I seem to have an issue with outgoing mail/sending mail, I've created a test user mailbox and I can connect to this using Outlook however then outgoing mail sending test fails.
In my /var/log/maillog it shows the incoming pop connection that appears ok however I then have a couple of warning lines see..
Aug 28 10:38:13 46 dovecot: pop3-login: Login: user=<test.generic>, method=PLAIN, rip=::ffff:80.176.130.114, lip=::ffff:46.38.189.235 Aug 28 10:38:13 46 dovecot: POP3(test.generic): Disconnected: Logged out top=0/0, retr=0/0, del=0/0, size=0 Aug 28 10:38:36 46 postfix/local[3213]: fatal: open database /etc/aliases.db: No such file or directory Aug 28 10:38:37 46 postfix/master[9206]: warning: process /usr/libexec/postfix/local pid 3213 exit status 1 Aug 28 10:38:37 46 postfix/master[9206]: warning: /usr/libexec/postfix/local: bad command startup -- throttling Aug 28 10:39:37 46 postfix/local[3242]: fatal: open database /etc/aliases.db: No such file or directory Aug 28 10:39:38 46 postfix/master[9206]: warning: process /usr/libexec/postfix/local pid 3242 exit status 1 Aug 28 10:39:38 46 postfix/master[9206]: warning: /usr/libexec/postfix/local: bad command startup -- throttling
Can anyone advise what is wrong/how to fix?
Ah, this could be an issue:
fatal: open database /etc/aliases.db: No such file or directory
To fix that, try running this command:
newaliases
Once you do that, restart Postfix with "/etc/init.d/postfix restart", and then check to see if any warnings show up.
If not, see if your email begins working as expected at that point.
-Eric
Hi Eric,
Thanks for your reply, I ran the commands however Outlook still cannot send, is says it cannot connect to the outgoing SMTP server, I am using the same server setting as for ncming would that be correct?
also told the system to resend the sign-up email however nothing has arrived, below is a futher extract from the maillog, does this give any lues as to what is wrong?
Aug 28 17:49:11 46 postfix/local[22407]: EDDA91C33B: to=root@generic.net-digital.co.uk, orig_to=, relay=local, delay=49609, delays=36060/13542/0/6.8, dsn=2.0.0, status=sent (delivered to command: /usr/bin/procmail-wrapper -o -a $DOMAIN -d $LOGNAME) Aug 28 17:49:11 46 postfix/qmgr[10779]: EDDA91C33B: removed Aug 28 17:49:25 46 postfix/postfix-script: stopping the Postfix mail system Aug 28 17:49:25 46 postfix/master[9206]: terminating on signal 15 Aug 28 17:49:26 46 postfix/postfix-script: starting the Postfix mail system Aug 28 17:49:26 46 postfix/master[22476]: daemon started -- version 2.3.3, configuration /etc/postfix Aug 28 17:51:30 46 dovecot: pop3-login: Login: user=<test.generic>, method=PLAIN, rip=::ffff:80.176.130.114, lip=::ffff:46.38.189.235 Aug 28 17:51:31 46 dovecot: POP3(test.generic): Disconnected: Logged out top=0/0, retr=0/0, del=0/0, size=0 Aug 28 17:55:24 46 postfix/pickup[22478]: 68AE21C34F: uid=0 from=<webmin@46.38.189.235> Aug 28 17:55:24 46 postfix/cleanup[22794]: 68AE21C34F: message-id=<1314554123.22787.1@46.38.189.235> Aug 28 17:55:24 46 postfix/qmgr[22479]: 68AE21C34F: from=<webmin@46.38.189.235>, size=1269, nrcpt=1 (queue active) Aug 28 17:55:26 46 postfix/smtp[22796]: 68AE21C34F: to=chris@net-digital.co.uk, relay=mx81.emailfiltering.com[194.116.198.78]:25, delay=2.4, delays=0.58/0.1/1.8/0, dsn=5.0.0, status=bounced (host