Sha256: b2227fa469d34e401adc5760fb5f16c48c0f249ffdc36ce4f0e5747b260279be
Contents?: true
Size: 1.56 KB
Versions: 9
Compression:
Stored size: 1.56 KB
Contents
Feature: request balancer health check In order to enhance system availability for customers RightSupport should track endpoint health when making load-balanced REST requests So apps do not become hung during network failures Scenario: mixed servers (overloaded, blackholed) using health check Given 4 overloaded servers And 4 blackholed servers And HealthCheck balancing policy When a client makes a load-balanced request to '/' with timeout 1 and open_timeout 2 Then the request should raise in less than 12 seconds Scenario: mixed servers (well-behaved, blackholed) using health check Given 4 servers that respond with 200 And 4 blackholed servers And HealthCheck balancing policy When a client makes a load-balanced request to '/' with timeout 1 and open_timeout 2 Then the request should complete in less than 8 seconds Scenario: mixed servers (overloaded, well-behaved, blackholed) using health check Given 1 overloaded server And 1 server that responds with 200 And 1 blackholed server And HealthCheck balancing policy When a client makes a load-balanced request to '/' with timeout 1 and open_timeout 2 Then the request should complete in less than 3 seconds Scenario: mixed servers (condition commented by Tony https://rightscale.acunote.com/projects/2091/tasks/23987#comments) using health check Given 3 overloaded servers And 1 server that responds with 200 And HealthCheck balancing policy When a client makes a load-balanced request to '/' with timeout 1 and open_timeout 2 Then the request should complete in less than 4 seconds
Version data entries
9 entries across 9 versions & 1 rubygems