[MCOL-3615] Table locks surviving cluster restarts Created: 2019-11-15  Updated: 2020-08-25  Resolved: 2019-11-15

Status: Closed
Project: MariaDB ColumnStore
Component/s: MariaDB Server
Affects Version/s: 1.2.5
Fix Version/s: Icebox

Type: Bug Priority: Major
Reporter: Rick Pizzi Assignee: Unassigned
Resolution: Not a Bug Votes: 0
Labels: None


 Description   

A table lock caused by a failed truncate will not be cleared even after a complete cluster restart:

Table LockID Process PID Session Txn CreationTime State DBRoots
 
dm_temp.tablename 1 DDLProc 5444 775 573937 Thu Nov 14 20:17:18 2019 LOADING 1,2

A cleartablelock will probably remove it, but it is not proper behaviour that a tablelock survives a cluster restart.
The table remains locked, affecting normal production workflow as data can't be inserted until the lock is cleared.

 Comments   
Comment by Andrew Hutchings (Inactive) [ 2019-11-15 ]

DDLProc table locks held during a restart can't be rolled back and have to be manually removed. This is for safety.

Generated at Thu Feb 08 02:44:07 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.