Description of problem:
Dynect was in maintenance from September 26, 2013 at 09:00 UTC until 10:00 UTC, causing its APIs to be unavailable. Moreover, the error was actually recorded in development.log:

^[[0;37m2013-09-26 05:01:56.865^[[0m [^[[0;37mDEBUG^[[0m] Retrying dyn_login after exception caught from DNS request: Login failed: Net::HTTPServiceUnavailable (pid:18844)
^[[0;37m2013-09-26 05:01:56.865^[[0m [^[[0;37mDEBUG^[[0m] DYNECT Login with path: /REST/Session/ (pid:18844)
^[[0;37m2013-09-26 05:01:56.955^[[0m [^[[0;37mDEBUG^[[0m] Execute DeregisterDnsOp (pid:8343)
^[[0;37m2013-09-26 05:01:56.969^[[0m [^[[0;37mDEBUG^[[0m] DYNECT Login with path: /REST/Session/ (pid:8343)
^[[0;37m2013-09-26 05:01:57.430^[[0m [^[[0;37mDEBUG^[[0m] Response code: 503 (pid:18844)
^[[0;37m2013-09-26 05:01:57.430^[[0m [^[[0;37mDEBUG^[[0m] Response body: {"status": "failure", "data": {}, "job_id": null, "msgs": [{"INFO": "The Dynect API is currently in maintenance mode.  Please try again later.", "SOURCE": "API-A", "LVL": "ERROR"}]} (pid:18844)
^[[0;37m2013-09-26 05:01:57.439^[[0m [^[[31mERROR^[[0m] Login failed: Net::HTTPServiceUnavailable (pid:18844)

However, the CLI only prompted the following error message when using rhc commands:

Login failed: Net::HTTPServiceUnavailable

which is very technical and user-unfriendly. It would be better if we could return a more readable and specific message.

Version-Release number of selected component (if applicable):
devenv_3832

How reproducible:
when Dynect is in maintenance

Steps to Reproduce:
1. Create an app via RHC when Dynect is in maintenance
2.
3.

Actual results:
The creation fails with the error:

[zhe@fedora runtst]$ rhc app create srb8 ruby-1.8
Application Options
-------------------
  Namespace:  test
  Cartridges: ruby-1.8
  Gear Size:  default
  Scaling:    no

Creating application 'srb8' ... Login failed: Net::HTTPServiceUnavailable


Expected results:
It could be better if the error message is more specific and user-friendly.

Additional info: