Indeed, even after used the new parameter, this Helm chart seem to doesn't work.
I have made some tests on my side and I can confirm to you that the root cause of this bug isn't linked to our managed Kubernetes cluster.
MariaDB pod seems to be in
CrashLoopBackOff status due to informations reported by his
Here, some logs from my local tests :
Liveness probe failed: mysqladmin: connect to server at 'localhost' failed
error: 'Access denied for user 'root'@'localhost' (using password: YES)'
This bug also impact the
WordPress pod which can't connect to his database :
Error executing 'postInstallation': Failed to connect to test-mariadb:3306 after 36 tries
This Github issue seems to be potentially linked to your problem : [stable/mariadb] probe failed if you set rootUser password
Can you tried to use a previous release of WordPress Helm Chart ?