I don't know what is this "Database HA" feature but I can tell you what I do.

I use Kafka channels and have multiple agents with the same configuration. In front of the agents I have an http load balancer.
That way, any agent can accept requests and any agent can process them once in the channel.

In addition, Cloudera Manager monitors are restarts agents if they die.

Regards,
Gonzalo

On 4 December 2015 at 16:18, Zhishan Li <zhishanlee@gmail.com> wrote:
How to implement the “Database HA” failover feature?

I have read https://flume.apache.org/FlumeUserGuide.html#handling-agent-failures but still without ideas. Please show me more details information?

Thanks