dnf update percona-mysql-router
systemctl restart mysqlrouter
On the primary node connected as icadmin via mysqlsh...
dba.getCluster().listRouters()
If you get this error...
"WARNING: Unable to fetch all configuration options: Please ensure the Router account has the right privileges using <Cluster>.setupRouterAccount() and re-bootstrap Router."
You may need to do this...
dba.getCluster().setupRouterAccount('icadmin', {'update':1})
NOTE: Only do this if you see the WARNINGThen on the machine running the MySQL Router...
mysqlrouter --bootstrap icadmin@node1:3306 --https-port 8444 --user mysqlrouter
NOTE: Only do this if you see the WARNINGdnf clean all
dnf update percona-mysql-shell --enablerepo=Percona-84-Local
dnf list installed | grep -i percona
Assumptions:
Single-Primary
Ensure you have a backup and you know how to restore it.
Ensure all important connections use MySQL Router.
Encourage developers to write idempotent code (an operation is idempotent if performing it multiple times has the same effect as performing it once).
On any node...
mysqlsh
\c icadmin@node1
where icadmin is the name of your cluster admin accountdba.getCluster().status()
Check "status" of the replica set is "OK" and the "status" of all nodes is "ONLINE".
Choose one of the nodes with a "memberRole" of "SECONDARY" as the first node to patch and, on that node, do (as root or via sudo)...
dnf clean all
dnf update percona-server-server percona-server-client percona-server-shared percona-icu-data-files --enablerepo=Percona-84-Local
dnf list installed | grep -i percona
On any node...
mysqlsh
\c icadmin@node1
where icadmin is the name of your cluster admin accountdba.getCluster().status()
Check "version" of the node you patched shows the correct version number.
Repeat for the rest of the SECONDARY nodes.
Finally, perform the same steps on the Primary node...
TODO
dnf clean all
dnf update percona-xtrabackup-84 --enablerepo=Percona-84-Local
dnf list installed | grep -i percona