Foreman 1.2 and Solaris Unattended Installs

Greetings,

I had previously posted a couple of messages about attempting to setup
foreman Fnattended provisioning for Solaris under version 1.1. It would
appear that some things have changed in 1.2, is that correct?

In my live environment, I have a smart-proxy DHCP/TFTP server on a secure
subnet for building and testing systems. I posted before about troubles
kickstarting OracleLinux on this subnet - I resolved those with a request
to firewall team. At anyrate, I'm moving on to Solaris, and I'm enjoying
how the 1.2 smart-proxy deploys the PXE boot kernels for solaris, whereas
in my lab under 1.1 I had to copy them manually.

Unfortunately, in attempting to rebuild an OracleLinux test box as Solaris

  • I simply edited the OperatingSystem parameters within Foreman and clicked
    Build. Watching the smart-proxy logs things look good. Booting the host,
    not so much. The host still wants to boot to its former syslinux file. I
    deleted that file, and now the host finds no suitable PXE boot solution.

Do I need to configure the DHCP server differently? I would think I'm
missing some kind of syslinux -> pxegrub chain loader, since dhcpd is
pointing all clients at pxelinux.0.

What am I missing here?

On a side note, I did have to go into the Provisioning Settings and set *
safemode_render* to false, in order to get the supplied Jumpstart Template
to render.

Looking spoofed at the PXEGrub, I get this:

There was an error rendering the Jumpstart default PXEGrub template: undefined local variable or method `multiboot' for #<#<Class:0x007fb80c0051b0>:0x007fb80dc338a0>

Perhaps that has something to do with my problem not being able to PXE boot
the solaris host?

ยทยทยท On Wednesday, August 7, 2013 10:51:49 AM UTC-4, Sean Alderman wrote: > > Greetings, > > I had previously posted a couple of messages about attempting to setup > foreman Fnattended provisioning for Solaris under version 1.1. It would > appear that some things have changed in 1.2, is that correct? > > In my live environment, I have a smart-proxy DHCP/TFTP server on a secure > subnet for building and testing systems. I posted before about troubles > kickstarting OracleLinux on this subnet - I resolved those with a request > to firewall team. At anyrate, I'm moving on to Solaris, and I'm enjoying > how the 1.2 smart-proxy deploys the PXE boot kernels for solaris, whereas > in my lab under 1.1 I had to copy them manually. > > Unfortunately, in attempting to rebuild an OracleLinux test box as Solaris > - I simply edited the OperatingSystem parameters within Foreman and clicked > Build. Watching the smart-proxy logs things look good. Booting the host, > not so much. The host still wants to boot to its former syslinux file. I > deleted that file, and now the host finds no suitable PXE boot solution. > > Do I need to configure the DHCP server differently? I would think I'm > missing some kind of syslinux -> pxegrub chain loader, since dhcpd is > pointing all clients at pxelinux.0. > > What am I missing here? >