In order of making it easier to differentiate between existing clients and
clients that I bring up only for installation and configuration I tried
the following configuration.
subnet 192.168.1.0 netmask 255.255.255.0 {
pool {
range 192.168.1.101 192.168.1.150;
default-lease-time 86400;
max-lease-time 604800;
}
pool {
range dynamic-bootp 192.168.1.151 192.168.1.180;
default-lease-time 600;
max-lease-time 86400;
next-server 192.168.1.231;
filename "pxelinux.0";
}
}
in the hope that bootp-clients would get an IP from the second IP range.
Documentation of dhcpd doesn't say if using two pools in one subnet is
actually allowed. It works. Sort of. Many hardware clients, clients from
VMWare server and from xen get an IP from the second range and pxe-boot.
But there are clients who should pxe-boot but get the first IP number and
won't pxe-boot. An example are HP Proliant DL140 servers.
Can I do it different, so it works for all clients? Does the HP server use
another method than bootp for getting the boot image?
Kai
--
Kai Schätzl, Berlin, Germany
Get your web at Conactive Internet Services: http://www.conactive.com
_______________________________________________
CentOS mailing list
CentOS@xxxxxxxxxx
http://lists.centos.org/mailman/listinfo/centos
