Although I have the alias defined in /etc/aliases and /etc/postfix/aliases, I'm not receiving root mail. Following the previous thread about unreceived logwatch mail, I tested with a manual run of logwatch, and found that my ISP is rejecting the mail because it is seeing an envelope carrying my local address. My suspicion is that the mail is going out via sendmail instead of postfix-sendmail. IIRC there used to be system-switch-mail or a similarly named package. I can't find any such package - and can't remember which file it edited. Can someone please point me to that file?
Anne
On Monday 22 Aug 2011 12:49:21 Anne Wilson wrote:
Although I have the alias defined in /etc/aliases and /etc/postfix/aliases, I'm not receiving root mail. Following the previous thread about unreceived logwatch mail, I tested with a manual run of logwatch, and found that my ISP is rejecting the mail because it is seeing an envelope carrying my local address. My suspicion is that the mail is going out via sendmail instead of postfix-sendmail. IIRC there used to be system-switch-mail or a similarly named package. I can't find any such package - and can't remember which file it edited. Can someone please point me to that file?
The actual return report says:
anne@xxxx.org: host mailhost.zen.co.uk[212.23.3.98] said: 550-Verification failed for anne@xxx.lan 550-Unrouteable address 550 Envelope Sender: Domain must resolve in DNS! (in reply to RCPT TO command)
Anne
anne@xxxx.org: host mailhost.zen.co.uk[212.23.3.98] said: 550-Verification failed for anne@xxx.lan 550-Unrouteable address 550 Envelope Sender: Domain must resolve in DNS! (in reply to RCPT TO command)
It looks like your outgoing mail from your local user(s) needs to be masqueraded or the whole server does.
On Mon, Aug 22, 2011 at 8:07 AM, Barry Brimer lists@brimer.org wrote:
anne@xxxx.org: host mailhost.zen.co.uk[212.23.3.98] said: 550-Verification failed for anne@xxx.lan 550-Unrouteable address 550 Envelope Sender: Domain must resolve in DNS! (in reply to RCPT TO command)
It looks like your outgoing mail from your local user(s) needs to be masqueraded or the whole server does.
I think Barry has it right here.
Also, if you're using postfix, you should "yum remove sendmail" to avoid any possible conflicts. There's no reason to have sendmail, postfix implements its own 'sendmail' command.
-☙ Brian Mathis ❧-
On Mon, 2011-08-22 at 13:07 +0100, Anne Wilson wrote:
anne@xxxx.org: host mailhost.zen.co.uk[212.23.3.98] said: 550-Verification failed for anne@xxx.lan 550-Unrouteable address 550 Envelope Sender: Domain must resolve in DNS! (in reply to RCPT TO command)
Once your email is out on the Internet is needs a genuine Internet email address. xxx.lan is known only to your internal system and it is not an Internet email address.
Care to show the headers of your outgoing email ? (cover-up the bits you want to keep private)
On Mon, Aug 22, 2011 at 01:12:54PM +0100, Always Learning wrote:
On Mon, 2011-08-22 at 13:07 +0100, Anne Wilson wrote:
anne@xxxx.org: host mailhost.zen.co.uk[212.23.3.98] said: 550-Verification failed for anne@xxx.lan 550-Unrouteable address 550 Envelope Sender: Domain must resolve in DNS! (in reply to RCPT TO command)
Once your email is out on the Internet is needs a genuine Internet email address. xxx.lan is known only to your internal system and it is not an Internet email address.
Care to show the headers of your outgoing email ? (cover-up the bits you want to keep private)
If you're running postfix, you might want to also edit /etc/aliases. The last line or so has a commented line, person who should get root's mail. (The name is marc, probably the person who first wrote it.)
After changing that, run newaliases /etc/alias to rebuild the /etc/alias.db.
To answer your other question, alternatives is the command you asked about in an earlier post
alternatives --config mta
This will show you if you're using postfix or sendmail.
On 22/08/2011, Scott Robbins scottro@nyc.rr.com wrote:
On Mon, Aug 22, 2011 at 01:12:54PM +0100, Always Learning wrote:
On Mon, 2011-08-22 at 13:07 +0100, Anne Wilson wrote:
anne@xxxx.org: host mailhost.zen.co.uk[212.23.3.98] said:
If you're running postfix, you might want to also edit /etc/aliases. The last line or so has a commented line, person who should get root's mail. (The name is marc, probably the person who first wrote it.)
I did that, some time ago. Postfix.conf points to /etc/postfix/aliases, though, so I made sure by editing both.
After changing that, run newaliases /etc/alias to rebuild the /etc/alias.db.
Yes, I did remember that
To answer your other question, alternatives is the command you asked about in an earlier post
alternatives --config mta
This will show you if you're using postfix or sendmail.