mx81.emailfiltering.com[194.116.198.78] said: 553 Syntax error (4.2.1): <webmin@46.38.189.235> (in reply to MAIL FROM command)) Aug 28 17:55:26 46 postfix/cleanup[22794]: 8C3DE1C35F: message-id=20110828175526.8C3DE1C35F@generic.net-digital.co.uk Aug 28 17:55:26 46 postfix/qmgr[22479]: 8C3DE1C35F: from=<>, size=3241, nrcpt=1 (queue active) Aug 28 17:55:26 46 postfix/bounce[22797]: 68AE21C34F: sender non-delivery notification: 8C3DE1C35F Aug 28 17:55:26 46 postfix/qmgr[22479]: 8C3DE1C35F: to=<webmin@46.38.189.235>, relay=none, delay=0, delays=0/0/0/0, dsn=5.1.3, status=bounced (bad address syntax) Aug 28 17:55:26 46 postfix/qmgr[22479]: 68AE21C34F: removed Aug 28 17:55:26 46 postfix/qmgr[22479]: 8C3DE1C35F: removed
Well, incoming and outgoing email are handled differently. Incoming is POP/IMAP, handled via Dovecot. Outgoing is connecting to Postfix via SMTP.
First, you may want to make sure Postfix is listening on port 25. You can do that with:
netstat -an | grep :25
If that works -- are you sure your ISP isn't blocking port 25?
-Eric
With that command I get:
[root@46 ~]# netstat -an | grep :25 tcp 0 0 127.0.0.1:25 0.0.0.0:* LISTEN tcp 0 0 ::ffff:46.38.189.235:22 ::ffff:80.176.130.114:25963 ESTABLISHED
Says establised so I guess this is working?
Further to my above post regarding the sending mail aspect of my issue I sent the sign-up email again to a gmail address and this time it did come throgh however I notice that the from address is webmin@46.38.189.235 as it isn't a proper domain which I am guessing is not correct, how do I set/control this?
Further to the above do I need to configure postfix or should it work out of the box? I checked with my ISP and they said "It looks like port 25 is unfiltered correctly, but nothing is listening on it."
I think there is something wrong with my postfix configuration as when I try and save the General Options in the postfix module I get the error message...
Failed to save options : Failed to query Postfix config command to get the current value of parameter myhostname: /usr/sbin/postconf: warning: valid_hostname: numeric hostname: 46.38.189.235 /usr/sbin/postconf: fatal: unable to use my own hostname
I've tried to view the /usr/sbin/postconf using FTP to edit and change the IP address to a hostname however the file looks encoded/encrypted i.e. not in plain text so I cannot edit?
It looks like it's not liking your hostname... what is that currently set to? You can determine that by running the "hostname" command.
-Eric
Says...
[root@46 ~]# hostname 46.38.189.235
How/where do I change this?
I think I've managed to change the hostname, first I changed in Webmin - Networking - Network Configuration - Hostname and DNS Client and use the field: Hostname
The I changed in Network interfaces for receiving mail within the Postfix module and now it's not complaining about the IP anymore however I still have the outgoing mailserver issue where Outlook cannot connect. As said above my ISP has confirmed that "It looks like port 25 is unfiltered correctly, but nothing is listening on it."
My maillog says...
Aug 29 17:07:16 46 postfix/master[22476]: warning: service smtp: ignoring inet_interfaces change Aug 29 17:07:16 46 postfix/master[22476]: warning: to change inet_interfaces, stop and start Postfix
Can anyone advise me how to fix and get Postfix listening?
