I just installed Red Hat 3.0.3 (was running 2.0). Now my sendmail daemon
takes 60 seconds to load. Any ideas what may be causing this? Not a big
problem, but would be nice if I could get it to load faster. Note that
this is on a standalone machine, used to connect to my ISP and do various
Internet stuff. It is a P100 w/ 24 megs of RAM. Thanks in advance for
your help. =8^)
Check if the lo interface is configured with network-address 127.0.0.1.
The interface lo should have this address, but the network address
should be 127.0.0.0. When I upgraded to kernel 2.0 and installed all
the new packages to support it I also needed to modify my /etc/hosts file.
I had to change the entry
192.168.0.1 onatopp.my.domain onatopp
Seems like sendmail needs the long version and syslogd the short one,
but I'm not sure if this is a problem if you only have a loopback interface.
Your /etc/networks file should have an entry like this:
Also make sure RedHat is not confused wether your loopback interface
is called lo or lo0. According to their manual and network-setup it's
lo0, but lo makes more sense. I use "lo" and have renamed the config
file for this interface /etc/sysconfig/network-scripts/ifcfg-lo. The
/etc/rc.d/init.d/network script that starts the different interfaces
looks for config-files called ifcfg-lo?, so you'll have to change that
in the script. Also make sure you have "NETWORK=127.0.0.0" and
"DEVICE=lo" in ifcfg-lo and that you're not trying a route "add -net
127.0.0.1" in /etc/sysconfig/network-scripts/ifup-lo. Maybe you can
use the lo0 name for your interface instead, if you want to, just make
sure you have consistency all over. What is definitely wrong is to use
"NETWORK=127.0.0.1" in ifcfg-lo, "route add -net 127.0.0.1" in ifup-lo
or "loopback 127.0.0.1" in /etc/networks.
Hope this helps. At least I had a lot of fun before I got it straigth
Jon Martin Solaas