hadoop-metrics 0.1.1 → 0.1.2

data/lib/hadoop_metrics/api.rb

@@ -3,22 +3,45 @@ require 'json'
 
 module HadoopMetrics
   module API
-    def jmx
-      jmx_json = HadoopMetrics.get_response(@jmx_endpoint)['beans'].first
-      @json_value_fields.each { |f|
-        jmx_json[f] = JSON.parse(jmx_json[f])
-      }
-      if @snake_case
-        jmx_json = HadoopMetrics.snake_cased(jmx_json)
-      end
-
-      jmx_json
+    def initialize(host, port, opts = {})
+      @endpoint = "#{host}:#{port}"
+      @metrics_endpoint = URI("http://#{@endpoint}/metrics?format=json")
+      @snake_case = opts[:snake_case] || true
+      @name = opts[:name] || host
     end
 
+    attr_reader :name
+
     def metrics
       HadoopMetrics.get_response(@metrics_endpoint)
     end
 
+    def gc
+      disable_snake_case {
+        result = via_jmx('java.lang:type=GarbageCollector,name=*').map { |jmx_gc_info|
+          gc_info = {'type' => (/PS Scavenge/.match(jmx_gc_info['name']) ? 'minor' : 'major')}
+          gc_info['estimated_time'] = jmx_gc_info['CollectionTime']
+          gc_info['count'] = jmx_gc_info['CollectionCount']
+          gc_info['last_start'] = jmx_gc_info['LastGcInfo']['startTime']
+          gc_info['last_duration'] = jmx_gc_info['LastGcInfo']['duration']
+          gc_info
+        }
+      }
+    end
+
+    def via_jmx(query, json_fields = [])
+      HadoopMetrics.get_response(URI("http://#{@endpoint}/jmx?qry=#{query}"))['beans'].map { |jmx_json|
+        json_fields.each { |f|
+          jmx_json[f] = JSON.parse(jmx_json[f])
+        }
+        if @snake_case
+          jmx_json = HadoopMetrics.snake_cased(jmx_json)
+        end
+
+        jmx_json
+      }
+    end
+
     private
 
     def group_by(category, target, column)
@@ -35,6 +58,15 @@ module HadoopMetrics
       }
     end
 
+    def disable_snake_case
+      old_snake_case = @snake_case
+      @snake_case = false
+
+      yield
+    ensure
+      @snake_case = old_snake_case
+    end
+
     def method_missing(method, *args)
       category, target = method.to_s.split('_', 2)
       group_by(category, target, *args)
@@ -62,10 +94,14 @@ module HadoopMetrics
   end
 
   def self.snake_cased(json)
-    snake_cased = {}
-    json.each_pair { |k, v|
-      snake_cased[HadoopMetrics.to_snake_case(k.dup)] = json[k]
+    snake_cased_json = {}
+    json.each_pair { |key, value|
+      v = json[key]
+      if v.is_a?(Hash)
+        v = snake_cased(v)
+      end
+      snake_cased_json[HadoopMetrics.to_snake_case(key.dup)] = v
     }
-    snake_cased
+    snake_cased_json
   end
 end

data/lib/hadoop_metrics/job_tracker.rb

@@ -4,11 +4,10 @@ module HadoopMetrics
   class JobTracker
     include API
 
-    def initialize(host, port, opts = {})
-      @jmx_endpoint = URI("http://#{host}:#{port}/jmx?qry=hadoop:service=JobTracker,name=JobTrackerInfo")
-      @metrics_endpoint = URI("http://#{host}:#{port}/metrics?format=json")
-      @json_value_fields = %W(SummaryJson AliveNodesInfoJson BlacklistedNodesInfoJson QueueInfoJson)
-      @snake_case = opts.has_key?(:snake_case) ? opts[:snake_case] : true
+    JSON_FILED_VALUES = %W(SummaryJson AliveNodesInfoJson BlacklistedNodesInfoJson QueueInfoJson)
+
+    def info
+      via_jmx('hadoop:service=JobTracker,name=JobTrackerInfo', JSON_FILED_VALUES).first
     end
 
     def fairscheduler_pools(column = 'name')

data/lib/hadoop_metrics/task_tracker.rb

@@ -4,11 +4,10 @@ module HadoopMetrics
   class TaskTracker
     include API
 
-    def initialize(host, port, opts = {})
-      @jmx_endpoint = URI("http://#{host}:#{port}/jmx?qry=hadoop:service=TaskTracker,name=TaskTrackerInfo")
-      @metrics_endpoint = URI("http://#{host}:#{port}/metrics?format=json")
-      @json_value_fields = %W(TasksInfoJson)
-      @snake_case = opts[:snake_case] || true
+    JSON_FILED_VALUES = %W(TasksInfoJson)
+
+    def info
+      via_jmx('hadoop:service=TaskTracker,name=TaskTrackerInfo', JSON_FILED_VALUES).first
     end
 
     def shuffle_output(column = 'sessionId')

data/lib/hadoop_metrics/version.rb

@@ -1,3 +1,3 @@
 module HadoopMetrics
-  VERSION = '0.1.1'
+  VERSION = '0.1.2'
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: hadoop-metrics
 version: !ruby/object:Gem::Version
-  version: 0.1.1
+  version: 0.1.2
   prerelease: 
 platform: ruby
 authors:
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-04-25 00:00:00.000000000 Z
+date: 2013-04-30 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake
@@ -58,7 +58,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: 460637235756246510
+      hash: -596763061643403807
 required_rubygems_version: !ruby/object:Gem::Requirement
   none: false
   requirements:
@@ -67,7 +67,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: 460637235756246510
+      hash: -596763061643403807
 requirements: []
 rubyforge_project: 
 rubygems_version: 1.8.23