>> Are you running IPchains or IPtables on the server? >> >> If so, make sure you open up both ports 67 and 68, for both TCP and UDP, >> on the interface you are using for the internal network. > Not on the server, by it turned out they were running on the client even > though I didn't want them to (which I seem to remember could also cause > problems like this, now that you mentioned it) - they had simply been > enabled by mistake. This was one difference from other hosts that I > failed to spot. > > However, event though I have now disabled the services, I'm still not > able to start the network. I have tried (in the order indicated): > 1. 'chkconfig iptables off; chkconfig ipchains off', then reboot > 2. 'anaconda --reconfig' - set "No firewall" in security config. > 3. Removed iptables, ipchains and related packages from the system via > 'rpm -e' > 4. Rebooted several times, retried config update etc. > > Is there anything else I must do to get rid of the packet filters? Or > could the DHCP failure have a different cause? It turns out that the host doesn't get the address properly when booting with the network installation CD (no, I wasn't going to re-install, I just wanted to test a different DHCP client. Really. I'm not a Windows user, after all), which I think rules out any client config problems.
In other words, there must be a problem with the hardware, the server, or the routing on the network, but what might cause this kind of behaviour? Is there any way a broken network card or similar could give results like this? (I'll try replacing some of the hardware tomorrow, I guess.) Note that I've also tried removing the host's "fixed-address" entry from the server config, so that an IP-address would be assigned from the dynamic pool, but I got exactly the same result when I did. - Toralf _______________________________________________ Redhat-list mailing list [EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
