fluent-plugin-webhdfs 0.7.1 → 1.0.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/.travis.yml +1 -10
- data/Appraisals +0 -2
- data/README.md +55 -45
- data/fluent-plugin-webhdfs.gemspec +4 -4
- data/lib/fluent/plugin/out_webhdfs.rb +193 -100
- data/lib/fluent/plugin/webhdfs_compressor_bzip2.rb +3 -7
- data/lib/fluent/plugin/webhdfs_compressor_gzip.rb +3 -3
- data/lib/fluent/plugin/webhdfs_compressor_lzo_command.rb +3 -3
- data/lib/fluent/plugin/webhdfs_compressor_snappy.rb +2 -2
- data/lib/fluent/plugin/webhdfs_compressor_text.rb +2 -2
- data/test/helper.rb +5 -0
- data/test/plugin/test_compressor.rb +3 -3
- data/test/plugin/test_out_webhdfs.rb +179 -105
- metadata +18 -24
@@ -1,14 +1,10 @@
|
|
1
|
-
module Fluent
|
2
|
-
class WebHDFSOutput <
|
1
|
+
module Fluent::Plugin
|
2
|
+
class WebHDFSOutput < Output
|
3
3
|
class Bzip2Compressor < Compressor
|
4
4
|
WebHDFSOutput.register_compressor('bzip2', self)
|
5
5
|
|
6
6
|
def initialize(options = {})
|
7
|
-
|
8
|
-
require "bzip2/ffi"
|
9
|
-
rescue LoadError
|
10
|
-
raise Fluent::ConfigError, "Install bzip2-ffi before use bzip2 compressor"
|
11
|
-
end
|
7
|
+
require "bzip2/ffi"
|
12
8
|
end
|
13
9
|
|
14
10
|
def ext
|
@@ -1,5 +1,5 @@
|
|
1
|
-
module Fluent
|
2
|
-
class WebHDFSOutput <
|
1
|
+
module Fluent::Plugin
|
2
|
+
class WebHDFSOutput < Output
|
3
3
|
class GzipCompressor < Compressor
|
4
4
|
WebHDFSOutput.register_compressor('gzip', self)
|
5
5
|
|
@@ -14,7 +14,7 @@ module Fluent
|
|
14
14
|
def compress(chunk, tmp)
|
15
15
|
w = Zlib::GzipWriter.new(tmp)
|
16
16
|
chunk.write_to(w)
|
17
|
-
w.
|
17
|
+
w.close
|
18
18
|
end
|
19
19
|
end
|
20
20
|
end
|
@@ -1,9 +1,9 @@
|
|
1
|
-
module Fluent
|
2
|
-
class WebHDFSOutput
|
1
|
+
module Fluent::Plugin
|
2
|
+
class WebHDFSOutput < Output
|
3
3
|
class LZOCommandCompressor < Compressor
|
4
4
|
WebHDFSOutput.register_compressor('lzo_command', self)
|
5
5
|
|
6
|
-
config_param :command_parameter, :string, :
|
6
|
+
config_param :command_parameter, :string, default: '-qf1'
|
7
7
|
|
8
8
|
def configure(conf)
|
9
9
|
super
|
data/test/helper.rb
CHANGED
@@ -8,10 +8,13 @@ rescue Bundler::BundlerError => e
|
|
8
8
|
exit e.status_code
|
9
9
|
end
|
10
10
|
require 'test/unit'
|
11
|
+
require 'test/unit/rr'
|
11
12
|
|
12
13
|
$LOAD_PATH.unshift(File.join(File.dirname(__FILE__), '..', 'lib'))
|
13
14
|
$LOAD_PATH.unshift(File.dirname(__FILE__))
|
14
15
|
require 'fluent/test'
|
16
|
+
require 'fluent/test/helpers'
|
17
|
+
require 'fluent/test/driver/output'
