stack elasticsearch + fluentd

3
votes

I am setting up fluentd and elasticsearch on a local VM in order to try the fluentd and ES stack.

OS: centos (recent)

[root@localhost data]# cat /etc/redhat-release 
CentOS release 6.5 (Final)

I am elasticsearch up and running on localhost (I used it with logstash with no issue)

[root@localhost data]# curl -X GET http://localhost:9200/
{
  "status" : 200,
  "name" : "Simon Williams",
  "version" : {
    "number" : "1.2.1",
    "build_hash" : "6c95b759f9e7ef0f8e17f77d850da43ce8a4b364",
    "build_timestamp" : "2014-06-03T15:02:52Z",
    "build_snapshot" : false,
    "lucene_version" : "4.8"
  },
  "tagline" : "You Know, for Search"
}

I have installed td-agent following the installation notes from fluentd website. I am using that configuration file:

<source>
  type tail
  path /tmp/data/log
  pos_file /tmp/data/log.pos
  format /^(?<host>[^ ]*) [^ ]* (?<user>[^ ]*) \[(?<time>[^\]]*)\] "(?<method>\S+)(?: +(?<path>[^ ]*) +\S*)?" (?<code>[^ ]*) (?<size>[^ ]*)(?: "(?<referer>[\
^\"]*)" "(?<agent>[^\"]*)")?/
  time_format %d/%b/%Y:%H:%M:%S %z
  tag front.nginx.access
</source>

<match front.nginx.access>
  type elasticsearch
  host localhost
  port 9200
  index_name fluentd
  type_name nginx
  include_tag_key

  # buffering                                                                                                                                                
  buffer_type file
  buffer_path /tmp/fluentd/buffer/
  flush_interval 10s
  buffer_chunk_limit 16m
  buffer_queue_limit 4096
  retry_wait 15s
</match>

Here is the start-up log:

2014-07-24 13:39:58 +0200 [info]: starting fluentd-0.10.50
2014-07-24 13:39:58 +0200 [info]: reading config file path="/etc/td-agent/td-agent.conf"
2014-07-24 13:39:58 +0200 [info]: gem 'fluent-mixin-config-placeholders' version '0.2.4'
2014-07-24 13:39:58 +0200 [info]: gem 'fluent-mixin-plaintextformatter' version '0.2.6'
2014-07-24 13:39:58 +0200 [info]: gem 'fluent-plugin-elasticsearch' version '0.3.1'
2014-07-24 13:39:58 +0200 [info]: gem 'fluent-plugin-flume' version '0.1.1'
2014-07-24 13:39:58 +0200 [info]: gem 'fluent-plugin-mongo' version '0.7.3'
2014-07-24 13:39:58 +0200 [info]: gem 'fluent-plugin-parser' version '0.3.4'
2014-07-24 13:39:58 +0200 [info]: gem 'fluent-plugin-rewrite-tag-filter' version '1.4.1'
2014-07-24 13:39:58 +0200 [info]: gem 'fluent-plugin-s3' version '0.4.0'
2014-07-24 13:39:58 +0200 [info]: gem 'fluent-plugin-scribe' version '0.10.10'
2014-07-24 13:39:58 +0200 [info]: gem 'fluent-plugin-td' version '0.10.20'
2014-07-24 13:39:58 +0200 [info]: gem 'fluent-plugin-td-monitoring' version '0.1.2'
2014-07-24 13:39:58 +0200 [info]: gem 'fluent-plugin-webhdfs' version '0.2.2'
2014-07-24 13:39:58 +0200 [info]: gem 'fluentd' version '0.10.50'
2014-07-24 13:39:58 +0200 [info]: using configuration file: <ROOT>
  <source>
    type tail
    path /tmp/data/log
    pos_file /tmp/data/log.pos
    format /^(?<host>[^ ]*) [^ ]* (?<user>[^ ]*) \[(?<time>[^\]]*)\] "(?<method>\S+)(?: +(?<path>[^ ]*) +\S*)?" (?<code>[^ ]*) (?<size>[^ ]*)(?: "(?<referer>[^\"]*)" "(?<agent>[^\"]*)")?/
    time_format %d/%b/%Y:%H:%M:%S %z
    tag front.nginx.access
  </source>
  <match front.nginx.access>
    type elasticsearch
    host localhost
    port 9200
    index_name fluentd
    type_name nginx
    include_tag_key 
    buffer_type file
    buffer_path /tmp/fluentd/buffer/
    flush_interval 10s
    buffer_chunk_limit 16m
    buffer_queue_limit 4096
    retry_wait 15s
  </match>
</ROOT>
2014-07-24 13:39:58 +0200 [info]: adding source type="tail"
2014-07-24 13:39:58 +0200 [info]: adding match pattern="front.nginx.access" type="elasticsearch"
2014-07-24 13:39:58 +0200 [info]: following tail of /tmp/data/log

