rhc commands took too long

[LiAh Sheep]

No description provided.

[LiAh Sheep]

Two approaches

1. Return { success: true, job_id: 'd7689' }. So the user needs to query /job/d7689 to see if the job has completed or not.
2. Add one more parameter callback=http://... so when the job has completed, it will call back using the url.

[Jimmy Chu]

Close due to duplicate with issue #20

[LiAh Sheep]

Not exactly duplicate. rhc commands load time may possibly be improved, without doing two phase notification.

[LiAh Sheep]

I try to isolate the problem. The bug is now reproducible in irc:

2.1.2p95 :005 > require 'httpclient'
 => true 

2.1.2p95 :006 > TCPSocket.new("dashboard.cnpaas.localhost", 80)
 => #<TCPSocket:fd 9> 

2.1.2p95 :007 > TCPSocket.new("dashboard.cnpaas.io", 80)
 => #<TCPSocket:fd 11> 

2.1.2p95 :008 > TCPSocket.new("www.google.com", 443)
 => #<TCPSocket:fd 13> 

2.1.2p95 :009 > TCPSocket.new("broker.openshift.local", 443)
 => #<TCPSocket:fd 14> 

While the first 3 commands return immediately, the last command takes about 5 seconds to complete. Might be due to misconfiguration of Vagrant box or OpenShift.

[LiAh Sheep]

Issue not reproducible on Jimmy’s machine.

[Jimmy Chu]

Let's verify the user creation / app creation / service creation speed is acceptable in staging server

[Jimmy Chu]

@liahsheep Check with you again, if you now agree this issue becomes a two-phase notification enhancement, please mark this issue duplicated.

[LiAh Sheep]

Duplicate of #20.