ignores proxy setting from config on startup #88
Description
So, for some reason codedeploy-agent doesn't read it's config before making the initial connect.
This is our agent config:
[root@azzzz codedeploy-agent]# cat /etc/codedeploy-agent/conf/codedeployagent.yml
---
:log_aws_wire: false
:log_dir: '/var/log/aws/codedeploy-agent/'
:pid_dir: '/opt/codedeploy-agent/state/.pid/'
:program_name: codedeploy-agent
:root_dir: '/opt/codedeploy-agent/deployment-root'
:verbose: false
:wait_between_runs: 1
proxy_uri: 'http://172.aa.bb.cc:3128'
:max_revisions: 5
Here's the initial startup log from /var/log/aws/codedeploy-agent/codedeploy-agent.log
2016-11-24 17:05:13 INFO [codedeploy-agent(3464)]: InstanceAgent::Plugins::CodeDeployPlugin::CommandPoller of master 1008: Received TERM - setting internal shutting down flag and possibly finishing last run
2016-11-24 17:06:12 INFO [codedeploy-agent(3464)]: [Aws::CodeDeployCommand::Client 0 80.349849 1 retries] poll_host_command(host_identifier:"arn:aws:ec2:eu-central-1:zzzzzz:instance/i-zzzzz") Errno::ENETUNREACH Network is unreachable - connect(2)
2016-11-24 17:06:12 ERROR [codedeploy-agent(3464)]: InstanceAgent::Plugins::CodeDeployPlugin::CommandPoller: Error during perform: Errno::ENETUNREACH - Network is unreachable - connect(2) - /usr/share/ruby/net/http.rb:878:in `initialize'
/usr/share/ruby/net/http.rb:878:in `open'
/usr/share/ruby/net/http.rb:878:in `block in connect'
/usr/share/ruby/timeout.rb:66:in `timeout'
/usr/share/ruby/net/http.rb:877:in `connect'
/usr/share/ruby/net/http.rb:862:in `do_start'
/usr/share/ruby/net/http.rb:857:in `start'
/opt/codedeploy-agent/vendor/gems/aws-sdk-core-2.6.11/lib/seahorse/client/net_http/connection_pool.rb:281:in `start_session'
/opt/codedeploy-agent/vendor/gems/aws-sdk-core-2.6.11/lib/seahorse/client/net_http/connection_pool.rb:93:in `session_for'
/opt/codedeploy-agent/vendor/gems/aws-sdk-core-2.6.11/lib/seahorse/client/net_http/handler.rb:116:in `session'
/opt/codedeploy-agent/vendor/gems/aws-sdk-core-2.6.11/lib/seahorse/client/net_http/handler.rb:68:in `transmit'
/opt/codedeploy-agent/vendor/gems/aws-sdk-core-2.6.11/lib/seahorse/client/net_http/handler.rb:42:in `call'
/opt/codedeploy-agent/vendor/gems/aws-sdk-core-2.6.11/lib/seahorse/client/plugins/content_length.rb:12:in `call'
/opt/codedeploy-agent/vendor/gems/aws-sdk-core-2.6.11/lib/aws-sdk-core/json/error_handler.rb:8:in `call'
/opt/codedeploy-agent/vendor/gems/aws-sdk-core-2.6.11/lib/aws-sdk-core/plugins/request_signer.rb:84:in `call'
/opt/codedeploy-agent/vendor/gems/aws-sdk-core-2.6.11/lib/aws-sdk-core/plugins/helpful_socket_errors.rb:10:in `call'
/opt/codedeploy-agent/vendor/gems/aws-sdk-core-2.6.11/lib/aws-sdk-core/plugins/retry_errors.rb:87:in `call'
/opt/codedeploy-agent/vendor/gems/aws-sdk-core-2.6.11/lib/aws-sdk-core/plugins/retry_errors.rb:118:in `retry_request'
/opt/codedeploy-agent/vendor/gems/aws-sdk-core-2.6.11/lib/aws-sdk-core/plugins/retry_errors.rb:101:in `retry_if_possible'
/opt/codedeploy-agent/vendor/gems/aws-sdk-core-2.6.11/lib/aws-sdk-core/plugins/retry_errors.rb:89:in `call'
/opt/codedeploy-agent/vendor/gems/codedeploy-commands-1.0.0/lib/aws/plugins/deploy_agent_version.rb:28:in `call'
/opt/codedeploy-agent/vendor/gems/aws-sdk-core-2.6.11/lib/aws-sdk-core/json/handler.rb:11:in `call'
/opt/codedeploy-agent/vendor/gems/aws-sdk-core-2.6.11/lib/aws-sdk-core/plugins/user_agent.rb:12:in `call'
/opt/codedeploy-agent/vendor/gems/aws-sdk-core-2.6.11/lib/seahorse/client/plugins/endpoint.rb:41:in `call'
/opt/codedeploy-agent/vendor/gems/aws-sdk-core-2.6.11/lib/aws-sdk-core/plugins/param_validator.rb:21:in `call'
After the startup I see successful http 200 states, so later it does respect the proxy, just not went it starts. Thing is, when it starts it needs to find it's instance / iam profile and that seems to be more fragile due to this error.