Aliyun ODPS Plugin for Fluentd

Getting Started


Introduction

  • ODPS-Open Data Processing Service is a massive data processing platform designed by alibaba.
  • DHS-ODPS DataHub Service is a service in Odps, which provides real-time upload and download functions for user.

Requirements

To get started using this plugin, you will need three things:

  1. Ruby 2.1.0 or later
  2. Gem 2.4.5 or later
  3. Fluentd-0.10.49 or later (Home Page)
  4. Protobuf-3.5.1 or later(Ruby protobuf)

Install the Plugin

install the project from gem or github:

$ gem install fluent-plugin-aliyun-odps
$ git clone https://github.com/aliyun/aliyun-odps-fluentd-plugin.git

Use gem to install dependency:

$ gem install protobuf
$ gem install fluentd --no-ri --no-rdoc

Your plugin is in aliyun-odps-fluentd-plugin/lib/fluent/plugin, entry file is out_odps.rb.

Use the Plugin

  • If you installed this plugin from gem, please ignore this step.
  • Move the plugin dir into the plugin directory of Fluentd.
  • (i.e., copy the folder aliyun-odps-fluentd-plugin/lib/fluent/plugin into YOUR_FLUENTD_DIRECTORY/lib/fluent/plugin).
$ cp aliyun-odps-fluentd-plugin/lib/fluent/plugin/* {YOUR_FLUENTD_DIRECTORY}/lib/fluent/plugin/ -r

ODPS Fluentd plugin now is available. Following is a simple example of how to write ODPS output configuration.

<source>
   type tail
   path /opt/log/in/in.log
   pos_file /opt/log/in/in.log.pos
   refresh_interval 5s
   tag in.log
   format /^(?<remote>[^ ]*) - - \[(?<datetime>[^\]]*)\] "(?<method>\S+)(?: +(?<path>[^\"]*?)(?: +\S*)?)?" (?<code>[^ ]*) (?<size>[^ ]*) "-" "(?<agent>[^\"]*)"$/
   time_format %Y%b%d %H:%M:%S %z
</source>
<match in.**>
  type aliyun_odps
  aliyun_access_id ************
  aliyun_access_key *********
  aliyun_odps_endpoint http://service.odps.aliyun.com/api
  aliyun_odps_hub_endpoint http://dh.odps.aliyun.com
  buffer_chunk_limit 2m
  buffer_queue_limit 128
  flush_interval 5s
  project your_projectName
  <table in.log>
    table your_tableName
    fields remote,method,path,code,size,agent
    partition ctime=${datetime.strftime('%Y%m%d')}
    time_format %d/%b/%Y:%H:%M:%S %z
    shard_number 1
  </table>
</match>

Parameters

  • type(Fixed): always be aliyun_odps.
  • aliyun_access_id(Required):your aliyun access id.
  • aliyun_access_key(Required):your aliyun access key.
  • aliyun_odps_hub_endpoint(Required):if you are using ECS, set it as http://dh-ext.odps.aliyun-inc.com, otherwise using http://dh.odps.aliyun.com.
  • aliyunodps_endpoint(Required):if you are using ECS, set it as http://odps-ext.aiyun-inc.com/api, otherwise using http://service.odps.aliyun.com/api .
  • buffer_chunk_limit(Optional):chunk size,??k?? (KB), ??m?? (MB), and ??g?? (GB) ??default 8MB??recommended number is 2MB.
  • buffer_queue_limit(Optional):buffer chunk size??example: buffer_chunk_limit2m??buffer_queue_limit 128??then the total buffer size is 2*128MB.
  • flush_interval(Optional):interval to flush data buffer, default 60s.
  • project(Required):your project name.
  • table(Required):your table name.
  • fields(Required): must match the keys in source.
  • partition(Optional)??set this if your table is partitioned.
    • partition format:
      • fix string: partition ctime=20150804
      • key words: partition ctime=$remote
      • key words int time format: partition ctime=$datetimedatetime.strftime('%Y%m%d')
  • time_format(Optional):
    • if the odps table is partitioned, you need to set the param
    • if you are using the key words to set your and the key word is in time format, please set the param . example: source[datetime] = "29/Aug/2015:11:10:16 +0800", and the param is "%d/%b/%Y:%H:%M:%S %z"
  • shard_number(Optional):less than the number you set when create the hubtable.

Authors && Contributors


License


licensed under the Apache License 2.0