MS SQL Server databases can be configured as a cluster. This provides a high-availability database environment for SQL Server clients. For example, if database node01 fails, the SQL Server clients connected to node01 are routed to the next active database node.
In Netprobe version GA6.2 (COL-11312) and above, you can now enable or disable the MultiSubnetFailover option in the SQL Toolkit plugin for the Microsoft SQL Server database. This setting is disabled by default.
Users with older Netprobe versions are recommended to upgrade, as SQL-Toolkit plugin previously did not support a MS SQL Server database cluster. Because of this, the error "Unable to complete login process due to delay in opening server connection" is encountered by the said plugin when its target database node01 fails or is offline.
One workaround suggested in the Microsoft website is to reconfigure the database cluster's HostRecordTTL so that clients can reconnect quickly to the next available node:
By default, clients cache cluster DNS records for 20 minutes. By reducing HostRecordTTL, the Time to Live (TTL), for the cached record, legacy clients may reconnect more quickly. However, reducing the HostRecordTTL setting may also result in increased traffic to the DNS servers.
Please sign in to leave a comment.