Greetings,
I'm running fetchmail via ppp on RedHat 6.2. To ensure that all the users on my
system automatically poll their individual mailservers on a regular
basis whenever my ppp-connection is up, I've added the following line (once for
each user) to /etc/rc.d/rc.local
su clause -c 'fetchmail -d 180 -M ppp0 -I ppp0/255.255.255.255/0.0.0.0 -s
--smtphost localhost --timeout 10'
However, the daemon is never started because it complains about:
fetchmail: couldn't find canonical DNS name of mail.my-isp.dk
Apparently fetchmail tries to get the DNS name of mail.my-isp.dk at startup,
when the ppp-connection is down and my isp's nameservers are unavailable. If I
insert the IP-address of mail.my-isp.dk in /etc/hosts everything is fine, but
I'd hate to "hardcode" the IP addresses requested by all users in /etc/hosts.
Is it possible (or shouldn't it be?) that when the interface option is used,
the address is not resolved until it has been verified that the connection is
actually up?
If I run the above command-line for fetchmail whenever my ppp-connection is up,
it appears to work perfectly.
Hope I made my problem clear.
Thanks,
Claus
OS: RedHat 6.2. (kernel 2.2.16-3)
Fetchmail release 5.6.2
Compiler: gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)