Description of problem:

Playing around with deployables, I found that dropping a character from either the image id or the build id is flagged as an error prior to adding the deployable, stating that the particular id could not be found.  But, if you do this with the hw profile name, it is not flagged as an error until you actually start the deployable which is a little late, especially in multi-assembly deployments.


Version-Release number of selected component (if applicable):
aeolus-configure-2.0.1-0.el6.20110628141215gitb8aaf85.noarch
rubygem-aeolus-cli-0.0.1-1.el6.20110701204941gitd66e27c.noarch
aeolus-conductor-doc-0.3.0-0.el6.20110701204941gitd66e27c.noarch
aeolus-conductor-daemons-0.3.0-0.el6.20110701204941gitd66e27c.noarch
aeolus-all-0.3.0-0.el6.20110701204941gitd66e27c.noarch
aeolus-conductor-0.3.0-0.el6.20110701204941gitd66e27c.noarch


Steps to Reproduce:
1.  Create a deployable xml with a bad build id
2.  Create a new deployment for a pool specifying xml from step 1
3.  Notice an error is thrown on a bad id and the deployable is not added
4.  Create a deployable xml and specify a non-existent hw profile
5.  Create a new deployment for a pool specifying xml from step 4
6.  Notice that the bad hw profile is not flagged as a error and you can proceed to starting the deployment.
  
Actual results:


Expected results:


Additional info: