problem with sendmail on obsd. .com.au turned into com.au.com.au
Craig Hammond
boxes.
The config for these different boxes hasn't changed in ages, but for
some
reason, sendmail on them all started up play up all at about the same
time.
Instead of relaying email to localhost as shown below:
Jul 22 20:00:04 int-firewall sm-mta[18685]: k6MA01XY018685:
from=<root@int-firew
all.sbisolutions.com.au>, size=4919, class=0, nrcpts=1,
msgid=<200607221000.k6MA
01Wo0...@int-firewall.sbisolutions.com.au>, proto=ESMTP, daemon=MTA,
relay=smm
s...@localhost.sbisolutions.com.au [127.0.0.1]
It started doing the following:
Jul 23 00:00:09 int-firewall sm-mta[29790]: k6ME010O004304:
to=<root@int-firewal
l.sbisolutions.com.au>, delay=00:00:03, xdelay=00:00:03, mailer=esmtp,
pri=32018
, relay=int-firewall.sbiso...ons.com.au.com.au. [204.13.161.23],
dsn=4.0.0, stat
=Deferred: Connection refused by
int-firewall.sbisolutions.com.au.com.au.
Note the ".com.au.com.au" at the end which obviously doesn't exist.
My configuration is unchanged and has been working for months. I don't
have
".com.au.com.au" listed in any of my config files.
Mail then just started queueing up in /var/spool/mqueue
What ever I've done wrong, I've done wrong in a consistent way as about
5 of
my other obsd boxes (both 3.8-stable & 3.9-stable) in other locations
all did
a similar thing at the same time.
I couldn't figure out what I did wrong, so just to get things working
again,
it tried adding in an entry to /etc/hosts pointing
int-firewall.sbisolutions.com.au.com.au to 127.0.0.1
This didn't work as I guess sendmail doesn't use /etc/hosts.
Finally, I added a the domain .com.au.com.au to named, and added in an
entry
for int-firewall.sbi.... to 127.0.0.1
This horrible hack got things working, but obviously I would like to
find
out what went wrong.
Any idea's
Thanks,
Craig.
knitti
> What ever I've done wrong, I've done wrong in a consistent way as about
> 5 of
> my other obsd boxes (both 3.8-stable & 3.9-stable) in other locations
> all did
> a similar thing at the same time.
I'd rather suspect soem DNS screw up, check whether the intended
name resolves on this box. (check twice ;-), check the bind instance
which should be queried by sendmail and check the name servers
configured in resolv.conf. sometimes this is not the same)
> I couldn't figure out what I did wrong, so just to get things working
> again,
> it tried adding in an entry to /etc/hosts pointing
> int-firewall.sbisolutions.com.au.com.au to 127.0.0.1
>
> This didn't work as I guess sendmail doesn't use /etc/hosts.
I _think_ this depends on your resolv.conf
--knitti
Darrin Chandler
> Note the ".com.au.com.au" at the end which obviously doesn't exist.
> My configuration is unchanged and has been working for months. I don't
> have
> ".com.au.com.au" listed in any of my config files.
>
> Mail then just started queueing up in /var/spool/mqueue
As already noted, this is almost certianly a name resolution problem.
Since it started affecting several machines around the same time you've
either changed resolv.conf/hosts on all those machines (unlikely, as I'm
sure this would have occured to you), or something is wacky with your
DNS server(s).
Missing a "." at the end of a PTR record causes fun problems, and it's
hard to spot. That's exactly the kind of thing you should look for,
though.
--
Darrin Chandler | Phoenix BSD Users Group
dwcha...@stilyagin.com | http://bsd.phoenix.az.us/
http://www.stilyagin.com/ |
mr2j...@gmail.com
The actual problem is because some retard registered au.com.au,
com.com.au, net.com.au, org.com.au, and then added wildcard DNS entries
for them..
Either your DNS server or the zone file for the external domain is
setup incorrectly.
I've managed to convince the domain registrar to demand that the client
remove the dns wildcard which I believe has been done, and everything
should be back to normal. How ever I would suggest you look into DNS
suffix order on your DNS server as this is generally the problem
Regards
James