logstash-output-azure 0.3.0 → 1.0.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +5 -5
- data/CHANGELOG.md +4 -1
- data/CONTRIBUTORS +1 -0
- data/lib/logstash/outputs/azure.rb +59 -81
- data/lib/logstash/outputs/blob/file_repository.rb +33 -18
- data/lib/logstash/outputs/blob/path_validator.rb +3 -3
- data/lib/logstash/outputs/blob/size_and_time_rotation_policy.rb +6 -4
- data/lib/logstash/outputs/blob/size_rotation_policy.rb +5 -4
- data/lib/logstash/outputs/blob/temporary_file.rb +28 -19
- data/lib/logstash/outputs/blob/temporary_file_factory.rb +28 -16
- data/lib/logstash/outputs/blob/time_rotation_policy.rb +5 -4
- data/lib/logstash/outputs/blob/uploader.rb +29 -22
- data/lib/logstash/outputs/blob/writable_directory_validator.rb +6 -7
- data/logstash-output-azure.gemspec +10 -10
- data/spec/outputs/azure_spec.rb +16 -18
- data/spec/outputs/blob/file_repository_spec.rb +35 -38
- data/spec/outputs/blob/size_and_time_rotation_policy_spec.rb +20 -21
- data/spec/outputs/blob/size_rotation_policy_spec.rb +13 -15
- data/spec/outputs/blob/temporary_file_factory_spec.rb +27 -28
- data/spec/outputs/blob/temporary_file_spec.rb +14 -15
- data/spec/outputs/blob/time_rotation_policy_spec.rb +17 -18
- data/spec/outputs/blob/uploader_spec.rb +28 -32
- data/spec/outputs/blob/writable_directory_validator_spec.rb +8 -9
- data/spec/spec_helper.rb +4 -5
- data/spec/supports/helpers.rb +12 -15
- metadata +9 -9
@@ -1,16 +1,16 @@
|
|
1
|
-
# encoding: utf-8
|
2
1
|
module LogStash
|
3
2
|
module Outputs
|
4
3
|
class LogstashAzureBlobOutput
|
5
4
|
# a sub class of +LogstashAzureBlobOutput+
|
6
5
|
# valdiates the path for the temporary directory
|
7
6
|
class PathValidator
|
8
|
-
INVALID_CHARACTERS = "\^`><"
|
9
|
-
|
7
|
+
INVALID_CHARACTERS = "\^`><".freeze
|
8
|
+
# boolean method to check if a name is valid
|
10
9
|
def self.valid?(name)
|
11
10
|
name.match(matches_re).nil?
|
12
11
|
end
|
13
12
|
|
13
|
+
# define the invalid characters that shouldn't be in the path name
|
14
14
|
def self.matches_re
|
15
15
|
/[#{Regexp.escape(INVALID_CHARACTERS)}]/
|
16
16
|
end
|
@@ -1,22 +1,24 @@
|
|
1
|
-
|
2
|
-
require
|
3
|
-
require "logstash/outputs/blob/time_rotation_policy"
|
1
|
+
require 'logstash/outputs/blob/size_rotation_policy'
|
2
|
+
require 'logstash/outputs/blob/time_rotation_policy'
|
4
3
|
|
5
4
|
module LogStash
|
6
5
|
module Outputs
|
7
6
|
class LogstashAzureBlobOutput
|
8
7
|
# a sub class of +LogstashAzureBlobOutput+
|
9
|
-
# sets the rotation policy
|
8
|
+
# sets the rotation policy
|
10
9
|
class SizeAndTimeRotationPolicy
|
10
|
+
# initialize the class
|
11
11
|
def initialize(file_size, time_file)
|
12
12
|
@size_strategy = SizeRotationPolicy.new(file_size)
|
13
13
|
@time_strategy = TimeRotationPolicy.new(time_file)
|
14
14
|
end
|
15
15
|
|
16
|
+
# check if it is time to rotate
|
16
17
|
def rotate?(file)
|
17
18
|
@size_strategy.rotate?(file) || @time_strategy.rotate?(file)
|
18
19
|
end
|
19
20
|
|
21
|
+
# boolean method
|
20
22
|
def needs_periodic?