|
15
18
|
unless ENV.has_key?('VERBOSE')
|
16
19
|
nulllogger = Object.new
|
17
20
|
nulllogger.instance_eval {|obj|
|
@@ -22,6 +25,8 @@ unless ENV.has_key?('VERBOSE')
|
|
22
25
|
$log = nulllogger
|
23
26
|
end
|
24
27
|
|
28
|
+
include Fluent::Test::Helpers
|
29
|
+
|
25
30
|
require 'fluent/plugin/out_webhdfs'
|
26
31
|
|
27
32
|
class Test::Unit::TestCase
|
@@ -16,11 +16,11 @@ class CompressorTest < Test::Unit::TestCase
|
|
16
16
|
def setup
|
17
17
|
omit unless Object.const_defined?(:Snappy)
|
18
18
|
Fluent::Test.setup
|
19
|
-
@compressor = Fluent::WebHDFSOutput::SnappyCompressor.new
|
19
|
+
@compressor = Fluent::Plugin::WebHDFSOutput::SnappyCompressor.new
|
20
20
|
end
|
21
21
|
|
22
|
-
def create_driver(conf=CONFIG
|
23
|
-
Fluent::Test::
|
22
|
+
def create_driver(conf = CONFIG)
|
23
|
+
Fluent::Test::Driver::Output.new(Fluent::Plugin::WebHDFSOutput).configure(conf)
|
24
24
|
end
|
25
25
|
|
26
26
|
def test_ext
|
@@ -1,169 +1,243 @@
|
|
1
1
|
require 'helper'
|
2
2
|
|
3
3
|
class WebHDFSOutputTest < Test::Unit::TestCase
|
4
|
-
|
5
|
-
|
6
|
-
|
7
|
-
|
4
|
+
CONFIG_DEFAULT = config_element("match", "", {"host" => "namenode.local", "path" => "/hdfs/path/file.%Y%m%d.log"})
|
5
|
+
|
6
|
+
CONFIG_COMPAT = config_element(
|
7
|
+
"ROOT", "", {
|
8
|
+
"output_data_type" => "",
|
9
|
+
"host" => "namenode.local",
|
10
|
+
"path" => "/hdfs/path/file.%Y%m%d.log"
|
11
|
+
})
|
12
|
+
|
8
13
|
def setup
|
9
14
|
Fluent::Test.setup
|
10
15
|
end
|
11
16
|
|
12
|
-
def create_driver(conf
|
13
|
-
Fluent::Test::
|
17
|
+
def create_driver(conf)
|
18
|
+
Fluent::Test::Driver::Output.new(Fluent::Plugin::WebHDFSOutput).configure(conf)
|
14
19
|
end
|
15
20
|
|
16
|
-
|
17
|
-
|
18
|
-
d = create_driver
|
21
|
+
sub_test_case "default configuration" do
|
22
|
+
test 'configured with standard out_file format with specified hdfs info' do
|
23
|
+
d = create_driver(CONFIG_DEFAULT)
|
24
|
+
assert_true d.instance.instance_eval{ @using_formatter_config }
|
25
|
+
|
19
26
|
assert_equal 'namenode.local', d.instance.instance_eval{ @namenode_host }
|
20
27
|
assert_equal 50070, d.instance.instance_eval{ @namenode_port }
|
21
28
|
assert_equal '/hdfs/path/file.%Y%m%d.log', d.instance.path
|
22
|
-
assert_equal '%Y%m%d', d.instance.time_slice_format
|
23
29
|
assert_equal false, d.instance.httpfs
|
24
30
|
assert_nil d.instance.username
|
25
31
|
assert_equal false, d.instance.ignore_start_check_error
|
26
32
|
|
27
|
-
assert_equal
|
28
|
-
assert_equal true, d.instance.
