Hi Loth, I know some setup with 3-nodes mariadb cluster and nginx (not haproxy) which work fine.
Could you please share more information, for example mariadb status, error logs in cloudstack management ? -Wei On Sun, 3 Oct 2021 at 07:29, Loth <k...@linuxdigital.net> wrote: > Hi Users, > > I am seeing some deadlocks in my sql error log; > "[ERROR] mysqld: Deadlock found when trying to get lock; try restarting > transaction" > > My environment is a 3-node Galera cluster behind HAProxy, with two as > "backup" and one active. Usually this performs fine, but recently these > deadlocks have been increasing. > However as far as I know this should not occur since we are running one > database at a time and usually these occur when modifying the same data. > Does anyone else run a similar setup or have any tips on how we can solve > this? > > Thanks for any help you can provide! >