I get that error:

2014-07-24 13:40:00 +0200 [warn]: temporarily failed to flush the buffer. next_retry=2014-07-24 13:40:13 +0200 error_class="Elasticsearch::Transport::Transport::Errors::ServiceUnavailable" error="[503] " instance=70247139359260
  2014-07-24 13:40:00 +0200 [warn]: /usr/lib64/fluent/ruby/lib/ruby/gems/1.9.1/gems/elasticsearch-transport-0.4.11/lib/elasticsearch/transport/transport/base.rb:132:in `__raise_transport_error'
  2014-07-24 13:40:00 +0200 [warn]: /usr/lib64/fluent/ruby/lib/ruby/gems/1.9.1/gems/elasticsearch-transport-0.4.11/lib/elasticsearch/transport/transport/base.rb:227:in `perform_request'
  2014-07-24 13:40:00 +0200 [warn]: /usr/lib64/fluent/ruby/lib/ruby/gems/1.9.1/gems/elasticsearch-transport-0.4.11/lib/elasticsearch/transport/transport/http/faraday.rb:20:in `perform_request'
  2014-07-24 13:40:00 +0200 [warn]: /usr/lib64/fluent/ruby/lib/ruby/gems/1.9.1/gems/elasticsearch-transport-0.4.11/lib/elasticsearch/transport/client.rb:92:in `perform_request'
  2014-07-24 13:40:00 +0200 [warn]: /usr/lib64/fluent/ruby/lib/ruby/gems/1.9.1/gems/elasticsearch-api-0.4.11/lib/elasticsearch/api/actions/ping.rb:19:in `ping'
  2014-07-24 13:40:00 +0200 [warn]: /usr/lib64/fluent/ruby/lib/ruby/gems/1.9.1/gems/fluent-plugin-elasticsearch-0.3.1/lib/fluent/plugin/out_elasticsearch.rb:46:in `client'
  2014-07-24 13:40:00 +0200 [warn]: /usr/lib64/fluent/ruby/lib/ruby/gems/1.9.1/gems/fluent-plugin-elasticsearch-0.3.1/lib/fluent/plugin/out_elasticsearch.rb:103:in `send'
  2014-07-24 13:40:00 +0200 [warn]: /usr/lib64/fluent/ruby/lib/ruby/gems/1.9.1/gems/fluent-plugin-elasticsearch-0.3.1/lib/fluent/plugin/out_elasticsearch.rb:98:in `write'
  2014-07-24 13:40:00 +0200 [warn]: /usr/lib64/fluent/ruby/lib/ruby/gems/1.9.1/gems/fluentd-0.10.50/lib/fluent/buffer.rb:296:in `write_chunk'
  2014-07-24 13:40:00 +0200 [warn]: /usr/lib64/fluent/ruby/lib/ruby/gems/1.9.1/gems/fluentd-0.10.50/lib/fluent/buffer.rb:276:in `pop'
  2014-07-24 13:40:00 +0200 [warn]: /usr/lib64/fluent/ruby/lib/ruby/gems/1.9.1/gems/fluentd-0.10.50/lib/fluent/output.rb:310:in `try_flush'
  2014-07-24 13:40:00 +0200 [warn]: /usr/lib64/fluent/ruby/lib/ruby/gems/1.9.1/gems/fluentd-0.10.50/lib/fluent/output.rb:132:in `run'

running tcpdump on port 9200, I get nothing...

tcpdump -x -X -i any 'port 9200'
1
elasticsearchfluentd
try 9300 - its default transport (not http) port – Konstantin V. Salikhov
I just tried and I get nothing as well on tcpdump, same error in log as before – Alexandre Mélard
Possibly seLinux getting in the way -- try setenforce Permissive and see if it helps. – Alcanzar
I've tryed setenforce Permissive with no success (note I have logstash running without problem) – Alexandre Mélard
Have you ever solved this? I'm running in the same issue and I am at loss... :( – mac

1 Answers

1
votes

I've found the problem.

Actually, I had not modified the default cluster name in ES. Another ES cluster existed on the same network. The clients used in this cluster where sending packets to my ES cluster with an ancient protocol.

I have corrected all issues by changing the ES cluster name.