OK making some progress I think I restarted Postfix and tried another send and receive this time I got a different message from Outlook which said none of the authentication methods supported by this client are supported by your server if I use outgoing mail authentication with the outgoing credentials and if I don't use outgoing mail authentication I get a message the server responded 450 4.2.0 , method=PLAIN, rip=::ffff:80.176.130.114, lip=::ffff:46.38.189.235
Aug 29 17:19:39 46 dovecot: POP3(test.generic): Disconnected: Logged out top=0/0, retr=0/0, del=0/1, size=981
Aug 29 17:19:39 46 postfix/smtpd[32224]: connect from demon.co.uk[80.176.130.114]
Aug 29 17:19:39 46 postfix/smtpd[32224]: lost connection after EHLO from demon.co.uk[80.176.130.114]
Aug 29 17:19:39 46 postfix/smtpd[32224]: disconnect from netdigital.demon.co.uk[80.176.130.114]
Aug 29 17:24:03 46 postfix/smtpd[32347]: NOQUEUE: reject: RCPT from netdigital.demon.co.uk[80.176.130.114]: 450 4.2.0 : Recipient address rejected: Greylisted, see http://postgrey.schweikert.ch/help/generic.net-digital.co.uk.html; from= to= proto=ESMTP helo=
Aug 29 17:24:03 46 postfix/smtpd[32347]: lost connection after RCPT from demon.co.uk[80.176.130.114]
Aug 29 17:24:03 46 postfix/smtpd[32347]: disconnect from demon.co.uk[80.176.130.114]
Aug 29 17:27:00 46 dovecot: pop3-login: Login: user=, method=PLAIN, rip=::ffff:80.176.130.114, lip=::ffff:46.38.189.235
Aug 29 17:27:01 46 dovecot: POP3(test.generic): Disconnected: Logged out top=0/0, retr=1/997, del=0/1, size=981
Aug 29 17:27:23 46 postfix/anvil[32349]: statistics: max connection rate 3/60s for (smtp:80.176.130.114) at Aug 29 17:24:02
Aug 29 17:27:23 46 postfix/anvil[32349]: statistics: max connection count 1 for (smtp:80.176.130.114) at Aug 29 17:23:34
Aug 29 17:27:23 46 postfix/anvil[32349]: statistics: max cache size 1 at Aug 29 17:23:34
As there was a message relating to grey listing I turned this off and tried to send again outlook didn't complain however no mail sent, below is my maillog now, any ideas whats wrong/how to fix?
Aug 29 20:10:44 46 postfix/smtpd[2828]: connect from netdigital.demon.co.uk[80.176.130.114]
Aug 29 20:10:44 46 postfix/trivial-rewrite[2830]: warning: do not list domain generic.net-digital.co.uk in BOTH mydestination and virtual_alias_domains
Aug 29 20:10:44 46 postfix/smtpd[2828]: NOQUEUE: reject: RCPT from demon.co.uk[80.176.130.114]: 554 5.7.1 : Relay access denied; from= to= proto=ESMTP helo=
OK I'm stuck now, as you can see from above I've been trying to work through this to solve the issue however I don't know what to do now, please can someone assist with the above?
Howdy,
That relay access denied error may mean that Outlook isn't configured to authenticate when sending an outgoing email. That's the Outlook default, to not authenticate.
However, since your server can't distinguish between legitimate users and a spammer, authentication is required... what you can do is go into the Outlook settings, and set it up to authenticate during Outgoing SMTP connections.
-Eric
Actually, weird changed nothing more and it seems to be working now, without outgoing authentication turned on in Outlook and if I actually tell Outllok to authnticate it stops working. Is there a setting I can use to require outgoing mail authntication?
If you're sending an email to an account that's local to the server, it would not require authentication. It only requires authentication for relaying an email to another server.
-Eric
Ok thanks however why would it fail to connect/send when I tell Outlook to use authentication (same credentials as pop user login) ?
What error do you get exactly? And what shows up in the mail logs when that happens?
-Eric
It says sending reported error (0x800CCC80): None of the authentication methods supported by this client are support by your server.
this is the maillog...
