Here is the web log for a successful edit:

Started GET "/hosts/homer.domain.com" for 172.x.x.x at Fri Aug 10 09:32:31 -0400 2012
  Processing by HostsController#show as HTML
  Parameters: {"id"=>"homer.domain.com"}
Rendered hosts/_overview.html.erb (13.8ms)
Rendered hosts/_metrics.html.erb (0.4ms)
Rendered bookmarks/_list.html.erb (2.3ms)
Rendered bookmarks/_list.html.erb (0.5ms)
Rendered home/_settings.html.erb (8.0ms)
Rendered home/_topbar.rhtml (19.0ms)
Rendered hosts/show.html.erb within layouts/application (526.2ms)
Completed 200 OK in 741ms (Views: 492.2ms | ActiveRecord: 44.6ms)

Note that this particular host (homer.domain.com) has a FQDN. Most of my hosts in Foreman/Puppet are listed with only their hostname and not the FQDN. It could be something with having hostname vs. FQDN.

I see the GET parameter for id is actually set in this successful request but not in my failed requests.

After looking into it some more, I see that the issue does in fact occur if the domain is not set on a host. Specifically, on the database side if the domain_id column is not set in the hosts table, then I get the 'Host not Found' error when submitting an edit to an individual host.

Unfortunately, I had to manually set the domain_id to 1 (which corresponds to my domain) in the database for all of my hosts and also set the default value to 1 for when new hosts are added. I tried setting the domain for an entire Hostgroup in the Foreman web gui and it appeared to set successfully, but when I went to the domains list, no additional hosts were listed for my particular domain. I could not set the domain on individual hosts in the web gui, of course, because of the Host not found issue in general.

Now when I make an edit to a host, Foreman converts the hostname to a FQDN. Thats OK, but looks like I need to create new certs for all my hosts now since the current certs don't have the FQDN.

Hope this info helps.

Scratch this comment:

Now when I make an edit to a host, Foreman converts the hostname to a FQDN. Thats OK, but looks like I need to create new certs for all my hosts now since the current certs don't have the FQDN.

The error I see is related to a conflicting Sshkey. My ssh_known_hosts file will need to be updated with fqdn's now, but thats a puppet issue.

err: Could not retrieve catalog from remote server: Error 400 on SERVER: Another local or imported resource exists with the type and title Sshkey[chicken] on node chicken