fluent-plugin-redshift 0.0.1
Sign up to get free protection for your applications and to get access to all the features.
- data/.gitignore +19 -0
- data/Gemfile +3 -0
- data/README.md +131 -0
- data/Rakefile +16 -0
- data/VERSION +1 -0
- data/fluent-plugin-redshift.gemspec +25 -0
- data/lib/fluent/plugin/out_redshift.rb +230 -0
- data/test/plugin/test_out_redshift.rb +395 -0
- data/test/test_helper.rb +8 -0
- metadata +152 -0
data/.gitignore
ADDED
data/Gemfile
ADDED
data/README.md
ADDED
@@ -0,0 +1,131 @@
|
|
1
|
+
Amazon Redshift output plugin for Fluentd
|
2
|
+
========
|
3
|
+
|
4
|
+
## Overview
|
5
|
+
|
6
|
+
Amazon Redshift output plugin uploads event logs to an Amazon Redshift Cluster. Supportted data formats are csv, tsv and json. An S3 bucket and a Redshift Cluster are required to use this plugin.
|
7
|
+
|
8
|
+
## Installation
|
9
|
+
|
10
|
+
gem install fluent-plugin-redshift
|
11
|
+
|
12
|
+
## Configuration
|
13
|
+
|
14
|
+
Format:
|
15
|
+
|
16
|
+
<match my.tag>
|
17
|
+
type redshift
|
18
|
+
|
19
|
+
# s3 (for copying data to redshift)
|
20
|
+
aws_key_id YOUR_AWS_KEY_ID
|
21
|
+
aws_sec_key YOUR_AWS_SECRET_KEY
|
22
|
+
s3_bucket YOUR_S3_BUCKET
|
23
|
+
s3_endpoint YOUR_S3_BUCKET_END_POINT
|
24
|
+
path YOUR_S3_PATH
|
25
|
+
timestamp_key_format year=%Y/month=%m/day=%d/hour=%H/%Y%m%d-%H%M
|
26
|
+
|
27
|
+
# redshift
|
28
|
+
redshift_host YOUR_AMAZON_REDSHIFT_CLUSTER_END_POINT
|
29
|
+
redshift_port YOUR_AMAZON_REDSHIFT_CLUSTER_PORT
|
30
|
+
redshift_dbname YOUR_AMAZON_REDSHIFT_CLUSTER_DATABASE_NAME
|
31
|
+
redshift_user YOUR_AMAZON_REDSHIFT_CLUSTER_USER_NAME
|
32
|
+
redshift_password YOUR_AMAZON_REDSHIFT_CLUSTER_PASSWORD
|
33
|
+
redshift_tablename YOUR_AMAZON_REDSHIFT_CLUSTER_TARGET_TABLE_NAME
|
34
|
+
file_type [tsv|csv|json]
|
35
|
+
|
36
|
+
# buffer
|
37
|
+
buffer_type file
|
38
|
+
buffer_path /var/log/fluent/redshift
|
39
|
+
flush_interval 15m
|
40
|
+
buffer_chunk_limit 1g
|
41
|
+
</match>
|
42
|
+
|
43
|
+
Example (watch and upload json formatted apache log):
|
44
|
+
|
45
|
+
<source>
|
46
|
+
type tail
|
47
|
+
path redshift_test.json
|
48
|
+
pos_file redshift_test_json.pos
|
49
|
+
tag redshift.json
|
50
|
+
format /^(?<log>.*)$/
|
51
|
+
</source>
|
52
|
+
|
53
|
+
<match redshift.json>
|
54
|
+
type redshift
|
55
|
+
|
56
|
+
# s3 (for copying data to redshift)
|
57
|
+
aws_key_id YOUR_AWS_KEY_ID
|
58
|
+
aws_sec_key YOUR_AWS_SECRET_KEY
|
59
|
+
s3_bucket hapyrus-example
|
60
|
+
s3_endpoint s3.amazonaws.com
|
61
|
+
path apache_json_log
|
62
|
+
timestamp_key_format year=%Y/month=%m/day=%d/hour=%H/%Y%m%d-%H%M
|
63
|
+
|
64
|
+
# redshift
|
65
|
+
redshift_host xxx-yyy-zzz.xxxxxxxxxx.us-east-1.redshift.amazonaws.com
|
66
|
+
redshift_port 5439
|
67
|
+
redshift_dbname fluent-redshift-test
|
68
|
+
redshift_user fluent
|
69
|
+
redshift_password fluent-password
|
70
|
+
redshift_tablename apache_log
|
71
|
+
file_type json
|
72
|
+
|
73
|
+
# buffer
|
74
|
+
buffer_type file
|
75
|
+
buffer_path /var/log/fluent/redshift
|
76
|
+
flush_interval 15m
|
77
|
+
buffer_chunk_limit 1g
|
78
|
+
<match>
|
79
|
+
|
80
|
+
+ `type` (required) : The value must be `redshift`.
|
81
|
+
|
82
|
+
+ `aws_key_id` (required) : AWS access key id to access s3 bucket.
|
83
|
+
|
84
|
+
+ `aws_sec_key` (required) : AWS securet key id to access s3 bucket.
|
85
|
+
|
86
|
+
+ `s3_bucket` (required) : s3 bucket name. S3 bucket must be same as the region of your Redshift cluster.
|
87
|
+
|
88
|
+
+ `s3_endpoint` : s3 endpoint.
|
89
|
+
|
90
|
+
+ `path` (required) : s3 path to input.
|
91
|
+
|
92
|
+
+ `timestamp_key_format` : The format of the object keys. It can include date-format directives.
|
93
|
+
|
94
|
+
- Default parameter is "year=%Y/month=%m/day=%d/hour=%H/%Y%m%d-%H%M"
|
95
|
+
- For example, the s3 path is as following with the above example configration.
|
96
|
+
<pre>
|
97
|
+
hapyrus-example/apache_json_log/year=2013/month=03/day=05/hour=12/20130305_1215_00.gz
|
98
|
+
hapyrus-example/apache_json_log/year=2013/month=03/day=05/hour=12/20130305_1230_00.gz
|
99
|
+
</pre>
|
100
|
+
|
101
|
+
+ `redshift_host` (required) : the end point(or hostname) of your Amazon Redshift cluster.