|
21
23
|
true
|
22
24
|
end
|
@@ -1,24 +1,25 @@
|
|
1
|
-
# encoding: utf-8
|
2
1
|
module LogStash
|
3
2
|
module Outputs
|
4
3
|
class LogstashAzureBlobOutput
|
5
4
|
# a sub class of +LogstashAzureBlobOutput+
|
6
|
-
# sets the rotation policy by size
|
5
|
+
# sets the rotation policy by size
|
7
6
|
class SizeRotationPolicy
|
8
7
|
attr_reader :size_file
|
9
|
-
|
8
|
+
# initialize the class
|
10
9
|
def initialize(size_file)
|
11
10
|
if size_file <= 0
|
12
|
-
raise LogStash::ConfigurationError
|
11
|
+
raise LogStash::ConfigurationError.new('`size_file` need to be greather than 0')
|
13
12
|
end
|
14
13
|
|
15
14
|
@size_file = size_file
|
16
15
|
end
|
17
16
|
|
17
|
+
# boolean method to check if it is time to rotate
|
18
18
|
def rotate?(file)
|
19
19
|
file.size >= size_file
|
20
20
|
end
|
21
21
|
|
22
|
+
# boolean method
|
22
23
|
def needs_periodic?
|
23
24
|
false
|
24
25
|
end
|
@@ -1,7 +1,6 @@
|
|
1
|
-
|
2
|
-
require
|
3
|
-
require
|
4
|
-
require "fileutils"
|
1
|
+
require 'thread'
|
2
|
+
require 'forwardable'
|
3
|
+
require 'fileutils'
|
5
4
|
|
6
5
|
module LogStash
|
7
6
|
module Outputs
|
@@ -16,6 +15,7 @@ module LogStash
|
|
16
15
|
|
17
16
|
attr_reader :fd
|
18
17
|
|
18
|
+
# initialize the class
|
19
19
|
def initialize(key, fd, temp_path)
|
20
20
|
@fd = fd
|
21
21
|
@key = key
|
@@ -23,27 +23,28 @@ module LogStash
|
|
23
23
|
@created_at = Time.now
|
24
24
|
end
|
25
25
|
|
26
|
+
# gets the created at time
|
26
27
|
def ctime
|
27
28
|
@created_at
|
28
29
|
end
|
29
30
|
|
30
|
-
|
31
|
-
|
32
|
-
end
|
31
|
+
# gets path to temporary directory
|
32
|
+
attr_reader :temp_path
|
33
33
|
|
34
|
+
# gets the size of file
|
34
35
|
def size
|
35
36
|
# Use the fd size to get the accurate result,
|
36
37
|
# so we dont have to deal with fsync
|
37
38
|
# if the file is close we will use the File::size
|
38
|
-
|
39
|
-
|
40
|
-
|
41
|
-
|
42
|
-
end
|
39
|
+
|
40
|
+
@fd.size
|
41
|
+
rescue IOError
|
42
|
+
::File.size(path)
|
43
43
|
end
|
44
44
|
|
45
|
+
# gets the key
|
45
46
|
def key
|
46
|
-
@key.gsub(/^\//,
|
47
|
+
@key.gsub(/^\//, '')
|
47
48
|
end
|
48
49
|
|
49
50
|
# Each temporary file is made inside a directory named with an UUID,
|
@@ -51,20 +52,28 @@ module LogStash
|
|
51
52
|
# we delete the root of the UUID, using a UUID also remove the risk of deleting unwanted file, it acts as
|
52
53
|
# a sandbox.
|
53
54
|
def delete!