|
29
|
-
|
33
|
+
assert_equal 'Fluent::Plugin::OutFileFormatter', d.instance.formatter.class.to_s
|
34
|
+
assert_equal true, d.instance.end_with_newline
|
35
|
+
|
36
|
+
# deprecated params
|
37
|
+
assert_nil d.instance.instance_eval{ @output_include_time }
|
38
|
+
assert_nil d.instance.instance_eval{ @output_include_tag }
|
30
39
|
assert_nil d.instance.remove_prefix
|
31
|
-
|
32
|
-
|
33
|
-
|
40
|
+
assert_nil d.instance.instance_eval{ @header_separator }
|
41
|
+
assert_nil d.instance.default_tag
|
42
|
+
end
|
43
|
+
end
|
44
|
+
|
45
|
+
sub_test_case "flat configuration" do
|
46
|
+
def test_default_for_traditional_config
|
47
|
+
d = create_driver(CONFIG_COMPAT)
|
48
|
+
assert_false d.instance.instance_eval{ @using_formatter_config }
|
49
|
+
|
50
|
+
assert_equal 'namenode.local', d.instance.instance_eval{ @namenode_host }
|
51
|
+
assert_equal 50070, d.instance.instance_eval{ @namenode_port }
|
52
|
+
assert_equal '/hdfs/path/file.%Y%m%d.log', d.instance.path
|
53
|
+
assert_equal false, d.instance.httpfs
|
54
|
+
assert_nil d.instance.username
|
55
|
+
assert_equal false, d.instance.ignore_start_check_error
|
56
|
+
|
57
|
+
assert_equal 'Fluent::Plugin::JSONFormatter', d.instance.formatter.class.to_s
|
58
|
+
assert_equal true, d.instance.end_with_newline
|
59
|
+
|
60
|
+
assert_equal true, d.instance.instance_eval{ @output_include_time }
|
61
|
+
assert_equal true, d.instance.instance_eval{ @output_include_tag }
|
62
|
+
assert_nil d.instance.instance_eval{ @remove_prefix }
|
63
|
+
assert_equal "\t", d.instance.instance_eval{ @header_separator }
|
64
|
+
assert_equal 'tag_missing', d.instance.instance_eval{ @default_tag }
|
34
65
|
end
|
35
66
|
|
36
67
|
def test_httpfs
|
37
|
-
|
38
|
-
|
39
|
-
|
40
|
-
|
41
|
-
|
42
|
-
|
68
|
+
conf = config_element(
|
69
|
+
"ROOT", "", {
|
70
|
+
"namenode" => "server.local:14000",
|
71
|
+
"path" => "/hdfs/path/file.%Y%m%d.%H%M.log",
|
72
|
+
"httpfs" => "yes",
|
73
|
+
"username" => "hdfs_user"
|
74
|
+
})
|
75
|
+
d = create_driver(conf)
|
76
|
+
|
43
77
|
assert_equal 'server.local', d.instance.instance_eval{ @namenode_host }
|
44
78
|
assert_equal 14000, d.instance.instance_eval{ @namenode_port }
|
45
79
|
assert_equal '/hdfs/path/file.%Y%m%d.%H%M.log', d.instance.path
|
46
|
-
assert_equal '%Y%m%d%H%M', d.instance.time_slice_format
|
47
80
|
assert_equal true, d.instance.httpfs
|
48
81
|
assert_equal 'hdfs_user', d.instance.username
|
49
82
|
end
|
50
83
|
|
51
84
|
def test_ssl
|
52
|
-
|
53
|
-
|
54
|
-
|
55
|
-
|
56
|
-
|
57
|
-
|
58
|
-
|
59
|
-
|
60
|
-
|
85
|
+
conf = config_element(
|
86
|
+
"ROOT", "", {
|
87
|
+
"namenode" => "server.local:14000",
|
88
|
+
"path" => "/hdfs/path/file.%Y%m%d.%H%M.log",
|
89
|
+
"ssl" => true,
|
90
|
+
"ssl_ca_file" => "/path/to/ca_file.pem",
|
91
|
+
"ssl_verify_mode" => "peer",
