Fresh ELK Install on Ubuntu 14.04, Kibana returns Connection Refused -

new server running ubuntu 14.04 server. fresh install of elk stack today. check elastic search spyder on home ubuntu machine. result (using python's requests package):

res = requests.get('http://' + falcon618 + ':' + port)  print(res.content) {   "status" : 200,   "name" : "six of nine",   "cluster_name" : "{my cluster name}",   "version" : {     "number" : "1.4.4",     "build_hash" : "c88f77ffc81301dfa9dfd81ca2232f09588bd512",     "build_timestamp" : "2015-02-19t13:05:36z",     "build_snapshot" : false,     "lucene_version" : "4.10.3"   },   "tagline" : "you know, search" } 

however, when go server_ip:5601 via chrome connection refused. when ask ubuntu kibana's status says running. when @ netstat -lnp there entry (listening) port 9200 (elaticsearch) no entry port 5601.

i have searched , searched google can't find kibana error log is. nothing in elasticsearch's logs suggest issue , home machine can see es via server's ip. i'm attempting run kibana same server. in kibana's config:

# kibana served end server. controls port use. port: 5601  # host bind server to. host: "{server ip}"  # elasticsearch instance use queries. elasticsearch_url: "http://{server ip}:9200" 

first, kibana log? second, ideas on why happening? i've read several threads on people having similar error message none of them quite match.

i followed page installing elk: elk on ubuntu 14.04

however, skipped nginx install , made es available on public ip. (using iptables reject not known server. iptables empty.)

--------- edit clarify comments below ----------

my first attempt @ elk stack included kibana on same server elasticsearch running on. however, since wanted elasticsearch available home servers told listen server's public ip. i've configured kibana on server elasticsearch on server's ip.

i test elasticsearch , have added approximately 3000 docs using home server. can retrieve docs home server well. set kibana on home server see if work. if kibana worked on lan server suggest maybe simple reinstall might fix kibana install on wan server. or possibly permissions or security issue.

the kibana install on wan server has returned err_connection_refused in chrome , still does. kibana install on lan server shows kibana won't let me designate default index pattern. first page come asks pattern there no button save pattern. have attempted add pattern via advanced setting , doesn't save.

running _cat/indices on elasticsearch returns following:

yellow open visitors 5 1 3282 0 54.8mb 54.8mb  yellow open .kibana  1 1    1 0  2.5kb  2.5kb 

so there .kibana index visitors index. first time ran api call took while display. however, since these indices have come fast.

----- edit #2 ------

i have tried install kibana on wan server lease. results identical lan server. loads there still no button save it. when enter in index pattern , hit return 3 white dots in upper right cycle. after long while "no living connections" error.

it doesn't appear able connect es seems timeout issue think. wan server #2 can curl es , proper response takes long time - possibly full minute. when use curl indices didn't time seemed take @ least couple of minutes.

on top of i've been wondering visitors index. has approximately 3000 entries take 3k of disk space each total of approximately 9m. however, in index take on 50m.

finally, while es appears working slow. when search right answer after several seconds. entire wordpress site can load faster (on wan server #2) can 3k worth of json text. (both wan server identical specs - 8 core amd, 32 gb ram, ubuntu 14.04 100 mbps pipe.)

-------- edit #3 --------

while still have not found solution have piece puzzle. digitalocean offers elk stack ready go got one. worked fine , see data in kibana , looked local server log data. great!

however, changed elasticsearch cluster name , restarted es. kibana looked have same issues describe above can't define default index pattern.

so rebooted machine. 502 bad gateway error. changed es default cluster name - elasticsearch - , rebooted. still 502.

hmmmm. here's hoping digitalocean's support can figure out.

-------- edit #4 ---------

after wasting way time on started digitalocean image of elk stack on ubuntu 14.04. broke changed cluster name in elsaticsearch.yml config file.

however, support's figured out cluster parameter needs added logstash config files well. use do's example of cluster named snuggles:

/etc/logstash/conf.d/30-lumberjack-output.conf:

output {   elasticsearch { host => localhost cluster => snuggles }   stdout { codec => rubydebug } } 

/etc/logstash/conf.d/01-lumberjack-input.conf:

input {   lumberjack {     port => 5000     type => "logs"     cluster => "snuggles"     ssl_certificate => "/etc/pki/tls/certs/logstash-forwarder.crt"     ssl_key => "/etc/pki/tls/private/logstash-forwarder.key"   }  } 

/etc/elasticsearch/elasticsearch.yml:

cluster.name: snuggles 

this got working on do's image. assume image based on tutorial: elk on ubuntu 14.04

i still have not gotten stack working on other machine. marvel plugin working on original - non-do - server , displaying nice data cluster... of 1.