Merge pull request #160 from zyjiaobj/dev_health

add health and cutoff config

jobs/eventgenerator/spec

@@ -173,4 +173,6 @@ properties:
     description: "Number of consecutive failure to trip the circuit down"
     default: 3
 
-
+  autoscaler.eventgenerator.health.port:
+    description: "the listening port of health endpoint"
+    default: 6204

jobs/eventgenerator/templates/eventgenerator.yml.erb

@@ -38,6 +38,8 @@ server:
 logging:
   level: <%= p("autoscaler.eventgenerator.logging.level") %>
 http_client_timeout: <%= p("autoscaler.eventgenerator.http_client_timeout") %>
+health:
+  port: <%= p("autoscaler.eventgenerator.health.port") %>
 
 db:
   policy_db:

jobs/metricscollector/spec

@@ -107,3 +107,6 @@ properties:
     description: "the method to collect metrics from loggregator"
     default: "streaming"
 
+  autoscaler.metricscollector.health.port:
+    description: "the listening port of health endpoint"
+    default: 6203

jobs/metricscollector/templates/metricscollector.yml.erb

@@ -44,6 +44,8 @@ server:
 logging:
   level: <%= p("autoscaler.metricscollector.logging.level") %>
 http_client_timeout: <%= p("autoscaler.metricscollector.http_client_timeout") %>
+health:
+  port: <%= p("autoscaler.metricscollector.health.port") %>
 
 db:
   policy_db:

jobs/operator/spec

@@ -123,23 +123,23 @@ properties:
   autoscaler.operator.instance_metrics_db.refresh_interval:
     description: "the time interval to prune the instancemetrics database"
     default: 24h
-  autoscaler.operator.instance_metrics_db.cutoff_days:
+  autoscaler.operator.instance_metrics_db.cutoff_duration:
     description: "the cutoff days when pruning instancemetrics database"
-    default: 30
+    default: 24h
 
   autoscaler.operator.app_metrics_db.refresh_interval:
     description: "the time interval to prune the appmetrics database"
     default: 24h
-  autoscaler.operator.app_metrics_db.cutoff_days:
+  autoscaler.operator.app_metrics_db.cutoff_duration:
     description: "the cutoff days when pruning appmetrics database"
-    default: 30
+    default: 48h
 
   autoscaler.operator.scaling_engine_db.refresh_interval:
     description: "the time interval to prune the scalingengine database"
     default: 24h
-  autoscaler.operator.scaling_engine_db.cutoff_days:
+  autoscaler.operator.scaling_engine_db.cutoff_duration:
     description: "the cutoff days when pruning scalingengine database"
-    default: 30
+    default: 240h
 
   autoscaler.operator.scaling_engine.host:
     description: "Host where the scaling engine is running"
@@ -222,5 +222,7 @@ properties:
   autoscaler.operator.enable_db_lock:
     description: "Enable database lock flag"
     default: false
-
-
+
+  autoscaler.operator.health.port:
+    description: "the listening port of health endpoint"
+    default: 6208

jobs/operator/templates/operator.yml.erb

@@ -49,6 +49,8 @@ cf:
 
 logging:
   level: <%= p("autoscaler.operator.logging.level") %>
+health:
+  port: <%= p("autoscaler.operator.health.port") %>
 http_client_timeout: <%= p("autoscaler.operator.http_client_timeout") %>
 instance_metrics_db:
   db:
@@ -57,7 +59,7 @@ instance_metrics_db:
     max_idle_connections: <%= p("autoscaler.instancemetrics_db_connection_config.max_idle_connections") %>
     connection_max_lifetime: <%= p("autoscaler.instancemetrics_db_connection_config.connection_max_lifetime") %>
   refresh_interval: <%= p("autoscaler.operator.instance_metrics_db.refresh_interval") %>
-  cutoff_days: <%= p("autoscaler.operator.instance_metrics_db.cutoff_days") %>
+  cutoff_duration: <%= p("autoscaler.operator.instance_metrics_db.cutoff_duration") %>
 
 app_metrics_db:
   db:
@@ -66,7 +68,7 @@ app_metrics_db:
     max_idle_connections: <%= p("autoscaler.appmetrics_db_connection_config.max_idle_connections") %>
     connection_max_lifetime: <%= p("autoscaler.appmetrics_db_connection_config.connection_max_lifetime") %>
   refresh_interval: <%= p("autoscaler.operator.app_metrics_db.refresh_interval") %>
-  cutoff_days: <%= p("autoscaler.operator.app_metrics_db.cutoff_days") %>
+  cutoff_duration: <%= p("autoscaler.operator.app_metrics_db.cutoff_duration") %>
 
