Description of problem: disklessrc only configures eth0. If pxe-booted from (what turns out to be) e.g. eth1, the netbooting stops and you get a simple bash prompt. Version-Release number of selected component (if applicable): 0.1.40 How reproducible: always Steps to Reproduce: 1. put utp cable in netboot-client, in eth port that won't be eth0 2. pxe boot 3. Actual results: simple bash prompt, no network Expected results: full net boot, root over nfs. Additional info: I have changed disklessrc so that it can use other eth devices than only eth0. It may be a bit quick and/or dirty, but it works for me. This is the diff (added indents and "\" for readability). The < is my version, > is old version. 195c195 < /bin/dhclient -1 -pf /tmp/dhclient.pid -lf /tmp/dhclient.leases \ >/tmp/dhclient.out 2>&1 --- > /bin/dhclient -1 -cf /tmp/dhclient.conf -pf /tmp/dhclient.pid -lf \ /tmp/dhclient.leases $ETHERNET >/tmp/dhclient.out 2>&1 231c231,233 < IP=$(ip addr show|grep "inet "|grep -v 127|awk '{ print $2 }' |awk -F/ '{ \ print $1 }') --- > IP=`/sbin/ip addr show ${ETHERNET} scope global primary | > grep '^[\ \ ]*inet.*scope global '${ETHERNET} | > sed 's/^.*inet\ //;s/[\/\ ].*$//'`;
I understand that you CAN use other interfaces but you have to tell the client that via the ETHERNET variable in the pxe boot conf. With my fix it doesn't matter where you put your utp cable, as long as the machine can pxe boot from it. Some Opteron boards use eth0 for the e100 and eth1 and eth2 for the tg3 driver. This triggered me to create this fix.
System-config-netboot is now in maintenance mode only, no further development is planned for RHEL 4.x. Closing as WONTFIX.