|
102
|
+
|
103
|
+
+ `redshift_port` (required) : port number.
|
104
|
+
|
105
|
+
+ `redshift_dbname` (required) : database name.
|
106
|
+
|
107
|
+
+ `redshift_user` (required) : user name.
|
108
|
+
|
109
|
+
+ `redshift_password` (required) : password for the user name.
|
110
|
+
|
111
|
+
+ `redshift_tablename` (required) : table name to store data.
|
112
|
+
|
113
|
+
+ `file_type` : file format of the source data. `csv`, `tsv` or `json` are available.
|
114
|
+
|
115
|
+
+ `delimiter` : delimiter of the source data. This option will be ignored if `file_type` is specified.
|
116
|
+
|
117
|
+
+ `buffer_type` : buffer type.
|
118
|
+
|
119
|
+
+ `buffer_path` : path prefix of the files to buffer logs.
|
120
|
+
|
121
|
+
+ `flush_interval` : flush interval.
|
122
|
+
|
123
|
+
+ `buffer_chunk_limit` : limit buffer size to chunk.
|
124
|
+
|
125
|
+
+ `utc` : utc time zone. This parameter affects `timestamp_key_format`.
|
126
|
+
|
127
|
+
## License
|
128
|
+
|
129
|
+
Copyright (c) 2013 [Hapyrus Inc](http://hapyrus.com)
|
130
|
+
|
131
|
+
[Apache License, Version 2.0](http://www.apache.org/licenses/LICENSE-2.0)
|
data/Rakefile
ADDED
@@ -0,0 +1,16 @@
|
|
1
|
+
require "bundler"
|
2
|
+
Bundler::GemHelper.install_tasks
|
3
|
+
require 'rake/testtask'
|
4
|
+
|
5
|
+
Rake::TestTask.new(:test) do |test|
|
6
|
+
test.libs << 'lib' << 'test'
|
7
|
+
test.test_files = FileList['test/plugin/*.rb']
|
8
|
+
test.verbose = true
|
9
|
+
end
|
10
|
+
|
11
|
+
task :coverage do |t|
|
12
|
+
ENV['COVERAGE'] = '1'
|
13
|
+
Rake::Task["test"].invoke
|
14
|
+
end
|
15
|
+
|
16
|
+
task :default => [:build]
|
data/VERSION
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
0.0.1
|
@@ -0,0 +1,25 @@
|
|
1
|
+
# -*- encoding: utf-8 -*-
|
2
|
+
$:.push File.expand_path('../lib', __FILE__)
|
3
|
+
|
4
|
+
Gem::Specification.new do |gem|
|
5
|
+
gem.name = "fluent-plugin-redshift"
|
6
|
+
gem.version = File.read("VERSION").strip
|
7
|
+
gem.authors = ["Masashi Miyazaki"]
|
8
|
+
gem.email = ["mmasashi@gmail.com"]
|
9
|
+
gem.description = %q{Amazon Redshift output plugin for Fluentd}
|
10
|
+
gem.summary = gem.description
|
11
|
+
gem.homepage = "https://github.com/hapyrus/fluent-plugin-redshift"
|
12
|
+
gem.has_rdoc = false
|
13
|
+
|
14
|
+
gem.files = `git ls-files`.split($/)
|
15
|
+
gem.executables = gem.files.grep(%r{^bin/}).map{ |f| File.basename(f) }
|
16
|
+
gem.test_files = gem.files.grep(%r{^(test|spec|features)/})
|
17
|
+
gem.require_paths = ["lib"]
|
18
|
+
|
19
|
+
gem.add_dependency "fluentd", "~> 0.10.0"
|
20
|
+
gem.add_dependency "aws-sdk", ">= 1.6.3"
|
21
|
+
gem.add_dependency "pg", "~> 0.14.0"
|
22
|
+
gem.add_development_dependency "rake"
|
23
|
+
gem.add_development_dependency "simplecov", ">= 0.5.4"
|
24
|
+
gem.add_development_dependency "flexmock", ">= 1.3.1"
|
25
|
+
end
|
@@ -0,0 +1,230 @@
|
|
1
|
+
module Fluent
|
2
|
+
|
3
|
+
|
4
|
+
class RedshiftOutput < BufferedOutput
|
5
|
+
Fluent::Plugin.register_output('redshift', self)
|
6
|
+
|
7
|
+
def initialize
|
8
|
+
super
|
9
|
+
require 'aws-sdk'
|
10
|
+
require 'zlib'
|
11
|
+
require 'time'
|
12
|
+
require 'tempfile'
|
13
|
+
require 'pg'
|
14
|
+
require 'json'
|
15
|
+
require 'csv'
|
16
|
+
end
|
17
|
+
|
18
|
+
config_param :record_log_tag, :string, :default => 'log'
|
19
|
+
# s3
|
20
|
+
config_param :aws_key_id, :string
|
21
|
+
config_param :aws_sec_key, :string
|
22
|
+
config_param :s3_bucket, :string
|
23
|
+
config_param :s3_endpoint, :string, :default => nil
|
24
|
+
config_param :path, :string, :default => ""
|
25
|
+
config_param :timestamp_key_format, :string, :default => 'year=%Y/month=%m/day=%d/hour=%H/%Y%m%d-%H%M'
|
26
|
+
config_param :utc, :bool, :default => false
|
27
|
+
# redshift
|
28
|
+
config_param :redshift_host, :string
|
29
|
+
config_param :redshift_port, :integer, :default => 5439
|
30
|
+
config_param :redshift_dbname, :string
|
31
|
+
config_param :redshift_user, :string
|
32
|
+
config_param :redshift_password, :string
|
33
|
+
config_param :redshift_tablename, :string
|
34
|
+
# file format
|
35
|
+
config_param :file_type, :string, :default => nil # json, tsv, csv
|
36
|
+
config_param :delimiter, :string, :default => nil
|
37
|
+
|
38
|
+
def configure(conf)
|
39
|
+
super
|
40
|
+
@path = "#{@path}/" if /.+[^\/]$/ =~ @path
|
41
|
+
@path = "" if @path == "/"
|
42
|
+
@utc = true if conf['utc']
|
43
|
+
@db_conf = {
|
44
|
+
host:@redshift_host,
|
45
|
+
port:@redshift_port,
|
46
|
+
dbname:@redshift_dbname,
|
47
|
+
user:@redshift_user,
|
48
|
+
password:@redshift_password
|
49
|
+
}
|
50
|
+
@delimiter = determine_delimiter(@file_type) if @delimiter.nil? or @delimiter.empty?
