A serverSetFailoverStatus event in the web server http_plugin.log means that at least one JVM in a cluster is down.

server01: Scanned /opt/WebSphere/Plugins/logs/webserver1/http_plugin.log from line 5064 to 5071. 
Found 2 occurences of the pattern [serverSetFailoverStatus].. 2 crossed warning ( ) or critical (0) threshold.

In the email alert will be a reference to an http_plugin.log file. Near the bottom of the http_plugin.log will be the impacted JVM. In this example, the JVM named Server01 is down.

[30/Nov/2017:03:00:12.94308] 00002509 965f6700 - ERROR: ws_server: serverSetFailoverStatus: Marking Node01_Server01 down

Check the JVMs SystemOut.log  to see if the JVM was automatically restarted. If the JVM is down or was not automatically restarted, determine the PID of the JVM.

ps -ef | grep -i jvm_name

Kill the PID.

kill -i the_pid

Start the JVM.

After the JVM is up and running again, check the SystemOut.log for errors.  Also check the log for stack trace errors. It is also a good idea to look at the log before the JVM was restarted, to see if you can determine what caused the JVM to go down.