Adjustment to Redis client's Seninel example in order to confirm that blocking operations are booted during a failover. (https://github.com/redis/redis-rb/blob/98e3e7a516fc9b4609bc8ab482605f835a4de621/examples/sentinel.rb)

sentinal.rb

require 'redis'

# This example creates a master-slave setup with a sentinel, then connects to
# it and sends write commands in a loop.
#
# After 30 seconds, the master dies. You will be able to see how a new master
# is elected and things continue to work as if nothing happened.
#
# To run this example:
#
#   $ ruby -I./lib examples/sentinel.rb
#

at_exit do
  begin
    Process.kill(:INT, $redises)
  rescue Errno::ESRCH
  end

  Process.waitall
end

$redises = spawn("examples/sentinel/start")

Sentinels = [{:host => "127.0.0.1", :port => 26379},
             {:host => "127.0.0.1", :port => 26380}]
r = Redis.new(:url => "redis://master1", :sentinels => Sentinels, :role => :master)

# Set keys into a loop.
#
# The example traps errors so that you can actually try to failover while
# running the script to see redis-rb reconfiguring.
(0..1000000).each{|i|
    begin
        # Still, write to the cluster (for replication purposes)
        r.set(i,i)
        $stdout.write("SET (#{i} times)\n") if i % 100 == 0

        $stdout.write("BLPOP #{i}\n")
        # This should block as long as the connection is open
        r.blpop("foo", 0)
    rescue => e
        $stdout.write("E\n")
        $stdout.write("#{e.message}\n\n")
    end
    sleep(1)
}