Originally when we configured our cluster database on EM12c we used IP addresses simply because DNS wasn’t reliable in the sense setup. Instead of fighting the battle we used IP addresses. Which works provided you never change your IP addresses. Three years later that’s exactly what we did.
During the configuration of the cluster database on EM12c, I realized I completely forgot how to make this happen. So this is really a note to myself for the next time, (side note, DNS is internal and stable so we used the host names) I may have to re-configure the cluster database within EM12c.
The first step of course is to install the agent on all cluster host. This is no different than any other host.
The second step adding the cluster database and high available is done through the manually add target. Remember to select cluster database and not database instance. The instances can’t be added until the cluster database.
After the configurations were complete, status remained pending for the cluster database and the targets all appeared down. What I discovered within the configuration EM12c used the listener machine name as the host for connection string purposes, but that didn’t occur to me until I set specific connection string for both of the instances within the cluster database. All that was required, change the listener machine name.
Why the listener machine name didn’t work? It was the VIP fully qualified. It should, but sadly the VIP fully qualified is not listed in the DNS nor the host file. I don’t have access to either, and the length of time to have this resolved put me on a different path. I changed the listener machine name to the host name. Success.
I didn’t just leave it there, though I put in the proper work order to have the VIPs added to the DNS which is estimated at about 20 business days. The work around provides us with EM working for the cluster database while we wait.