|
51
|
+
$log.debug "redshift file_type:#{@file_type} delimiter:'#{@delimiter}'"
|
52
|
+
@copy_sql_template = "copy #{@redshift_tablename} from '%s' CREDENTIALS 'aws_access_key_id=#{@aws_key_id};aws_secret_access_key=%s' delimiter '#{@delimiter}' REMOVEQUOTES GZIP;"
|
53
|
+
end
|
54
|
+
|
55
|
+
def start
|
56
|
+
super
|
57
|
+
# init s3 conf
|
58
|
+
options = {
|
59
|
+
:access_key_id => @aws_key_id,
|
60
|
+
:secret_access_key => @aws_sec_key
|
61
|
+
}
|
62
|
+
options[:s3_endpoint] = @s3_endpoint if @s3_endpoint
|
63
|
+
@s3 = AWS::S3.new(options)
|
64
|
+
@bucket = @s3.buckets[@s3_bucket]
|
65
|
+
end
|
66
|
+
|
67
|
+
def format(tag, time, record)
|
68
|
+
(json?) ? record.to_msgpack : "#{record[@record_log_tag]}\n"
|
69
|
+
end
|
70
|
+
|
71
|
+
def write(chunk)
|
72
|
+
# create a gz file
|
73
|
+
tmp = Tempfile.new("s3-")
|
74
|
+
tmp = (json?) ? create_gz_file_from_json(tmp, chunk, @delimiter)
|
75
|
+
: create_gz_file_from_msgpack(tmp, chunk)
|
76
|
+
|
77
|
+
# no data -> skip
|
78
|
+
unless tmp
|
79
|
+
$log.debug "received no valid data. "
|
80
|
+
return
|
81
|
+
end
|
82
|
+
|
83
|
+
# create a file path with time format
|
84
|
+
s3path = create_s3path(@bucket, @path)
|
85
|
+
|
86
|
+
# upload gz to s3
|
87
|
+
@bucket.objects[s3path].write(Pathname.new(tmp.path),
|
88
|
+
:acl => :bucket_owner_full_control)
|
89
|
+
# copy gz on s3 to redshift
|
90
|
+
s3_uri = "s3://#{@s3_bucket}/#{s3path}"
|
91
|
+
sql = @copy_sql_template % [s3_uri, @aws_sec_key]
|
92
|
+
$log.debug "start copying. s3_uri=#{s3_uri}"
|
93
|
+
conn = nil
|
94
|
+
begin
|
95
|
+
conn = PG.connect(@db_conf)
|
96
|
+
conn.exec(sql)
|
97
|
+
$log.info "completed copying to redshift. s3_uri=#{s3_uri}"
|
98
|
+
rescue PG::Error => e
|
99
|
+
$log.error "failed to copy data into redshift. sql=#{s3_uri}", :error=>e.to_s
|
100
|
+
raise e if e.result.nil? # retry if connection errors
|
101
|
+
ensure
|
102
|
+
conn.close rescue nil if conn
|
103
|
+
end
|
104
|
+
end
|
105
|
+
|
106
|
+
private
|
107
|
+
def json?
|
108
|
+
@file_type == 'json'
|
109
|
+
end
|
110
|
+
|
111
|
+
def create_gz_file_from_msgpack(dst_file, chunk)
|
112
|
+
gzw = nil
|
113
|
+
begin
|
114
|
+
gzw = Zlib::GzipWriter.new(dst_file)
|
115
|
+
chunk.write_to(gzw)
|
116
|
+
ensure
|
117
|
+
gzw.close rescue nil if gzw
|
118
|
+
end
|
119
|
+
dst_file
|
120
|
+
end
|
121
|
+
|
122
|
+
def create_gz_file_from_json(dst_file, chunk, delimiter)
|
123
|
+
# fetch the table definition from redshift
|
124
|
+
redshift_table_columns = fetch_table_columns
|
125
|
+
if redshift_table_columns == nil
|
126
|
+
raise "failed to fetch the redshift table definition."
|
127
|
+
elsif redshift_table_columns.empty?
|
128
|
+
$log.warn "no table on redshift. table_name=#{@redshift_tablename}"
|
129
|
+
return nil
|
130
|
+
end
|
131
|
+
|
132
|
+
# convert json to tsv format text
|
133
|
+
table_texts = ""
|
134
|
+
chunk.msgpack_each do |record|
|
135
|
+
begin
|
136
|
+
table_texts << json_to_table_text(redshift_table_columns, record[@record_log_tag], delimiter)
|
137
|
+
rescue => e
|
138
|
+
$log.error "failed to create table text from json. text=(#{record[@record_log_tag]})", :error=>$!.to_s
|
139
|
+
$log.error_backtrace
|
140
|
+
end
|
141
|
+
end
|
142
|
+
return nil if table_texts.empty?
|
143
|
+
|
144
|
+
# create gz
|
145
|
+
gzw = nil
|
146
|
+
begin
|
147
|
+
gzw = Zlib::GzipWriter.new(dst_file)
|
148
|
+
gzw.write(table_texts)
|
149
|
+
ensure
|
150
|
+
gzw.close rescue nil if gzw
|
151
|
+
end
|
152
|
+
dst_file
|
153
|
+
end
|
154
|
+
|
155
|
+
def determine_delimiter(file_type)
|
156
|
+
case file_type
|
157
|
+
when 'json', 'tsv'
|
158
|
+
"\t"
|
159
|
+
when "csv"
|
160
|
+
','
|
161
|
+
else
|
162
|
+
raise Fluent::ConfigError, "Invalid file_type:#{file_type}."
