Just a FYI for anyone else who hits this: I just upgraded to logstash 2.3 and my ELK pipeline broke because my indexers were not able to read from my redis instances, which were running redis 2.4.10 (Default RHEL 6 version). The errors included connection errors, and exception regarding an unknown command:
{:timestamp=>"2016-04-04T08:44:26.088000-0700", :message=>"Redis connection problem", :exception=>#<Redis::CommandError: ERR unknown command 'script'>, :level=>:warn}
{:timestamp=>"2016-04-04T08:44:26.088000-0700", :message=>"Redis connection problem", :exception=>#<Redis::CommandError: ERR unknown command 'script'>, :level=>:warn}
{:timestamp=>"2016-04-04T08:44:26.088000-0700", :message=>"Redis connection problem", :exception=>#<Redis::CommandError: ERR unknown command 'script'>, :level=>:warn}
{:timestamp=>"2016-04-04T08:44:26.089000-0700", :message=>"Redis connection problem", :exception=>#<Redis::CannotConnectError: Error connecting to Redis on xxxxxxx:6379 (Errno::ECONNREFUSED)>, :level=>:warn}
{:timestamp=>"2016-04-04T08:44:26.089000-0700", :message=>"Redis connection problem", :exception=>#<Redis::CannotConnectError: Error connecting to Redis on xxxxxx:6379 (Errno::ECONNREFUSED)>, :level=>:warn}
{:timestamp=>"2016-04-04T08:44:26.089000-0700", :message=>"Redis connection problem", :exception=>#<Redis::CannotConnectError: Error connecting to Redis on xxxxxxx:6379 (Errno::ECONNREFUSED)>, :level=>:warn}
{:timestamp=>"2016-04-04T08:44:26.089000-0700", :message=>"Redis connection problem", :exception=>#<Redis::CannotConnectError: Error connecting to Redis on xxxxxx:6379 (Errno::ECONNREFUSED)>, :level=>:warn}
{:timestamp=>"2016-04-04T08:44:26.090000-0700", :message=>"Redis connection problem", :exception=>#<Redis::CannotConnectError: Error connecting to Redis on xxxxxxx:6379 (Errno::ECONNREFUSED)>, :level=>:warn}
I was able to resolve this by upgrading to redis 3.0.6. The redis input docs say that redis 2.6+ is recommended, but I think it is a requirement....
John Ouellette