Recently we've had an automatic NameNode failover (no traces why it happended, no warnings/errors or similar). Actually, everything played as it should - standby NN was promoted to an active state and the previously active one went into standby mode. But, this didn't produce any alerts that would notify us about the action. So, I'm wondering if this is actually covered by CM's alerting subsystem?
I've checked CM/CDH's settings and noticed that HDFS configuration parameter "Enable Configuration Change Alerts" is disabled. Based on the description I'd say that turning it on would trigger an alert for this kind of an action. Could anybody share some info on this? Is my assumption correct or ..?
Here's an description for the param: "When set, Cloudera Manager will send alerts when this entity's configuration changes."
Our CDH version is: Cloudera Express 5.4.9 (#9 built by jenkins on 20151214-2009 git: cdc5995a94dd30d599f3534c7a59a68d0744f2f8)