Aug 30 23:32:04 46 postfix/smtpd[19998]: connect from demon.co.uk[80.176.130.114]
Aug 30 23:32:04 46 postfix/smtpd[19998]: lost connection after EHLO from demon.co.uk[80.176.130.114]
Aug 30 23:32:04 46 postfix/smtpd[19998]: disconnect from demon.co.uk[80.176.130.114]
(note if I try and send email to a remote address without using authentication I get the error message - Server error: '554 5.7.1: Relay access denied'
postfix/smtpd[19998]: NOQUEUE: reject: RCPT from demon.co.uk[80.176.130.114]: 554 5.7.1 to= proto=ESMTP helo=
I really need to get this issue solved now, can anyone help is there some additional configuration that needs to be done if so are there any guidance notes?
What port do you have Outlook setup to connect on when sending an email when authentication is setup?
Also, what do these two commands show:
netstat -an | grep 465
netstat -an | grep 587
I'm using the standard port 25 for SMTP, the two commands didn't return anything...
[root@babyandtoddlerworld ~]# netstat -an | grep 465
[root@babyandtoddlerworld ~]# netstat -an | grep 587
[root@babyandtoddlerworld ~]#
?
It doesn't sound like Outlook is happy with the authentication methods available by default on port 25.
We can enable some additional options -- what distribution is it that you're using?
-Eric
Webmin version 1.561 Virtualmin version 3.87.gpl GPL Operating system CentOS Linux 5.6
Port 587 (Submission) should be enabled by default on a new CentOS install... you may want to make sure that saslauthd is running. You can do that by running this command:
/etc/init.d/saslauthd restart
After that, do you see port 587 in use:
netstat -an | grep 587
OK ran the commands got...
[root@babyandtoddlerworld ~]# /etc/init.d/saslauthd restart
Stopping saslauthd: [ OK ]
Starting saslauthd: [ OK ]
[root@babyandtoddlerworld ~]# netstat -an | grep 587
[root@babyandtoddlerworld ~]#
Please advise?
Hi Eric,
Any news on how I enable some additional options?
Can anyone help me with this?
I really need to get some help with this to finally resolve the problem can anyone advise/assist?
Do you have a Virtual Server that you can enable SSL for?
You can use that to get "Submission" on port 587 up and running.
So, once you pick a Virtual Server to enable SSL -- go into Edit Virtual Server -> Enabled Features, and enable "SSL Website Enabled".
Then, go into Server Configuration -> Manage SSL Certificate, and click "Copy to Postfix".
Just to be sure that took properly, restart Postfix by running "/etc/init.d/postfix restart".
After you do that, what does this command show:
netstat -an | grep 587
I have one Virtual Server set-up, does it have to run with SSL as we don't have an SSL certificate for the domain concerned?
That's okay, it'll setup a self-signed certificate for you, you don't need to purchase an SSL certificate.
-Eric
OK done this is what happened after running the commands...
[root@babyandtoddlerworld ~]# /etc/init.d/postfix restart
Shutting down postfix: [ OK ]
Starting postfix: [ OK ]
[root@babyandtoddlerworld ~]# netstat -an | grep 587
unix 3 [ ] STREAM CONNECTED 1044587
[root@babyandtoddlerworld ~]#
Now, you'll need to edit /etc/postfix/master.cf, find the sections relating to SMTPS and Submission, and uncomment those lines. After doing that, restart Postfix, and try running that "netstat" command again.
-Eric
I've just search /etc/postfix/master.cf for 'SMTPS' however there is nothing found?
The line is likely all lower case. If you searched on upper-case only, the search wouldn't likely see it.
-Eric
Nope sorry searched in lowercase for 'smtps' as well and still not found?
Well, let's just ignore smtps for now... do you see "submission" in there?
-Eric
I don't think so the only line in the file that mentions the word submission is my line 643 which is a commented line...
# setgid_group: The group for mail submission and queue management
?
Hrm, that's unusual, those entries should both be in everything master.cf file by default.
Can you attach your /etc/postfix/master.cf file to the thread here?
-Eric
Attached as a text file
That's your main.cf file, the Submission/SMTPS stuff we're talking about would be in your master.cf file :-)
-Eric
Doh sorry!