|
163
|
+
end
|
164
|
+
end
|
165
|
+
|
166
|
+
def fetch_table_columns
|
167
|
+
fetch_columns_sql = "select column_name from INFORMATION_SCHEMA.COLUMNS where table_name = '#{@redshift_tablename}' order by ordinal_position;"
|
168
|
+
conn = PG.connect(@db_conf)
|
169
|
+
begin
|
170
|
+
columns = nil
|
171
|
+
conn.exec(fetch_columns_sql) do |result|
|
172
|
+
columns = result.collect{|row| row['column_name']}
|
173
|
+
end
|
174
|
+
columns
|
175
|
+
ensure
|
176
|
+
conn.close rescue nil
|
177
|
+
end
|
178
|
+
end
|
179
|
+
|
180
|
+
def json_to_table_text(redshift_table_columns, json_text, delimiter)
|
181
|
+
return "" if json_text.nil? or json_text.empty?
|
182
|
+
|
183
|
+
# parse json text
|
184
|
+
json_obj = nil
|
185
|
+
begin
|
186
|
+
json_obj = JSON.parse(json_text)
|
187
|
+
rescue => e
|
188
|
+
$log.warn "failed to parse json. ", :error=>e.to_s
|
189
|
+
return ""
|
190
|
+
end
|
191
|
+
return "" unless json_obj
|
192
|
+
|
193
|
+
# extract values from json
|
194
|
+
val_list = redshift_table_columns.collect do |cn|
|
195
|
+
val = json_obj[cn]
|
196
|
+
val = nil unless val and not val.to_s.empty?
|
197
|
+
val = JSON.generate(val) if val.kind_of?(Hash) or val.kind_of?(Array)
|
198
|
+
val.to_s unless val.nil?
|
199
|
+
end
|
200
|
+
if val_list.all?{|v| v.nil? or v.empty?}
|
201
|
+
$log.warn "no data match for table columns on redshift. json_text=#{json_text} table_columns=#{redshift_table_columns}"
|
202
|
+
return ""
|
203
|
+
end
|
204
|
+
|
205
|
+
# generate tsv text
|
206
|
+
begin
|
207
|
+
CSV.generate(:col_sep=>delimiter, :quote_char => '"') do |row|
|
208
|
+
row << val_list # inlude new line
|
209
|
+
end
|
210
|
+
rescue => e
|
211
|
+
$log.debug "failed to generate csv val_list:#{val_list} delimiter:(#{delimiter})"
|
212
|
+
raise e
|
213
|
+
end
|
214
|
+
end
|
215
|
+
|
216
|
+
def create_s3path(bucket, path)
|
217
|
+
timestamp_key = (@utc) ? Time.now.utc.strftime(@timestamp_key_format) : Time.now.strftime(@timestamp_key_format)
|
218
|
+
i = 0
|
219
|
+
begin
|
220
|
+
suffix = "_#{'%02d' % i}"
|
221
|
+
s3path = "#{path}#{timestamp_key}#{suffix}.gz"
|
222
|
+
i += 1
|
223
|
+
end while bucket.objects[s3path].exists?
|
224
|
+
s3path
|
225
|
+
end
|
226
|
+
|
227
|
+
end
|
228
|
+
|
229
|
+
|
230
|
+
end
|
@@ -0,0 +1,395 @@
|
|
1
|
+
require 'test_helper'
|
2
|
+
|
3
|
+
require 'fluent/test'
|
4
|
+
require 'fluent/plugin/out_redshift'
|
5
|
+
require 'flexmock/test_unit'
|
6
|
+
require 'zlib'
|
7
|
+
|
8
|
+
|
9
|
+
class RedshiftOutputTest < Test::Unit::TestCase
|
10
|
+
def setup
|
11
|
+
require 'aws-sdk'
|
12
|
+
require 'pg'
|
13
|
+
require 'csv'
|
14
|
+
Fluent::Test.setup
|
15
|
+
end
|
16
|
+
|
17
|
+
CONFIG_BASE= %[
|
18
|
+
aws_key_id test_key_id
|
19
|
+
aws_sec_key test_sec_key
|
20
|
+
s3_bucket test_bucket
|
21
|
+
path log
|
22
|
+
redshift_host test_host
|
23
|
+
redshift_dbname test_db
|
24
|
+
redshift_user test_user
|
25
|
+
redshift_password test_password
|
26
|
+
redshift_tablename test_table
|
27
|
+
buffer_type memory
|
28
|
+
utc
|
29
|
+
]
|
30
|
+
CONFIG_CSV= %[
|
31
|
+
#{CONFIG_BASE}
|
32
|
+
file_type csv
|
33
|
+
]
|
34
|
+
CONFIG_TSV= %[
|
35
|
+
#{CONFIG_BASE}
|
36
|
+
file_type tsv
|
37
|
+
]
|
38
|
+
CONFIG_JSON = %[
|
39
|
+
#{CONFIG_BASE}
|
40
|
+
file_type json
|
41
|
+
]
|
42
|
+
CONFIG_PIPE_DELIMITER= %[
|
43
|
+
#{CONFIG_BASE}
|
44
|
+
delimiter |
|
45
|
+
]
|
46
|
+
CONFIG_PIPE_DELIMITER_WITH_NAME= %[
|
47
|
+
#{CONFIG_BASE}
|
48
|
+
file_type pipe
|
49
|
+
delimiter |
|
50
|
+
]
|
51
|
+
CONFIG=CONFIG_CSV
|
52
|
+
|
53
|
+
RECORD_CSV_A = {"log" => %[val_a,val_b,val_c,val_d]}
|
54
|
+
RECORD_CSV_B = {"log" => %[val_e,val_f,val_g,val_h]}
|
55
|
+
RECORD_TSV_A = {"log" => %[val_a\tval_b\tval_c\tval_d]}
|
56
|
+
RECORD_TSV_B = {"log" => %[val_e\tval_f\tval_g\tval_h]}
|
57
|