|
54
|
-
|
55
|
-
|
55
|
+
begin
|
56
|
+
@fd.close
|
57
|
+
rescue
|
58
|
+
IOError
|
59
|
+
end
|
60
|
+
FileUtils.rm_r(@temp_path, secure: true)
|
56
61
|
end
|
57
62
|
|
63
|
+
# boolean method to determine if the file is empty
|
58
64
|
def empty?
|
59
|
-
size
|
65
|
+
size.zero?
|
60
66
|
end
|
61
67
|
|
68
|
+
# creates the temporary file in an existing temporary directory from existing file
|
69
|
+
# @param file_path [String] path to the file
|
70
|
+
# @param temporary_folder [String] path to the temporary folder
|
62
71
|
def self.create_from_existing_file(file_path, temporary_folder)
|
63
72
|
key_parts = Pathname.new(file_path).relative_path_from(temporary_folder).to_s.split(::File::SEPARATOR)
|
64
73
|
|
65
|
-
TemporaryFile.new(key_parts.slice(1, key_parts.size).join(
|
66
|
-
|
67
|
-
|
74
|
+
TemporaryFile.new(key_parts.slice(1, key_parts.size).join('/'),
|
75
|
+
::File.open(file_path, 'r'),
|
76
|
+
::File.join(temporary_folder, key_parts.slice(0, 1)))
|
68
77
|
end
|
69
78
|
end
|
70
79
|
end
|
@@ -1,9 +1,8 @@
|
|
1
|
-
|
2
|
-
require
|
3
|
-
require
|
4
|
-
require
|
5
|
-
require
|
6
|
-
require "forwardable"
|
1
|
+
require 'socket'
|
2
|
+
require 'securerandom'
|
3
|
+
require 'fileutils'
|
4
|
+
require 'zlib'
|
5
|
+
require 'forwardable'
|
7
6
|
|
8
7
|
module LogStash
|
9
8
|
module Outputs
|
@@ -11,14 +10,15 @@ module LogStash
|
|
11
10
|
# a sub class of +LogstashAzureBlobOutput+
|
12
11
|
# creates the temporary files to write and later upload
|
13
12
|
class TemporaryFileFactory
|
14
|
-
FILE_MODE =
|
15
|
-
GZIP_ENCODING =
|
16
|
-
GZIP_EXTENSION =
|
17
|
-
TXT_EXTENSION =
|
18
|
-
STRFTIME =
|
13
|
+
FILE_MODE = 'a'.freeze
|
14
|
+
GZIP_ENCODING = 'gzip'.freeze
|
15
|
+
GZIP_EXTENSION = 'txt.gz'.freeze
|
16
|
+
TXT_EXTENSION = 'txt'.freeze
|
17
|
+
STRFTIME = '%Y-%m-%dT%H.%M'.freeze
|
19
18
|
|
20
19
|
attr_accessor :counter, :tags, :prefix, :encoding, :temporary_directory, :current
|
21
20
|
|
21
|
+
# initialize the class
|
22
22
|
def initialize(prefix, tags, encoding, temporary_directory)
|
23
23
|
@counter = 0
|
24
24
|
@prefix = prefix
|
@@ -31,41 +31,49 @@ module LogStash
|
|
31
31
|
rotate!
|
32
32
|
end
|
33
33
|
|
34
|
+
# do the rotation
|
34
35
|
def rotate!
|
35
|
-
@lock.synchronize
|
36
|
+
@lock.synchronize do
|
36
37
|
@current = new_file
|
37
38
|
increment_counter
|
38
39
|
@current
|
39
|
-
|
40
|
+
end
|
40
41
|
end
|
41
42
|
|
42
43
|
private
|
44
|
+
|
45
|
+
# if it is not gzip ecoding, then it is txt extension
|
43
46
|
def extension
|
44
47
|
gzip? ? GZIP_EXTENSION : TXT_EXTENSION
|
45
48
|
end
|
46
49
|
|
50
|
+
# boolean method to check if its gzip encoding
|
47
51
|
def gzip?