OK uncommented...
submission inet n - n - - smtpd
The line
smtps inet n - n - - smtpd
is already uncommented
[root@babyandtoddlerworld ~]# /etc/init.d/postfix restart
Shutting down postfix: [ OK ]
Starting postfix: [ OK ]
[root@babyandtoddlerworld ~]# netstat -an | grep 587
tcp 0 0 46.38.189.235:587 0.0.0.0:* LISTEN
[root@babyandtoddlerworld ~]#
Okay, that's good!
Now, configure Outlook so that when it connects, it connects to port 587, and uses SSL or TLS. If port 587 doesn't work, try port 465.
-Eric
It's still not liking it tried both SSL & TSL ports 587 & 465 and each time I get the same message…
'Outlook cannot connect to the Outgoing SMTP email server'
The maillog on the server is only showing the POP connects and disconnects not SMTP attempts?
Whenever I try to connect to port 587 on your IP address, I receive a "No route to host" error.
Are you sure there isn't a router or firewall in front of your server blocking that port?
-Eric
Got in contact with ISP and they advised me to add..
-A RH-Firewall-1-INPUT -m tcp -p tcp --dport 587 -j ACCEPT
-A RH-Firewall-1-INPUT -m tcp -p tcp --dport 465 -j ACCEPT
into the IP tables and restart the service.
Now seems to connect on port 465 with SSL, just doing some sending tests, just noticed a couple of things in the maillog, a warning and another line saying removed, are these issues I need to sort, if so how?
Sep 6 16:28:04 babyandtoddlerworld postfix/trivial-rewrite[9062]: warning: do not list domain babyandtoddlerworld.co.uk in BOTH mydestination and virtual_alias_domains
Sep 6 16:28:07 babyandtoddlerworld postfix/smtp[9064]: AC2A4130602: to=, relay=gmail-smtp-in.l.google.com[209.85.227.27]:25, delay=2.6, delays=0.39/0.03/0.4/1.8, dsn=2.0.0, status=sent (250 2.0.0 OK 1315322877 et19si8816166wbb.3)
Sep 6 16:28:07 babyandtoddlerworld postfix/qmgr[13177]: AC2A4130602: removed
Ah, so that means you're running a firewall on your own server.
That's not the default... so that means anytime you're trying to run a new service on your system, you'll need to add a line similar to what you did above for that new service.
As far as your mail log entries go -- those are all okay. The warning is safe to ignore, and the other entries are typical entries that you'd see in the mail log.
-Eric
OK thanks, I'm nearly there now I think, sending seems to be working after a couple of test, just trying to work out what's going on with the incoming mail seems to be hitting the server however not arriving in the mailbox for collection see this extract from the maillog below, what is wrong?
Sep 6 16:56:17 babyandtoddlerworld spamd[1214]: spamd: clean message (3.6/5.0) for marketing.btw:502 in 0.5 seconds, 12159 bytes.
Sep 6 16:56:17 babyandtoddlerworld spamd[1214]: spamd: result: . 3 - EXTRA_MPART_TYPE,HTML_MESSAGE,RCVD_NUMERIC_HELO scantime=0.5,size=12159,user=marketing.btw,uid=502,required_score=5.0,rhost=localhost,raddr=127.0.0.1,rport=59924,mid=,autolearn=no
Sep 6 16:56:17 babyandtoddlerworld postfix/local[9943]: D7AA3130600: to=, orig_to=, relay=local, delay=1.1, delays=0.06/0.01/0/1.1, dsn=2.0.0, status=sent (delivered to command: /usr/bin/procmail-wrapper -o -a $DOMAIN -d $LOGNAME)
Sep 6 16:56:17 babyandtoddlerworld postfix/qmgr[13177]: D7AA3130600: removed
Sep 6 16:56:17 babyandtoddlerworld spamd[1143]: prefork: child states: II
Well, it looks like it's being delivered... the next step would be to look at /var/log/procmail.log to see where exactly procmail is delivering it.
-Eric
Pages