+
RECORD_JSON_A = {"log" => %[{"key_a" : "val_a", "key_b" : "val_b"}]}
|
58
|
+
RECORD_JSON_B = {"log" => %[{"key_c" : "val_c", "key_d" : "val_d"}]}
|
59
|
+
DEFAULT_TIME = Time.parse("2013-03-06 12:15:02 UTC").to_i
|
60
|
+
|
61
|
+
def create_driver(conf = CONFIG, tag='test.input')
|
62
|
+
Fluent::Test::BufferedOutputTestDriver.new(Fluent::RedshiftOutput, tag).configure(conf)
|
63
|
+
end
|
64
|
+
|
65
|
+
def create_driver_no_write(conf = CONFIG, tag='test.input')
|
66
|
+
Fluent::Test::BufferedOutputTestDriver.new(Fluent::RedshiftOutput, tag) do
|
67
|
+
def write(chunk)
|
68
|
+
chunk.read
|
69
|
+
end
|
70
|
+
end.configure(conf)
|
71
|
+
end
|
72
|
+
|
73
|
+
def test_configure
|
74
|
+
assert_raise(Fluent::ConfigError) {
|
75
|
+
d = create_driver('')
|
76
|
+
}
|
77
|
+
assert_raise(Fluent::ConfigError) {
|
78
|
+
d = create_driver(CONFIG_BASE)
|
79
|
+
}
|
80
|
+
d = create_driver(CONFIG_CSV)
|
81
|
+
assert_equal "test_key_id", d.instance.aws_key_id
|
82
|
+
assert_equal "test_sec_key", d.instance.aws_sec_key
|
83
|
+
assert_equal "test_bucket", d.instance.s3_bucket
|
84
|
+
assert_equal "log/", d.instance.path
|
85
|
+
assert_equal "test_host", d.instance.redshift_host
|
86
|
+
assert_equal 5439, d.instance.redshift_port
|
87
|
+
assert_equal "test_db", d.instance.redshift_dbname
|
88
|
+
assert_equal "test_user", d.instance.redshift_user
|
89
|
+
assert_equal "test_password", d.instance.redshift_password
|
90
|
+
assert_equal "test_table", d.instance.redshift_tablename
|
91
|
+
assert_equal "csv", d.instance.file_type
|
92
|
+
assert_equal ",", d.instance.delimiter
|
93
|
+
assert_equal true, d.instance.utc
|
94
|
+
end
|
95
|
+
def test_configure_localtime
|
96
|
+
d = create_driver(CONFIG_CSV.gsub(/ *utc */, ''))
|
97
|
+
assert_equal false, d.instance.utc
|
98
|
+
end
|
99
|
+
def test_configure_no_path
|
100
|
+
d = create_driver(CONFIG_CSV.gsub(/ *path *.+$/, ''))
|
101
|
+
assert_equal "", d.instance.path
|
102
|
+
end
|
103
|
+
def test_configure_root_path
|
104
|
+
d = create_driver(CONFIG_CSV.gsub(/ *path *.+$/, 'path /'))
|
105
|
+
assert_equal "", d.instance.path
|
106
|
+
end
|
107
|
+
def test_configure_path_with_slash
|
108
|
+
d = create_driver(CONFIG_CSV.gsub(/ *path *.+$/, 'path log/'))
|
109
|
+
assert_equal "log/", d.instance.path
|
110
|
+
end
|
111
|
+
def test_configure_tsv
|
112
|
+
d1 = create_driver(CONFIG_TSV)
|
113
|
+
assert_equal "tsv", d1.instance.file_type
|
114
|
+
assert_equal "\t", d1.instance.delimiter
|
115
|
+
end
|
116
|
+
def test_configure_json
|
117
|
+
d2 = create_driver(CONFIG_JSON)
|
118
|
+
assert_equal "json", d2.instance.file_type
|
119
|
+
assert_equal "\t", d2.instance.delimiter
|
120
|
+
end
|
121
|
+
def test_configure_original_file_type
|
122
|
+
d3 = create_driver(CONFIG_PIPE_DELIMITER)
|
123
|
+
assert_equal nil, d3.instance.file_type
|
124
|
+
assert_equal "|", d3.instance.delimiter
|
125
|
+
|
126
|
+
d4 = create_driver(CONFIG_PIPE_DELIMITER_WITH_NAME)
|
127
|
+
assert_equal "pipe", d4.instance.file_type
|
128
|
+
assert_equal "|", d4.instance.delimiter
|
129
|
+
end
|
130
|
+
|
131
|
+
def emit_csv(d)
|
132
|
+
d.emit(RECORD_CSV_A, DEFAULT_TIME)
|
133
|
+
d.emit(RECORD_CSV_B, DEFAULT_TIME)
|
134
|
+
end
|
135
|
+
def emit_tsv(d)
|
136
|
+
d.emit(RECORD_TSV_A, DEFAULT_TIME)
|
137
|
+
d.emit(RECORD_TSV_B, DEFAULT_TIME)
|
138
|
+
end
|
139
|
+
def emit_json(d)
|
140
|
+
d.emit(RECORD_JSON_A, DEFAULT_TIME)
|
141
|
+
d.emit(RECORD_JSON_B, DEFAULT_TIME)
|
142
|
+
end
|
143
|
+
|
144
|
+
def test_format_csv
|
145
|
+
d_csv = create_driver_no_write(CONFIG_CSV)
|
146
|
+
emit_csv(d_csv)
|
147
|
+
d_csv.expect_format RECORD_CSV_A['log'] + "\n"
|
148
|
+
d_csv.expect_format RECORD_CSV_B['log'] + "\n"
|
149
|
+
d_csv.run
|
150
|
+
end
|
151
|
+
def test_format_tsv
|
152
|
+
d_tsv = create_driver_no_write(CONFIG_TSV)