|
92
|
+
"kerberos" => true
|
93
|
+
})
|
94
|
+
d = create_driver(conf)
|
61
95
|
|
62
96
|
assert_equal 'server.local', d.instance.instance_eval{ @namenode_host }
|
63
97
|
assert_equal 14000, d.instance.instance_eval{ @namenode_port }
|
64
98
|
assert_equal '/hdfs/path/file.%Y%m%d.%H%M.log', d.instance.path
|
65
|
-
assert_equal '%Y%m%d%H%M', d.instance.time_slice_format
|
66
99
|
assert_equal true, d.instance.ssl
|
67
100
|
assert_equal '/path/to/ca_file.pem', d.instance.ssl_ca_file
|
68
101
|
assert_equal :peer, d.instance.ssl_verify_mode
|
69
102
|
assert_equal true, d.instance.kerberos
|
70
|
-
assert_equal '/path/to/kerberos.keytab', d.instance.kerberos_keytab
|
71
103
|
end
|
72
104
|
|
73
|
-
data(gzip: [
|
74
|
-
bzip2: [
|
75
|
-
snappy: [
|
76
|
-
lzo: [
|
105
|
+
data(gzip: [:gzip, Fluent::Plugin::WebHDFSOutput::GzipCompressor],
|
106
|
+
bzip2: [:bzip2, Fluent::Plugin::WebHDFSOutput::Bzip2Compressor],
|
107
|
+
snappy: [:snappy, Fluent::Plugin::WebHDFSOutput::SnappyCompressor],
|
108
|
+
lzo: [:lzo_command, Fluent::Plugin::WebHDFSOutput::LZOCommandCompressor])
|
77
109
|
def test_compress(data)
|
78
110
|
compress_type, compressor_class = data
|
79
111
|
begin
|
80
|
-
|
81
|
-
|
82
|
-
|
83
|
-
|
84
|
-
|
112
|
+
conf = config_element(
|
113
|
+
"ROOT", "", {
|
114
|
+
"namenode" => "server.local:14000",
|
115
|
+
"path" => "/hdfs/path/file.%Y%m%d.%H%M.log",
|
116
|
+
"compress" => compress_type
|
117
|
+
})
|
118
|
+
d = create_driver(conf)
|
85
119
|
rescue Fluent::ConfigError => ex
|
86
120
|
omit ex.message
|
87
121
|
end
|
88
122
|
assert_equal 'server.local', d.instance.instance_eval{ @namenode_host }
|
89
123
|
assert_equal 14000, d.instance.instance_eval{ @namenode_port }
|
90
124
|
assert_equal '/hdfs/path/file.%Y%m%d.%H%M.log', d.instance.path
|
91
|
-
assert_equal '%Y%m%d%H%M', d.instance.time_slice_format
|
92
125
|
assert_equal compress_type, d.instance.compress
|
93
126
|
assert_equal compressor_class, d.instance.compressor.class
|
94
127
|
end
|
95
128
|
|
96
|
-
def
|
97
|
-
|
98
|
-
|
99
|
-
|
100
|
-
|
101
|
-
|
129
|
+
def test_placeholders_old_style
|
130
|
+
conf = config_element(
|
131
|
+
"ROOT", "", {
|
132
|
+
"hostname" => "testing.node.local",
|
133
|
+
"namenode" => "server.local:50070",
|
134
|
+
"path" => "/hdfs/${hostname}/file.%Y%m%d%H.log"
|
135
|
+
})
|
136
|
+
d = create_driver(conf)
|
102
137
|
assert_equal '/hdfs/testing.node.local/file.%Y%m%d%H.log', d.instance.path
|
103
138
|
end
|
104
139
|
|
105
|
-
|
106
|
-
|
107
|
-
|
108
|
-
|
109
|
-
|
110
|
-
|
111
|
-
|
140
|
+
data("%Y%m%d" => ["/hdfs/path/file.%Y%m%d.log", "/hdfs/path/file.20120718.log"],
|
141
|
+
"%Y%m%d.%H%M" => ["/hdfs/path/file.%Y%m%d.%H%M.log", "/hdfs/path/file.20120718.1503.log"])
|
142
|
+
test "generate_path" do |(path, expected)|
|
143
|
+
conf = config_element(
