Hi Rob,
I've been bitten by a problem where my whole list of mail-accounts hasn't
been checked for a time because one mailserver wasn't resolvable due to
provider problems.
I always got the message "couldn't find canonical DNS name of xxx" directly
after startup and fetchmail exited with PS_DNS.
I think the correct behavior should be to print out an errormessage like
that and skip this host but poll the others.
I think you'll find this was patched in 5.8.2:
* Don't cough and die from failure to resolve a skipped host. Resolves
Debian bug #92530
thanks for the hint.
But look at this (using the clean fetchmail 5.8.7 without my patch):
root(_at_)water /etc]# fetchmail -V
This is fetchmail release 5.8.7+SSL
Linux water.local 2.2.19-1 #1 Wed May 30 00:04:44 CEST 2001 i686 unknown
[root(_at_)water /etc]# cat /etc/fetchmailrc
defaults proto pop3
smtphost fire.local
poll notindns.com user xyz password abc to root
poll pop.t-online.de user me password mypwd to root
[root(_at_)water /etc]# fetchmail -v -v -a -f /etc/fetchmailrc
fetchmail: couldn't find canonical DNS name of notindns.com
[root(_at_)water /etc]#
Might it be that the bugfix in 5.8.2 is not complete? Or is my configuration
somehow broken?
Regards,
Gerd