[Spacewalk-list] Re-provisioning issues

[Tomi Salmi]

Hi,

I'm noticing some unexpected behaviour when re-provisioning physical Fedora
16 systems from Spacewalk 1.7. While the following problems are most likely
not directly related to each other, they are a part of the same workflow.

1) In my environment, most network interfaces get renamed to p5p1 or
similar during the initial bare-metal install, which works flawlessly.
However, re-provisioning a system from Spacewalk causes Cobbler to add an
additional eth0 interface to the profile with the same preferences as the
original interface has. This will generate a duplicate entry in dnsmasq
(managed by cobbler), which then causes dnsmasq to crash/stop.

2) The webgui does not update the "deploy configuration files" and
"kickstart complete" checkmarks when re-provisioning a system, while the
client has finished all steps succesfully. I have added "reset_base_channel
= 0" to /usr/share/rhn/config-defaults/rhn_server.conf but it should be the
cause of this issue?

Basically, re-provisioning a system from SW leaves me with dnsmasq stopped
and a never ending kickstart task in the webgui.

Any advice or suggestions are welcome.

Best regards,
Tomi Salmi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/spacewalk-list/attachments/20121025/dcf215e5/attachment.htm>