.