The address which is supplied to the node when PXE booting does not have to be the same as the interface's address as configured by kickstart. You should be able to set up your DHCP server to supply an address on your provisioning network during PXE boot, then configure kickstart to configure the interface with it's final address. Configure kickstart to shut the machine down when the initial configuration is done. When the machine shuts down move the interface to its production network and reboot.
----- Original Message ----- > I'm building out a foreman environment (technically, RH Satellite > 6.1) > to do provisioning and management of Linux VMs at my current client > (soon-to-be employer). At the moment, they do the majority of their > builds by cloning an existing server (and bringing along all the junk > and baggage that entails). Rarely, they use an older Satellite 5 > server > to do their builds (if there isn't a known server they can clone). In > the Satellite5 case, they login to vCenter, create the VM on a build > network, then login to the console. PXE then presents them with a > list > of possible images and a kickstart pre-script prompts for hostname, > final IP address, DNS, etc. After they input that information, the > build happens, then the server reboots with the final IP address and > hopefully everything configured properly. They have to manually > change > the VM network segment in vCenter and fix up any issues. > In my Satellite6 setup, we obviously configure all the information in > the Foreman client which has vCenter rights to create the VM and boot > it > on the build network. Either kickstart or puppet does all the initial > configuration and patching. So far, a much simplier process. However, > once the VM is built, I need to migrate it to its final network home. > I > was hoping to just be able to update the record in Foreman and have > it > automatically update the VM and the OS. This doesn't happen, though. > :-( > Is anyone using Foreman/Satellite6 with this separate build network > and > have ideas on how to automatically migrate it to the final home? > Surely, we can't be the only ones who do our builds on a centralized > network. > Thanks, > -spp > _______________________________________________ > Tech mailing list > Tech@lists.lopsa.org > https://lists.lopsa.org/cgi-bin/mailman/listinfo/tech > This list provided by the League of Professional System > Administrators > http://lopsa.org/ -- Patrick Landry Director, UCSS University of Louisiana at Lafayette p...@louisiana.edu
_______________________________________________ Tech mailing list Tech@lists.lopsa.org https://lists.lopsa.org/cgi-bin/mailman/listinfo/tech This list provided by the League of Professional System Administrators http://lopsa.org/
