Join this online group to communicate across IBM product users and experts by sharing advice and best practices with peers and staying up to date regarding product enhancements.
Hi,
I am wondering how they would have avoid the single point of failuer using broker is being connected to all the clients.
In other words, how can we implement clustering concepts in the broker to avoid single point of failuer.
Thanks in advance…
-Kali
Use an OS active/passive cluster for the Broker. If the active node fails, the OS will make the passive node the active node. The clients will reconnect to the newly active node. Normally the cutover happens within 5-10 minutes (depending upon cluster configuration). This results in service delay, rather than an outage.