|
153
|
+
emit_tsv(d_tsv)
|
154
|
+
d_tsv.expect_format RECORD_TSV_A['log'] + "\n"
|
155
|
+
d_tsv.expect_format RECORD_TSV_B['log'] + "\n"
|
156
|
+
d_tsv.run
|
157
|
+
end
|
158
|
+
def test_format_json
|
159
|
+
d_json = create_driver_no_write(CONFIG_JSON)
|
160
|
+
emit_json(d_json)
|
161
|
+
d_json.expect_format RECORD_JSON_A.to_msgpack
|
162
|
+
d_json.expect_format RECORD_JSON_B.to_msgpack
|
163
|
+
d_json.run
|
164
|
+
end
|
165
|
+
|
166
|
+
class PGConnectionMock
|
167
|
+
def initialize(return_keys=['key_a', 'key_b', 'key_c', 'key_d', 'key_e', 'key_f', 'key_g', 'key_h'])
|
168
|
+
@return_keys = return_keys
|
169
|
+
end
|
170
|
+
def exec(sql, &block)
|
171
|
+
if block_given? and /^select column_name from/ =~ sql
|
172
|
+
yield @return_keys.collect{|key| {'column_name' => key}}
|
173
|
+
end
|
174
|
+
end
|
175
|
+
def close
|
176
|
+
end
|
177
|
+
end
|
178
|
+
|
179
|
+
def setup_pg_mock
|
180
|
+
# create mock of PG
|
181
|
+
def PG.connect(dbinfo)
|
182
|
+
return PGConnectionMock.new
|
183
|
+
end
|
184
|
+
end
|
185
|
+
|
186
|
+
def setup_s3_mock(expected_data)
|
187
|
+
current_time = Time.now
|
188
|
+
|
189
|
+
# create mock of s3 object
|
190
|
+
s3obj = flexmock(AWS::S3::S3Object)
|
191
|
+
s3obj.should_receive(:exists?).with_any_args.and_return { false }
|
192
|
+
s3obj.should_receive(:write).with(
|
193
|
+
# pathname
|
194
|
+
on { |pathname|
|
195
|
+
data = nil
|
196
|
+
pathname.open { |f|
|
197
|
+
gz = Zlib::GzipReader.new(f)
|
198
|
+
data = gz.read
|
199
|
+
gz.close
|
200
|
+
}
|
201
|
+
assert_equal expected_data, data
|
202
|
+
},
|
203
|
+
:acl => :bucket_owner_full_control
|
204
|
+
).and_return { true }
|
205
|
+
|
206
|
+
# create mock of s3 object collection
|
207
|
+
s3obj_col = flexmock(AWS::S3::ObjectCollection)
|
208
|
+
s3obj_col.should_receive(:[]).with(
|
209
|
+
on { |key|
|
210
|
+
expected_key = current_time.utc.strftime("log/year=%Y/month=%m/day=%d/hour=%H/%Y%m%d-%H%M_00.gz")
|
211
|
+
key == expected_key
|
212
|
+
}).
|
213
|
+
and_return {
|
214
|
+
s3obj
|
215
|
+
}
|
216
|
+
|
217
|
+
# create mock of s3 bucket
|
218
|
+
flexmock(AWS::S3::Bucket).new_instances do |bucket|
|
219
|
+
bucket.should_receive(:objects).with_any_args.
|
220
|
+
and_return {
|
221
|
+
s3obj_col
|
222
|
+
}
|
223
|
+
end
|
224
|
+
end
|
225
|
+
|
226
|
+
def setup_mocks(expected_data)
|
227
|
+
setup_pg_mock
|
228
|
+
setup_s3_mock(expected_data) end
|
229
|
+
|
230
|
+
def test_write_with_csv
|
231
|
+
setup_mocks(%[val_a,val_b,val_c,val_d\nval_e,val_f,val_g,val_h\n])
|
232
|
+
d_csv = create_driver
|
233
|
+
emit_csv(d_csv)
|
234
|
+
d_csv.run
|
235
|
+
end
|
236
|
+
|
237
|
+
def test_write_with_json
|
238
|
+
setup_mocks(%[val_a\tval_b\t\t\t\t\t\t\n\t\tval_c\tval_d\t\t\t\t\n])
|
239
|
+
d_json = create_driver(CONFIG_JSON)
|
240
|
+
emit_json(d_json)
|
241
|
+
d_json.run
|
242
|
+
end
|
243
|
+
|
244
|
+
def test_write_with_json_hash_value
|
245
|
+
setup_mocks("val_a\t\"{\"\"foo\"\":\"\"var\"\"}\"\t\t\t\t\t\t\n\t\tval_c\tval_d\t\t\t\t\n")
|
246
|
+
d_json = create_driver(CONFIG_JSON)
|
247
|
+
d_json.emit({"log" => %[{"key_a" : "val_a", "key_b" : {"foo" : "var"}}]} , DEFAULT_TIME)
|
248
|
+
d_json.emit(RECORD_JSON_B, DEFAULT_TIME)
|
249
|
+
d_json.run
|
250
|
+
end
|
251
|
+
|
252
|
+
def test_write_with_json_array_value
|
253
|
+
setup_mocks("val_a\t\"[\"\"foo\"\",\"\"var\"\"]\"\t\t\t\t\t\t\n\t\tval_c\tval_d\t\t\t\t\n")
|
254
|
+
d_json = create_driver(CONFIG_JSON)
|
255
|
+
d_json.emit({"log" => %[{"key_a" : "val_a", "key_b" : ["foo", "var"]}]} , DEFAULT_TIME)
|
256
|
+
d_json.emit(RECORD_JSON_B, DEFAULT_TIME)
|
257
|
+
d_json.run
|
258
|
+
end
|
259
|
+
|
260
|
+
def test_write_with_json_no_data
|
261
|
+
setup_mocks("")
|
262
|
+
d_json = create_driver(CONFIG_JSON)
|
263
|
+
d_json.emit("", DEFAULT_TIME)