|
48
52
|
encoding == GZIP_ENCODING
|
49
53
|
end
|
50
54
|
|
55
|
+
# increment the counter in 1 unit
|
51
56
|
def increment_counter
|
52
57
|
@counter += 1
|
53
58
|
end
|
54
59
|
|
60
|
+
# gets the current time
|
55
61
|
def current_time
|
56
62
|
Time.now.strftime(STRFTIME)
|
57
63
|
end
|
58
64
|
|
65
|
+
# method that generate the name of the file to be saved in blob storage
|
59
66
|
def generate_name
|
60
67
|
filename = "#{current_time}.#{SecureRandom.uuid}"
|
61
68
|
|
62
|
-
if tags.
|
69
|
+
if !tags.empty?
|
63
70
|
"#{filename}.tag_#{tags.join('.')}.part#{counter}.#{extension}"
|
64
71
|
else
|
65
72
|
"#{filename}.part#{counter}.#{extension}"
|
66
73
|
end
|
67
74
|
end
|
68
75
|
|
76
|
+
# create the file to be saved in blob storage
|
69
77
|
def new_file
|
70
78
|
uuid = SecureRandom.uuid
|
71
79
|
name = generate_name
|
@@ -85,25 +93,28 @@ module LogStash
|
|
85
93
|
TemporaryFile.new(key, io, path)
|
86
94
|
end
|
87
95
|
|
88
|
-
# clas for the
|
96
|
+
# clas for the encoding
|
89
97
|
class IOWrappedGzip
|
90
98
|
extend Forwardable
|
91
99
|
|
92
100
|
def_delegators :@gzip_writer, :write, :close
|
93
101
|
attr_reader :file_io, :gzip_writer
|
94
102
|
|
103
|
+
# initialize the class for encoding
|
95
104
|
def initialize(file_io)
|
96
105
|
@file_io = file_io
|
97
106
|
@gzip_writer = Zlib::GzipWriter.open(file_io)
|
98
107
|
end
|
99
108
|
|
109
|
+
# gets the path
|
100
110
|
def path
|
101
111
|
@gzip_writer.to_io.path
|
102
112
|
end
|
103
113
|
|
114
|
+
# gets the file size
|
104
115
|
def size
|
105
116
|
# to get the current file size
|
106
|
-
if @gzip_writer.pos
|
117
|
+
if @gzip_writer.pos.zero?
|
107
118
|
# Ensure a zero file size is returned when nothing has
|
108
119
|
# yet been written to the gzip file.
|
109
120
|
0
|
@@ -113,6 +124,7 @@ module LogStash
|
|
113
124
|
end
|
114
125
|
end
|
115
126
|
|
127
|
+
# gets the fsync
|
116
128
|
def fsync
|
117
129
|
@gzip_writer.to_io.fsync
|
118
130
|
end
|
@@ -1,4 +1,3 @@
|
|
1
|
-
# encoding: utf-8
|
2
1
|
module LogStash
|
3
2
|
module Outputs
|
4
3
|
class LogstashAzureBlobOutput
|
@@ -6,19 +5,21 @@ module LogStash
|
|
6
5
|
# sets the policy for time rotation
|
7
6
|
class TimeRotationPolicy
|
8
7
|
attr_reader :time_file
|
9
|
-
|
8
|
+
# initialize the class and validate the time file
|
10
9
|
def initialize(time_file)
|
11
10
|
if time_file <= 0
|
12
|
-
raise LogStash::ConfigurationError
|
11
|
+
raise LogStash::ConfigurationError.new('`time_file` need to be greather than 0')
|
13
12
|
end
|
14
13
|
|
15
14
|
@time_file = time_file * 60
|
16
15
|
end
|
17
16
|
|
17
|
+
# rotates based on time policy
|
18
18
|
def rotate?(file)
|
19
|
-
file.
|
19
|
+
!file.empty? && (Time.now - file.ctime) >= time_file
|
20
20
|
end
|
21
21
|
|
22
|
+
# boolean method
|
22
23
|
def needs_periodic?