|
144
|
+
"ROOT", "", {
|
145
|
+
"namenode" => "server.local:14000",
|
146
|
+
"path" => path
|
147
|
+
})
|
148
|
+
d = create_driver(conf)
|
149
|
+
formatter = Fluent::Timezone.formatter("+0900", path)
|
150
|
+
mock(Fluent::Timezone).formatter(Time.now.strftime("%z"), path) { formatter }
|
151
|
+
time = event_time("2012-07-18 15:03:00 +0900")
|
152
|
+
metadata = d.instance.metadata("test", time, {})
|
153
|
+
chunk = d.instance.buffer.generate_chunk(metadata)
|
154
|
+
assert_equal expected, d.instance.generate_path(chunk)
|
155
|
+
end
|
112
156
|
|
113
|
-
|
114
|
-
|
115
|
-
|
116
|
-
|
117
|
-
|
118
|
-
|
119
|
-
|
120
|
-
|
157
|
+
data(path: { "append" => false },
|
158
|
+
ssl: { "ssl" => true, "ssl_verify_mode" => "invalid" },
|
159
|
+
compress: { "compress" => "invalid" })
|
160
|
+
test "invalid" do |attr|
|
161
|
+
conf = config_element(
|
162
|
+
"ROOT", "", {
|
163
|
+
"namenode" => "server.local:14000",
|
164
|
+
"path" => "/hdfs/path/file.%Y%m%d.%H%M.log"
|
165
|
+
})
|
166
|
+
conf += config_element("", "", attr)
|
167
|
+
assert_raise Fluent::ConfigError do
|
168
|
+
create_driver(conf)
|
121
169
|
end
|
122
170
|
end
|
171
|
+
end
|
123
172
|
|
124
|
-
|
125
|
-
|
126
|
-
|
127
|
-
|
128
|
-
|
129
|
-
|
130
|
-
|
131
|
-
|
132
|
-
|
133
|
-
|
173
|
+
sub_test_case "sub section configuration" do
|
174
|
+
def test_time_key
|
175
|
+
conf = config_element(
|
176
|
+
"ROOT", "", {
|
177
|
+
"host" => "namenode.local",
|
178
|
+
"path" => "/hdfs/path/file.%Y%m%d.log"
|
179
|
+
}, [
|
180
|
+
config_element(
|
181
|
+
"buffer", "time", {
|
182
|
+
"timekey" => 1
|
183
|
+
})
|
184
|
+
]
|
185
|
+
)
|
186
|
+
d = create_driver(conf)
|
187
|
+
time = event_time("2012-07-18 15:03:00 +0900")
|
188
|
+
metadata = d.instance.metadata("test", time, {})
|
189
|
+
chunk = d.instance.buffer.generate_chunk(metadata)
|
190
|
+
assert_equal 1, d.instance.buffer_config.timekey
|
191
|
+
assert_equal "/hdfs/path/file.20120718.log", d.instance.generate_path(chunk)
|
192
|
+
end
|
193
|
+
end
|
134
194
|
|
135
|
-
|
136
|
-
|
137
|
-
|
138
|
-
|
139
|
-
|
140
|
-
|
141
|
-
|
142
|
-
|
143
|
-
|
144
|
-
end
|
195
|
+
sub_test_case "using format subsection" do
|
196
|
+
test "blank format means default format 'out_file' with UTC timezone" do
|
197
|
+
format_section = config_element("format", "", {}, [])
|
198
|
+
conf = config_element("match", "", {"host" => "namenode.local", "path" => "/hdfs/path/file.%Y%m%d%H.log"}, [format_section])
|
199
|
+
d = create_driver(conf)
|
200
|
+
time = event_time("2017-01-24 13:10:30 -0700")
|
201
|
+
line = d.instance.format("test.now", time, {"message" => "yay", "name" => "tagomoris"})
|
202
|
+
assert_equal "2017-01-24T20:10:30Z\ttest.now\t{\"message\":\"yay\",\"name\":\"tagomoris\"}\n", line
|
203
|
+
end
|
145
204
|
|
146
|
-
|
147
|
-
