> During a new installation I noticed that the installation was "hanging" > for quite a while during the installation of portmap. This seems to be caused by the network configuration. I was seeing it for some installs and not for others. The difference is in the /etc/hosts file generated for DHCP and static IP configuration. For DCHP: 127.0.0.1 localhost 127.0.1.1 <hostname>.<domain> <hostname> For static IP: 127.0.0.1 localhost <static IP> <hostname>.<domain> <hostname> The other important factor is that the lo network interface is not up during installation. Note that 'rpcinfo -p' is actually the same as 'rpcinfo -p $(hostname)'. During an installation with DHCP, 'rpcinfo -p' returns: rpcinfo: can't contact portmapper: RPC: Remote system error - Connection refused This can be explained as 'ping 127.0.1.1' returns nothing. If lo is brought up with a netmask of 255.0.0.0 (as on an installed system), it does replies to ping. During an installation with static IP, 'rpcinfo -p' will hang. Reason is that the IP address is resolvable (eth0 has it), but cannot be reached because lo is not up (ping hangs too). Bringing lo up makes 'rpcinfo -p' reply the same as in the DHCP case. During the installation of the standard task, 'rpcinfo -p' will continue to fail as portmap is installed, but never actually started. I'm not sure what the correct solution is here: - set up /etc/hosts the same as in the DHCP case - make sure that lo is up during installation with correct netmask
Attachment:
pgpa1yLMxWpAg.pgp
Description: PGP signature