|
23
24
|
true
|
24
25
|
end
|
@@ -1,6 +1,5 @@
|
|
1
|
-
|
2
|
-
require
|
3
|
-
require "azure"
|
1
|
+
require 'logstash/util'
|
2
|
+
require 'azure'
|
4
3
|
|
5
4
|
module LogStash
|
6
5
|
module Outputs
|
@@ -9,22 +8,24 @@ module LogStash
|
|
9
8
|
# this class uploads the files to Azure cloud
|
10
9
|
class Uploader
|
11
10
|
TIME_BEFORE_RETRYING_SECONDS = 1
|
12
|
-
DEFAULT_THREADPOOL = Concurrent::ThreadPoolExecutor.new(
|
13
|
-
|
14
|
-
|
15
|
-
|
16
|
-
|
17
|
-
|
18
|
-
|
19
|
-
|
20
|
-
|
21
|
-
|
11
|
+
DEFAULT_THREADPOOL = Concurrent::ThreadPoolExecutor.new(min_threads: 1,
|
12
|
+
max_threads: 8,
|
13
|
+
max_queue: 1,
|
14
|
+
fallback_policy: :caller_runs)
|
15
|
+
|
16
|
+
attr_accessor :upload_options, :logger, :container_name, :blob_account
|
17
|
+
|
18
|
+
# Initializes the class
|
19
|
+
# @param blob_account [Object] endpoint to azure gem
|
20
|
+
# @param container_name [String] name of the container in azure blob, at this point, if it doesn't exist, it was already created
|
21
|
+
def initialize(blob_account, container_name, logger, threadpool = DEFAULT_THREADPOOL)
|
22
22
|
@blob_account = blob_account
|
23
23
|
@workers_pool = threadpool
|
24
24
|
@logger = logger
|
25
|
-
@container_name = container_name
|
26
|
-
|
25
|
+
@container_name = container_name
|
26
|
+
end
|
27
27
|
|
28
|
+
# Create threads to upload the file to the container
|
28
29
|
def upload_async(file, options = {})
|
29
30
|
@workers_pool.post do
|
30
31
|
LogStash::Util.set_thread_name("LogstashAzureBlobOutput output uploader, file: #{file.path}")
|
@@ -32,32 +33,38 @@ module LogStash
|
|
32
33
|
end
|
33
34
|
end
|
34
35
|
|
36
|
+
# Uploads the file to the container
|
35
37
|
def upload(file, options = {})
|
36
38
|
upload_options = options.fetch(:upload_options, {})
|
37
39
|
|
38
40
|
begin
|
39
|
-
content = Object::File.open(file.path,
|
40
|
-
|
41
|
-
|
42
|
-
|
43
|
-
|
41
|
+
content = Object::File.open(file.path, 'rb').read
|
42
|
+
filename = Object::File.basename file.path
|
43
|
+
puts filename
|
44
|
+
blob = blob_account.create_block_blob(container_name, filename, content)
|
45
|
+
puts blob.name
|
44
46
|
rescue => e
|
45
47
|
# When we get here it usually mean that LogstashAzureBlobOutput tried to do some retry by himself (default is 3)
|
46
48
|
# When the retry limit is reached or another error happen we will wait and retry.
|
47
49
|
#
|
48
50
|
# Thread might be stuck here, but I think its better than losing anything
|
49
51
|
# its either a transient errors or something bad really happened.
|
50
|
-
logger.error(
|
52
|
+
logger.error('Uploading failed, retrying', exception: e.class, message: e.message, path: file.path, backtrace: e.backtrace)
|
51
53
|
retry
|
52
54
|
end
|
53
55
|
|
54
56
|
options[:on_complete].call(file) unless options[:on_complete].nil?