|
264
|
+
d_json.emit("", DEFAULT_TIME)
|
265
|
+
d_json.run
|
266
|
+
end
|
267
|
+
|
268
|
+
def test_write_with_json_invalid_one_line
|
269
|
+
setup_mocks(%[\t\tval_c\tval_d\t\t\t\t\n])
|
270
|
+
d_json = create_driver(CONFIG_JSON)
|
271
|
+
d_json.emit({"log" => %[}}]}, DEFAULT_TIME)
|
272
|
+
d_json.emit(RECORD_JSON_B, DEFAULT_TIME)
|
273
|
+
d_json.run
|
274
|
+
end
|
275
|
+
|
276
|
+
def test_write_with_json_no_available_data
|
277
|
+
setup_mocks(%[val_a\tval_b\t\t\t\t\t\t\n])
|
278
|
+
d_json = create_driver(CONFIG_JSON)
|
279
|
+
d_json.emit(RECORD_JSON_A, DEFAULT_TIME)
|
280
|
+
d_json.emit({"log" => %[{"key_o" : "val_o", "key_p" : "val_p"}]}, DEFAULT_TIME)
|
281
|
+
d_json.run
|
282
|
+
end
|
283
|
+
|
284
|
+
def test_write_redshift_connection_error
|
285
|
+
def PG.connect(dbinfo)
|
286
|
+
return Class.new do
|
287
|
+
def initialize(return_keys=[]); end
|
288
|
+
def exec(sql)
|
289
|
+
raise PG::Error, "redshift connection error"
|
290
|
+
end
|
291
|
+
def close; end
|
292
|
+
end.new
|
293
|
+
end
|
294
|
+
setup_s3_mock(%[val_a,val_b,val_c,val_d\nval_e,val_f,val_g,val_h\n])
|
295
|
+
|
296
|
+
d_csv = create_driver
|
297
|
+
emit_csv(d_csv)
|
298
|
+
assert_raise(PG::Error) {
|
299
|
+
d_csv.run
|
300
|
+
}
|
301
|
+
end
|
302
|
+
|
303
|
+
def test_write_redshift_logic_error
|
304
|
+
PG::Error.module_eval { attr_accessor :result}
|
305
|
+
def PG.connect(dbinfo)
|
306
|
+
return Class.new do
|
307
|
+
def initialize(return_keys=[]); end
|
308
|
+
def exec(sql)
|
309
|
+
error = PG::Error.new("redshift logic error")
|
310
|
+
error.result = "logic error"
|
311
|
+
raise error
|
312
|
+
end
|
313
|
+
def close; end
|
314
|
+
end.new
|
315
|
+
end
|
316
|
+
setup_s3_mock(%[val_a,val_b,val_c,val_d\nval_e,val_f,val_g,val_h\n])
|
317
|
+
|
318
|
+
d_csv = create_driver
|
319
|
+
emit_csv(d_csv)
|
320
|
+
assert_nothing_raised {
|
321
|
+
d_csv.run
|
322
|
+
}
|
323
|
+
end
|
324
|
+
|
325
|
+
def test_write_with_json_redshift_connection_error
|
326
|
+
def PG.connect(dbinfo)
|
327
|
+
return Class.new do
|
328
|
+
def initialize(return_keys=[]); end
|
329
|
+
def exec(sql, &block)
|
330
|
+
error = PG::Error.new("redshift connection error")
|
331
|
+
raise error
|
332
|
+
end
|
333
|
+
def close; end
|
334
|
+
end.new
|
335
|
+
end
|
336
|
+
setup_s3_mock(%[val_a,val_b,val_c,val_d\nval_e,val_f,val_g,val_h\n])
|
337
|
+
|
338
|
+
d_json = create_driver(CONFIG_JSON)
|
339
|
+
emit_json(d_json)
|
340
|
+
assert_raise(PG::Error) {
|
341
|
+
d_json.run
|
342
|
+
}
|
343
|
+
end
|
344
|
+
|
345
|
+
def test_write_with_json_no_table_on_redshift
|
346
|
+
def PG.connect(dbinfo)
|
347
|
+
return Class.new do
|
348
|
+
def initialize(return_keys=[]); end
|
349
|
+
def exec(sql, &block)
|
350
|
+
yield [] if block_given?
|
351
|
+
end
|
352
|
+
def close; end
|
353
|
+
end.new
|
354
|
+
end
|
355
|
+
setup_s3_mock(%[val_a,val_b,val_c,val_d\nval_e,val_f,val_g,val_h\n])
|
356
|
+
|
357
|
+
d_json = create_driver(CONFIG_JSON)
|
358
|
+
emit_json(d_json)
|
359
|
+
assert_nothing_raised {
|
360
|
+
d_json.run
|
361
|
+
}
|
362
|
+
end
|
363
|
+
|
364
|
+
def test_write_with_json_failed_to_get_columns
|
365
|
+
def PG.connect(dbinfo)
|
366
|
+
return Class.new do
|
367
|
+
def initialize(return_keys=[]); end
|
368
|
+
def exec(sql, &block)
|
369
|
+
end
|
370
|
+
def close; end
|
371
|
+
end.new
|
372
|
+
end
|
373
|
+
setup_s3_mock("")
|
374
|
+
|
375
|
+
d_json = create_driver(CONFIG_JSON)
|
376
|
+
emit_json(d_json)
|
377
|
+
assert_raise(RuntimeError, "failed to fetch the redshift table definition.") {
|
378
|
+
d_json.run
|
379
|
+
}
|
380
|
+
end
|
381
|
+
|
382
|
+
def test_write_with_json_failed_to_generate_tsv
|
383
|
+
flexmock(CSV).should_receive(:generate).with_any_args.
|
384
|
+
and_return {
|
385
|
+
raise "failed to generate tsv."
