Has anyone successfully set up Db2 11.5.5 in a Windows Server Failover Clustering (WSFC) environment? Do DRDA connections work for you?
Several years ago, I had no problem setting up Db2 11.1 (though it was a tough slog) and documented the setup here:
However, I am not having similar success with 11.5.5fp1 - I have set up the cluster and can successfully move the role between nodes and can access the database with command line (e.g. "db2 connect to MYDB") or with an ODBC connection that uses DBALIAS=MYDB.
However, when I try to access the database using DRDA (JCC driver, our own ODBC driver "StarSQL", or the IBM Db2 ODBC driver configured to use host/port/database parameters rather than DBALIAS), I get the following error:
[IBM][CLI Driver] SQL1031N The database directory cannot be found on the indicated file system. SQLSTATE=58031
It does not matter whether I use the cluster name, individual hostname, or localhost - the problem seems to be on the Db2 side, but there is nothing in db2diag.log.
IBM Support has not been helpful and says they don't have anyone familiar enough with MCSC to understand what is going on.
So I am reaching out to see if anyone else has been successful (i.e. I would like confirmation that it should actually work, or not), or if anyone else is encountering the above problem.
Hi,that's interesting because the client does not really know that you are using a cluster. Like you we used MSCS years ago without problems. Did you use the highly availabe IP adress in the JDBC driver and the Db2 node directory which was defined here https://www.ibm.com/docs/en/db2/11.5?topic=commands-db2iclus-microsoft-cluster-server under point 10?From Cluster Administrator, create a new resource type of type "IP Address" called "mscs5" that resides on the Public Network. This resource should also belong to Db2 Group. This will be a highly available IP address, and this address should not correspond to any machine on the network. Bring the IP Address resource type online and ensure that the address can be pinged from a remote machine.RegardsGerald
Did you use Virtual IP in defining connection string using host/port?
This Virtual IP - when defined as floating resource in MSCS should be following current primary host in my opinion - but I have only experience with TSM resources.
you have to use static IP adresses in DHCP for all cluster members and the virtual IP adress must be defined in the Db2 node directory or in the JDBC connection string with port 50000(=default).