|
55
57
|
blob
|
56
58
|
rescue => e
|
57
|
-
logger.error(
|
59
|
+
logger.error('An error occured in the `on_complete` uploader',
|
60
|
+
exception: e.class,
|
61
|
+
message: e.message,
|
62
|
+
path: file.path,
|
63
|
+
backtrace: e.backtrace)
|
58
64
|
raise e # reraise it since we don't deal with it now
|
59
65
|
end
|
60
66
|
|
67
|
+
# stop threads
|
61
68
|
def stop
|
62
69
|
@workers_pool.shutdown
|
63
70
|
@workers_pool.wait_for_termination(nil) # block until its done
|
@@ -1,4 +1,3 @@
|
|
1
|
-
# encoding: utf-8
|
2
1
|
module LogStash
|
3
2
|
module Outputs
|
4
3
|
class LogstashAzureBlobOutput
|
@@ -6,13 +5,13 @@ module LogStash
|
|
6
5
|
# validates that the specified tmeporary directory can be accesed with
|
7
6
|
# write permission
|
8
7
|
class WritableDirectoryValidator
|
8
|
+
# Checks if a path is valid
|
9
|
+
# @param path [String] String that represents the path
|
9
10
|
def self.valid?(path)
|
10
|
-
|
11
|
-
|
12
|
-
|
13
|
-
|
14
|
-
false
|
15
|
-
end
|
11
|
+
FileUtils.mkdir_p(path) unless Dir.exist?(path)
|
12
|
+
::File.writable?(path)
|
13
|
+
rescue
|
14
|
+
false
|
16
15
|
end
|
17
16
|
end
|
18
17
|
end
|
@@ -1,25 +1,25 @@
|
|
1
1
|
Gem::Specification.new do |s|
|
2
2
|
s.name = 'logstash-output-azure'
|
3
|
-
s.version = '0.
|
3
|
+
s.version = '1.0.0'
|
4
4
|
s.licenses = ['Apache-2.0']
|
5
5
|
s.summary = 'Plugin for logstash to send output to Microsoft Azure Blob'
|
6
|
-
#s.description = 'TODO: Write a longer description or delete this line.'
|
7
|
-
#s.homepage = 'TODO: Put your plugin''s website or public repo URL here.'
|
6
|
+
# s.description = 'TODO: Write a longer description or delete this line.'
|
7
|
+
# s.homepage = 'TODO: Put your plugin''s website or public repo URL here.'
|
8
8
|
s.authors = ['Tuffk']
|
9
9
|
s.email = 'tuffkmulhall@gmail.com'
|
10
10
|
s.require_paths = ['lib']
|
11
11
|
|
12
12
|
# Files
|
13
|
-
s.files = Dir['lib/**/*','spec/**/*','vendor/**/*','*.gemspec','*.md','CONTRIBUTORS','Gemfile','LICENSE','NOTICE.TXT']
|
14
|
-
|
13
|
+
s.files = Dir['lib/**/*', 'spec/**/*', 'vendor/**/*', '*.gemspec', '*.md', 'CONTRIBUTORS', 'Gemfile', 'LICENSE', 'NOTICE.TXT']
|
14
|
+
# Tests
|
15
15
|
s.test_files = s.files.grep(%r{^(test|spec|features)/})
|
16
16
|
|
17
17
|
# Special flag to let us know this is actually a logstash plugin
|
18
|
-
s.metadata = {
|
18
|
+
s.metadata = { 'logstash_plugin' => 'true', 'logstash_group' => 'output' }
|
19
19
|
|
20
20
|
# Gem dependencies
|
21
|
-
s.add_runtime_dependency
|
22
|
-
s.add_runtime_dependency
|
23
|
-
s.add_runtime_dependency
|
24
|
-
s.add_development_dependency
|
21
|
+
s.add_runtime_dependency 'azure', '~> 0.7'
|
22
|
+
s.add_runtime_dependency 'logstash-codec-plain'
|
23
|
+
s.add_runtime_dependency 'logstash-core-plugin-api', '~> 2.0'
|
24
|
+
s.add_development_dependency 'logstash-devutils'
|
25
25
|
end
|