|
386
|
+
}
|
387
|
+
setup_s3_mock("")
|
388
|
+
|
389
|
+
d_json = create_driver(CONFIG_JSON)
|
390
|
+
emit_json(d_json)
|
391
|
+
assert_nothing_raised {
|
392
|
+
d_json.run
|
393
|
+
}
|
394
|
+
end
|
395
|
+
end
|
data/test/test_helper.rb
ADDED
metadata
ADDED
@@ -0,0 +1,152 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: fluent-plugin-redshift
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.0.1
|
5
|
+
prerelease:
|
6
|
+
platform: ruby
|
7
|
+
authors:
|
8
|
+
- Masashi Miyazaki
|
9
|
+
autorequire:
|
10
|
+
bindir: bin
|
11
|
+
cert_chain: []
|
12
|
+
date: 2013-03-07 00:00:00.000000000 Z
|
13
|
+
dependencies:
|
14
|
+
- !ruby/object:Gem::Dependency
|
15
|
+
name: fluentd
|
16
|
+
requirement: !ruby/object:Gem::Requirement
|
17
|
+
none: false
|
18
|
+
requirements:
|
19
|
+
- - ~>
|
20
|
+
- !ruby/object:Gem::Version
|
21
|
+
version: 0.10.0
|
22
|
+
type: :runtime
|
23
|
+
prerelease: false
|
24
|
+
version_requirements: !ruby/object:Gem::Requirement
|
25
|
+
none: false
|
26
|
+
requirements:
|
27
|
+
- - ~>
|
28
|
+
- !ruby/object:Gem::Version
|
29
|
+
version: 0.10.0
|
30
|
+
- !ruby/object:Gem::Dependency
|
31
|
+
name: aws-sdk
|
32
|
+
requirement: !ruby/object:Gem::Requirement
|
33
|
+
none: false
|
34
|
+
requirements:
|
35
|
+
- - ! '>='
|
36
|
+
- !ruby/object:Gem::Version
|
37
|
+
version: 1.6.3
|
38
|
+
type: :runtime
|
39
|
+
prerelease: false
|
40
|
+
version_requirements: !ruby/object:Gem::Requirement
|
41
|
+
none: false
|
42
|
+
requirements:
|
43
|
+
- - ! '>='
|
44
|
+
- !ruby/object:Gem::Version
|
45
|
+
version: 1.6.3
|
46
|
+
- !ruby/object:Gem::Dependency
|
47
|
+
name: pg
|
48
|
+
requirement: !ruby/object:Gem::Requirement
|
49
|
+
none: false
|
50
|
+
requirements:
|
51
|
+
- - ~>
|
52
|
+
- !ruby/object:Gem::Version
|
53
|
+
version: 0.14.0
|
54
|
+
type: :runtime
|
55
|
+
prerelease: false
|
56
|
+
version_requirements: !ruby/object:Gem::Requirement
|
57
|
+
none: false
|
58
|
+
requirements:
|
59
|
+
- - ~>
|
60
|
+
- !ruby/object:Gem::Version
|
61
|
+
version: 0.14.0
|
62
|
+
- !ruby/object:Gem::Dependency
|
63
|
+
name: rake
|
64
|
+
requirement: !ruby/object:Gem::Requirement
|
65
|
+
none: false
|
66
|
+
requirements:
|
67
|
+
- - ! '>='
|
68
|
+
- !ruby/object:Gem::Version
|
69
|
+
version: '0'
|
70
|
+
type: :development
|
71
|
+
prerelease: false
|
72
|
+
version_requirements: !ruby/object:Gem::Requirement
|
73
|
+
none: false
|
74
|
+
requirements:
|
75
|
+
- - ! '>='
|
76
|
+
- !ruby/object:Gem::Version
|
77
|
+
version: '0'
|
78
|
+
- !ruby/object:Gem::Dependency
|
79
|
+
name: simplecov
|
80
|
+
requirement: !ruby/object:Gem::Requirement
|
81
|
+
none: false
|
82
|
+
requirements:
|
83
|
+
- - ! '>='
|
84
|
+
- !ruby/object:Gem::Version
|
85
|
+
version: 0.5.4
|
86
|
+
type: :development
|
87
|
+
prerelease: false
|
88
|
+
version_requirements: !ruby/object:Gem::Requirement
|
89
|
+
none: false
|
90
|
+
requirements:
|
91
|
+
- - ! '>='
|
92
|
+
- !ruby/object:Gem::Version
|
93
|
+
version: 0.5.4
|
94
|
+
- !ruby/object:Gem::Dependency
|
95
|
+
name: flexmock
|
96
|
+
requirement: !ruby/object:Gem::Requirement
|
97
|
+
none: false
|
98
|
+
requirements:
|
99
|
+
- - ! '>='
|
100
|
+
- !ruby/object:Gem::Version
|
101
|
+
version: 1.3.1
|
102
|
+
type: :development
|
103
|
+
prerelease: false
|
104
|
+
version_requirements: !ruby/object:Gem::Requirement
|
105
|
+
none: false
|
106
|
+
requirements:
|
107
|
+
- - ! '>='
|
108
|
+
- !ruby/object:Gem::Version
|
109
|
+
version: 1.3.1
|
110
|
+
description: Amazon Redshift output plugin for Fluentd
|
111
|
+
email:
|
112
|
+
- mmasashi@gmail.com
|
113
|
+
executables: []
|
114
|
+
extensions: []
|
115
|
+
extra_rdoc_files: []
|
116
|
+
files:
|
117
|
+
- .gitignore
|
118
|
+
- Gemfile
|
119
|
+
- README.md
|
120
|
+
- Rakefile
|
121
|
+
- VERSION
|
122
|
+
- fluent-plugin-redshift.gemspec
|
123
|
+
- lib/fluent/plugin/out_redshift.rb
|
124
|
+
- test/plugin/test_out_redshift.rb
|
125
|
+
- test/test_helper.rb
|
126
|
+
homepage: https://github.com/hapyrus/fluent-plugin-redshift
|
127
|
+
licenses: []
|
128
|
+
post_install_message:
|
129
|
+
rdoc_options: []
|
130
|
+
require_paths:
|
131
|
+
- lib
|
132
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
133
|
+
none: false
|
134
|
+
requirements:
|
135
|
+
- - ! '>='
|
136
|
+
- !ruby/object:Gem::Version
|
137
|
+
version: '0'
|
138
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
139
|
+
none: false
|
140
|
+
requirements:
|
141
|
+
- - ! '>='
|
142
|
+
- !ruby/object:Gem::Version
|
143
|
+
version: '0'
|
144
|
+
requirements: []
|
145
|
+
rubyforge_project:
|
146
|
+
rubygems_version: 1.8.23
|
147
|
+
signing_key:
|
148
|
+
specification_version: 3
|
149
|
+
summary: Amazon Redshift output plugin for Fluentd
|
150
|
+
test_files:
|
151
|
+
- test/plugin/test_out_redshift.rb
|
152
|
+
- test/test_helper.rb
|