Continual wsrep in pre-commit stage lock ups across several clusters
We are testing with different nodes. It seems no matter what these clusters lock up on this:
wsrep in pre-commit stage
We need answers of why this happens, and if Galera cluster is a sustainable technology
From what I see the problem is similar to what you described in the other thread, http://support.severalnines.com/entries/83817265
Our questions are still valid, please provide us with answers because they are needed to actually start assessing the problem. Culprit may be hidden and not easy to find so I can't tell if this data will be enough but right now we don't have any informations of what is going on in your system.
Can you please provide following data:
SHOW FULL PROCESSLIST;
SHOW GLOBAL VARIABLES\G
SHOW STATUS LIKE 'wsrep%';
SHOW ENGINE InnoDB STATUS\G
Please sign in to leave a comment.