[Spacewalk-list] Deploy configs during kickstart fails

[Giovanni Torres]

I get the following error during kickstarts:
Summary: Deploy config files to system scheduled by (none)
Details:...
This action's status is: FailedClient execution returned "Failed deployment and rollback, information on user 'user1' could not be found" (code 50)

Here's what I am doing in the %post section:
# Creating user1groupadd -g 500 user1useradd -u 500 -g 500 -m user1
# Pulling down configs (some are owned by user1)/usr/bin/rhncfg-client get


This is suspicion:The activation key I am using subscribes to a couple configuration channels.  At the time of activation, it tries to pull down the configs and fails because /home/user1 does not exist, but one of the configuration channels includes a custom /home/user1/.bashrc.  This is where the error is log in the Events tab of the host in the web UI.
Then, later in the kickstart, I create user1 and pull down the configs again manually, as shown above.  At this point, the config deployment works, but does not get logged in the web UI.

Can someone confirm that is indeed the behaviour?   If this is the case, is there a good way in Spacewalk to make sure the a user's directory exists before trying to apply a config deployment?
Thanks,Giovanni
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/spacewalk-list/attachments/20130121/7fe5b411/attachment.htm>