yahoo-eng-team team mailing list archive

[Dolph Mathews <1149987@xxxxxxxxxxxxxxxxxx>]

Reverting the above patch as it caused subsequent issues when combined
with patchset 16 of https://review.openstack.org/#/c/50491/

After discussing the issue further, this bug report & patch doesn't
address the underlying issue which is that we're raising 400's based on
typeerrors in the first place, rather than doing actual request
validation in a single phase, and emitting 400's there (and only there).

** Changed in: keystone
       Status: Fix Committed => Won't Fix

-- 
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to Keystone.
https://bugs.launchpad.net/bugs/1149987

Title:
  TypeErrors caught by keystone.common.wsgi.Application could return
  better error messages

Status in OpenStack Identity (Keystone):
  Won't Fix

Bug description:
  Sending an invalid request that raises a TypeError (such as a POST w/
  empty body) against the controller method results in an error message
  that's not particularly helpful.

  POST http://localhost:35357/v3/users
  X-Auth-Token: 1dedbe6f49c14a5a82f3cc7765317c2b

  
  400 Bad Request
  Date: Wed, 06 Mar 2013 17:21:52 GMT
  Vary: X-Auth-Token
  Content-Length: 112
  Status: 400
  Content-Type: application/json

  {
    "error": {
      "message": "create_user() takes exactly 3 arguments (2 given)", 
      "code": 400, 
      "title": "Bad Request"
    }
  }

  The message returned is quite vague, albeit better than a 500. I'm
  wondering if we could inspect the method signature to figure out
  what's wrong to provide more useful error feedback?

To manage notifications about this bug go to:
https://bugs.launchpad.net/keystone/+bug/1149987/+subscriptions