 scaling_engine_db:
   db:
@@ -75,7 +77,7 @@ scaling_engine_db:
     max_idle_connections: <%= p("autoscaler.scalingengine_db_connection_config.max_idle_connections") %>
     connection_max_lifetime: <%= p("autoscaler.scalingengine_db_connection_config.connection_max_lifetime") %>
   refresh_interval: <%= p("autoscaler.operator.scaling_engine_db.refresh_interval") %>
-  cutoff_days: <%= p("autoscaler.operator.scaling_engine_db.cutoff_days") %>
+  cutoff_duration: <%= p("autoscaler.operator.scaling_engine_db.cutoff_duration") %>
 
 scaling_engine:
   scaling_engine_url: https://<%= p("autoscaler.operator.scaling_engine.host") %>:<%= p("autoscaler.operator.scaling_engine.port") %>

jobs/scalingengine/spec

@@ -106,10 +106,7 @@ properties:
 
   autoscaler.scalingengine.health.port:
     description: "the listening port of health endpoint"
-    default: 8081
-  autoscaler.scalingengine.health.emit_interval:
-    description: "the time interval to emit health metrics"
-    default: 15s
+    default: 6204
 
   autoscaler.scalingengine.consul.cluster:
     description: "Consul Cluster configuration. If not specified, scaling engine will not register itself with consul."

jobs/scalingengine/templates/scalingengine.yml.erb

@@ -48,7 +48,6 @@ logging:
 http_client_timeout: <%= p("autoscaler.scalingengine.http_client_timeout") %>
 health:
   port: <%= p("autoscaler.scalingengine.health.port") %>
-  emit_interval: <%= p("autoscaler.scalingengine.health.emit_interval") %>
 
 db:
   policy_db:

jobs/scheduler/spec

@@ -83,5 +83,8 @@ properties:
   autoscaler.policy_db.roles:
     description: "The list of database roles used in policydb database including name/password"  
 
+  autoscaler.scheduler.health.port:
+    description: "the listening port of health endpoint"
+    default: 6204
 
 

jobs/scheduler/templates/application.properties.erb

@@ -63,6 +63,7 @@ org.quartz.scheduler.instanceId=<%= spec.id %>
 
 # scheduler port
 server.port=<%=p('autoscaler.scheduler.port') %>
+scheduler.healthserver.port=<%=p('autoscaler.scheduler.health.port') %>
 
 spring.application.name=scheduler
 spring.mvc.servlet.load-on-startup=1

packages/eventgenerator/spec

@@ -10,7 +10,6 @@ files:
 - app-autoscaler/src/autoscaler/db/sqldb/*.go # gosub
 - app-autoscaler/src/autoscaler/eventgenerator/*.go # gosub
 - app-autoscaler/src/autoscaler/eventgenerator/aggregator/*.go # gosub
-- app-autoscaler/src/autoscaler/eventgenerator/aggregator/fakes/*.go # gosub
 - app-autoscaler/src/autoscaler/eventgenerator/cmd/eventgenerator/*.go # gosub
 - app-autoscaler/src/autoscaler/eventgenerator/config/*.go # gosub
 - app-autoscaler/src/autoscaler/eventgenerator/generator/*.go # gosub

packages/metricscollector/spec

@@ -15,7 +15,6 @@ files:
 - app-autoscaler/src/autoscaler/metricscollector/cmd/metricscollector/*.go # gosub
 - app-autoscaler/src/autoscaler/metricscollector/collector/*.go # gosub
 - app-autoscaler/src/autoscaler/metricscollector/config/*.go # gosub
-- app-autoscaler/src/autoscaler/metricscollector/fakes/*.go # gosub
 - app-autoscaler/src/autoscaler/metricscollector/noaa/*.go # gosub
 - app-autoscaler/src/autoscaler/metricscollector/server/*.go # gosub
 - app-autoscaler/src/autoscaler/metricscollector/testhelpers/*.go # gosub

packages/operator/spec

@@ -15,7 +15,6 @@ files:
 - app-autoscaler/src/autoscaler/operator/*.go # gosub
 - app-autoscaler/src/autoscaler/operator/cmd/operator/*.go # gosub
 - app-autoscaler/src/autoscaler/operator/config/*.go # gosub
-- app-autoscaler/src/autoscaler/operator/fakes/*.go # gosub
 - app-autoscaler/src/autoscaler/routes/*.go # gosub
 - app-autoscaler/src/autoscaler/sync/*.go # gosub
 - app-autoscaler/src/code.cloudfoundry.org/cfhttp/*.go # gosub

packages/scalingengine/spec

@@ -16,7 +16,6 @@ files:
 - app-autoscaler/src/autoscaler/scalingengine/*.go # gosub
 - app-autoscaler/src/autoscaler/scalingengine/cmd/scalingengine/*.go # gosub
 - app-autoscaler/src/autoscaler/scalingengine/config/*.go # gosub
-- app-autoscaler/src/autoscaler/scalingengine/fakes/*.go # gosub
 - app-autoscaler/src/autoscaler/scalingengine/schedule/*.go # gosub
 - app-autoscaler/src/autoscaler/scalingengine/server/*.go # gosub
 - app-autoscaler/src/code.cloudfoundry.org/cfhttp/*.go # gosub

templates/app-autoscaler-deployment-fewer.yml

@@ -136,7 +136,6 @@ instance_groups:
             port: &scalingEnginePort 6104
           health:
             port: &scalingEngineHealthPort 6204
-            emit_interval: 15s  
           defaultCoolDownSecs: 300
           lockSize: 32
           ca_cert: ((scalingengine_ca.ca))
@@ -149,6 +148,8 @@ instance_groups:
       autoscaler:
         scheduler:
           port: &schedulerPort 6102
+          health:
+            port: &schedulerHealthPort 6202
           http_client_timeout: 60
           job_reschedule_interval_millisecond: 10000
           job_reschedule_maxcount: 6
@@ -196,6 +197,8 @@ instance_groups:
           logging:
             level: debug 
           http_client_timeout: 60s
+          health:
+            port: &operatorHealthPort 6208
           require_consul: false
 
 # asapi Instance Group : apiserver&servicebroker
@@ -368,6 +371,8 @@ instance_groups:
           http_client_timeout: 60s
           server:
             port: &metricsCollectorPort 6103
+          health:
+            port: &metricsCollectorHealthPort 6203
           collector:
             refresh_interval: 60s
             collect_interval: 30s
@@ -390,6 +395,8 @@ instance_groups:
           http_client_timeout: 60s
           server:
             port: &eventGeneratorPort 6105
+          health:
+            port: &eventGeneratorHealthPort 6205
           ca_cert: ((eventgenerator_ca.ca))
           server_cert: ((eventgenerator_server.certificate))
           server_key: ((eventgenerator_server.private_key))
@@ -404,8 +411,8 @@ instance_groups:
             evaluation_manager_execute_interval: 40s
             evaluator_count: 20
             trigger_array_channel_size: 200
-          defaultStatWindowSecs: 300
-          defaultBreachDurationSecs: 300
+          defaultStatWindowSecs: 120
+          defaultBreachDurationSecs: 120
           circuitBreaker:
             back_off_initial_interval: 5m
             back_off_max_interval: 120m

templates/app-autoscaler-deployment.yml

@@ -188,6 +188,8 @@ instance_groups:
       autoscaler:
         scheduler:
           port: &schedulerPort 6102
+          health:
+            port: &schedulerHealthPort 6202
           http_client_timeout: 60
           job_reschedule_interval_millisecond: 10000
           job_reschedule_maxcount: 6
@@ -222,7 +224,7 @@ instance_groups:
       consul:
         agent:
           services:
-            scalingengine: 
+            scalingengine:
               check:
                 tcp: 127.0.0.1:6104
                 interval: 30s
@@ -246,7 +248,6 @@ instance_groups:
             port: &scalingEnginePort 6104
           health:
             port: &scalingEngineHealthPort 6204
-            emit_interval: 15s
           defaultCoolDownSecs: 300
           lockSize: 32
           ca_cert: ((scalingengine_ca.ca))
@@ -363,6 +364,8 @@ instance_groups:
           http_client_timeout: 60s
           server:
             port: &metricsCollectorPort 6103
+          health:
+            port: &metricsCollectorHealthPort 6203
           collector:
             refresh_interval: 60s
             collect_interval: 30s
@@ -410,6 +413,8 @@ instance_groups:
           http_client_timeout: 60s
           server:
             port: &eventGeneratorPort 6105
+          health:
+            port: &eventGeneratorHealthPort 6205
           ca_cert: ((eventgenerator_ca.ca))
           server_cert: ((eventgenerator_server.certificate))
           server_key: ((eventgenerator_server.private_key))
@@ -498,6 +503,8 @@ instance_groups:
           enable_db_lock: false
           logging:
             level: debug 
+          health:
+            port: &operatorHealthPort 6208