|
148
|
-
|
149
|
-
|
150
|
-
|
151
|
-
|
152
|
-
|
153
|
-
|
154
|
-
end
|
205
|
+
test "specifying timezone works well in format section" do
|
206
|
+
format_section = config_element("format", "", {"timezone" => "+0100"}, [])
|
207
|
+
conf = config_element("match", "", {"host" => "namenode.local", "path" => "/hdfs/path/file.%Y%m%d%H.log"}, [format_section])
|
208
|
+
d = create_driver(conf)
|
209
|
+
time = event_time("2017-01-24 13:10:30 -0700")
|
210
|
+
line = d.instance.format("test.now", time, {"message" => "yay", "name" => "tagomoris"})
|
211
|
+
assert_equal "2017-01-24T21:10:30+01:00\ttest.now\t{\"message\":\"yay\",\"name\":\"tagomoris\"}\n", line
|
212
|
+
end
|
155
213
|
|
156
|
-
|
157
|
-
|
158
|
-
|
159
|
-
|
160
|
-
|
161
|
-
|
162
|
-
|
163
|
-
|
164
|
-
|
165
|
-
|
166
|
-
|
214
|
+
test "specifying formatter type LTSV for records, without tag and timezone" do
|
215
|
+
format_section = config_element("format", "", {"@type" => "ltsv"}, [])
|
216
|
+
conf = config_element("match", "", {"host" => "namenode.local", "path" => "/hdfs/path/file.%Y%m%d%H.log"}, [format_section])
|
217
|
+
d = create_driver(conf)
|
218
|
+
time = event_time("2017-01-24 13:10:30 -0700")
|
219
|
+
line = d.instance.format("test.now", time, {"message" => "yay", "name" => "tagomoris"})
|
220
|
+
assert_equal "message:yay\tname:tagomoris\n", line
|
221
|
+
end
|
222
|
+
|
223
|
+
test "specifying formatter type LTSV for records, with inject section to insert tag and time" do
|
224
|
+
inject_section = config_element("inject", "", {"tag_key" => "tag", "time_key" => "time", "time_type" => "string", "localtime" => "false"})
|
225
|
+
format_section = config_element("format", "", {"@type" => "ltsv"}, [])
|
226
|
+
conf = config_element("match", "", {"host" => "namenode.local", "path" => "/hdfs/path/file.%Y%m%d%H.log"}, [inject_section, format_section])
|
227
|
+
d = create_driver(conf)
|
228
|
+
time = event_time("2017-01-24 13:10:30 -0700")
|
229
|
+
line = d.instance.format("test.now", time, {"message" => "yay", "name" => "tagomoris"})
|
230
|
+
assert_equal "message:yay\tname:tagomoris\ttag:test.now\ttime:2017-01-24T20:10:30Z\n", line
|
231
|
+
end
|
232
|
+
end
|
233
|
+
|
234
|
+
sub_test_case "using older configuration" do
|
235
|
+
test "output_data_type json is same with out_file with UTC timezone" do
|
236
|
+
conf = config_element("match", "", {"host" => "namenode.local", "path" => "/hdfs/path/file.%Y%m%d%H.log", "output_data_type" => "json"}, [])
|
237
|
+
d = create_driver(conf)
|
238
|
+
time = event_time("2017-01-24 13:10:30 -0700")
|
239
|
+
line = d.instance.format("test.now", time, {"message" => "yay", "name" => "tagomoris"})
|
240
|
+
assert_equal "2017-01-24T20:10:30Z\ttest.now\t{\"message\":\"yay\",\"name\":\"tagomoris\"}\n", line
|
167
241
|
end
|
168
242
|
end
|
169
243
|
end
|