Sendmail via relay server

[Acxiom Unix Team]

I am having problems in using sendmail on a box. When i am sending an email and it goes via the RELAY HOST it fails and all i get is this error in the mail.log...

Please see the attached document with more details. The basics of it is i know sendmail works locally on both boxes involved and i know when coming from the originating server it works in a "fashion"......I am guessing it has something to do with host names and resolution of them but wonder if anyone can offer some help.

Just FYI

Originating server is running 11.31 and the relay host is running 11.11.

not sure if that is an issue....

[TTr]

What does the /etc/hosts look like on the originating server dalmas for the dalmas name itself and the localhost?

It should be

172.30.5.205 dalmas.co-op.co.uk dalmas

(Again, this is the /etc/hosts on the dalmas server)

[Acxiom Unix Team]

If i look in /etc/hosts on dalmas this is the result.

dalmas:/var/adm/syslog> cat /etc/hosts | grep -e dalmas
172.30.2.231 dalmas2.co-op.co.uk dalmas2 # RETAIL UX

172.30.5.205 dalmas.co-op.co.uk dalmas

11.11.11.5 dalmas11.co-op.co.uk dalmas11
12.12.12.5 dalmas12.co-op.co.uk dalmas12

The bottom 2 entries are used by ServiceGuard.......

But as you can see there is an entry in as you mentioned but its not the first entry for this box....

[TTr]

Why do you have dalmas5 for the originating server on the mail relay server?

The dsn=5.6.0 error is cause among other things by a hostname resolution issue. You may want to try making the name resolution of the originating server consistent both in itself and in the mail relay server.

Google says that another cause may be the localhost, mailhost line. What does your 127.0.0.1 line look like on each server?

[Steven E. Protter]

Shalom,

andy.pilgrim@co-operative.coop

That is not a valid domain name.

That domain name lookup will fail on the relay server, even if you try and make it valid in /etc/hosts

In this case, no mail delivery.

SEP

[Acxiom Unix Team]

Two answers here.

TTr, I dont know why they have dalmas5 in the host file. Only been at this place 1 week so got to find that out, but i guessed that might be causing issues...

Anyway, localhost on both servers look like this;

dalmas:/root> grep localhost /etc/hosts
127.0.0.1 localhost.co-op.co.uk localhost loopback

daly:/root> grep localhost /etc/hosts
127.0.0.1 localhost loopback


SEP,

How do you know that andy.pilgrim@co-operative.coop is not a valid domain name??? Might be a daft question but i can 100% send emails to and from that address both internally and externally.

[TTr]

> dalmas:/root> grep localhost /etc/hosts
> 127.0.0.1 localhost.co-op.co.uk localhost loopback

I know it is not your doing but I don't understand why people mess around with the loopback name/address. It is there for a reason and it is there on each and every server.


> Feb 11 14:20:25 daly sendmail[11290]: n1BEK3J11282: to=andy.pilgrim@co-operative.coop, delay=00:00:09, xdelay=00:00:00, mailer=esmtp, pri=120012, relay=mailer1.co-op.co.uk. [217.33.9.74], dsn=5.6.0, stat=Data format error

In your attachment you showed us the "to" record. There is a "from" record as well before this for the same mail message ID n1BEK3J11282. Take a look at that and see what hostname shows for the dalmas server. This name must be a valid (officially resolvable) name by the relay server. This is where the problem might be. You probably need to fix the hostname for the dalmas server.

[Acxiom Unix Team]

TTr,

I had a discussion (polite term for argument) with the so-called Technical Architect who set all this up and have left it with him to figure out...

I am going to close this call now, but thanks for the pointers towards possible solutions. I will just use them as notes for myself of things to look for in the future.

Thanks

[Acxiom Unix Team]

The solution i have found is, i have left it with the Technical Architect who set it all up to figure out since he wont listen to me.......