I found postfix in the partial headers in the bounce-message
-- Scott Robbins PGP keyID EB3467D6 ( 1B48 077D 66F6 9DB0 FDC2 A409 FA54 EB34 67D6 ) gpg --keyserver pgp.mit.edu --recv-keys EB3467D6
Spike: Where have you been pet? Drusilla: I went for a walk. I met an old man. I didn't like him, he got stuck in my teeth. _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
On Monday 22 Aug 2011 13:12:54 Always Learning wrote:
Once your email is out on the Internet is needs a genuine Internet email address. xxx.lan is known only to your internal system and it is not an Internet email address.
Exactly - and my problem is knowing where it is getting this from.
Care to show the headers of your outgoing email ? (cover-up the bits you want to keep private)
Not sure where I can find the outgoing mail but I can give more info that might help.
It's a long story, but probably relevant, so -
We first set up a family LAN around 12 years ago, and called it xxx.net. Eventually I realised that xxx.net was actually a TLD name, so not a good idea. Actually I now own both xxx.org and xxx.net, but I decided that with this install I'd correct what had long been our practice.
On the new CentOS install I set the server name to borg.xxx.lan. I then changed every reference to xxx.net in /etc/hosts, and set about changing the Postfix config files. (I know now that it is using postfix.sendmail,) Somewhere I either have some other file still pointing to the old name, or, more likely, postfix is still using it somewhere.
I have been careful to re-map transports and restart postfix after any changes to the config files. The following from maillog may shed some light as to where the problem lies:
Aug 22 14:02:11 borg sendmail[1711]: p7MD29Lf001711: from=anne, size=6877, class=0, nrcpts=1, msgid=201108221302.p7MD29Lf001711@borg.xxx.lan, relay=root@localhost Aug 22 14:02:11 borg postfix/smtpd[2067]: connect from borg.xxx.net[127.0.0.1] Aug 22 14:02:11 borg postfix/smtpd[2067]: B4693A377C: client=borg.xxx.net[127.0.0.1] Aug 22 14:02:11 borg postfix/cleanup[2070]: B4693A377C: message- id=201108221302.p7MD29Lf001711@borg.xxx.lan Aug 22 14:02:11 borg postfix/qmgr[1499]: B4693A377C: from=anne@borg.xxx.lan, size=7487, nrcpt=1 (queue active) Aug 22 14:02:11 borg sendmail[1711]: p7MD29Lf001711: to=anne@xxx.org, ctladdr=anne (500/100), delay=00:00:02, xdelay=00:00:00, mailer=relay, pri=36877, relay=[127.0.0.1] [127.0.0.1], dsn=2.0.0, stat=Sent (Ok: queued as B4693A377C) Aug 22 14:02:11 borg postfix/smtpd[2067]: disconnect from borg.xxx.net[127.0.0.1] Aug 22 14:02:12 borg postfix/smtp[2071]: B4693A377C: to=anne@xxx.org, relay=mailhost.zen.co.uk[212.23.3.98]:25, delay=0.43, delays=0.09/0.02/0.15/0.18, dsn=5.0.0, status=bounced (host mailhost.zen.co.uk[212.23.3.98] said: 550-Verification failed for anne@borg.xxx.lan 550-Unrouteable address 550 Envelope Sender: Domain must resolve in DNS! (in reply to RCPT TO command)) Aug 22 14:02:12 borg postfix/cleanup[2070]: 2EB94A371B: message- id=20110822130212.2EB94A371B@borg Aug 22 14:02:12 borg postfix/bounce[2072]: B4693A377C: sender non-delivery notification: 2EB94A371B Aug 22 14:02:12 borg postfix/qmgr[1499]: 2EB94A371B: from=<>, size=9481, nrcpt=1 (queue active) Aug 22 14:02:12 borg postfix/qmgr[1499]: B4693A377C: removed Aug 22 14:02:12 borg postfix/smtp[2071]: 2EB94A371B: to=anne@borg.xxx.lan, relay=mailhost.zen.co.uk[212.23.3.98]:25, delay=0.75, delays=0.04/0/0.16/0.54, dsn=2.0.0, status=sent (250 OK id=1QvU8e-00085m-9o) Aug 22 14:02:12 borg postfix/qmgr[1499]: 2EB94A371B: removed
The old transport read:
xxx.net local: .xxx.net local: * smtp:[mailhost.zen.co.uk] .* smtp:[mailhost.zen.co.uk]
Those lines have been left and
xxx.lan local: .xxx.lan local:
added in.
A more likely suspect, I think, is main.cf - the relevant lines in the old one being
mydestination = $myhostname, localhost.$mydomain, $mydomain, xxx.lan myhostname = borg.xxx.net mydomain = xxx.net myorigin = $mydomain masquerade_domains = $mydomain #masquerade_domains = |borg.xxx.net mynetworks = 192.168.0.0/24,127.0.0.0/8
and the latest version (after many thrashing edits)
mydestination = $myhostname, localhost.$mydomain, $mydomain, xxx.lan myhostname = borg mydomain = xxx.org myorigin = $mydomain # masquerade_domains = $mydomain mynetworks = 192.168.0.0/24,127.0.0.0/8
Sorry this was so long, but it seemed important to give you as much info as I could.
Anne
On 22/08/2011 15:29, Anne Wilson wrote:
On Monday 22 Aug 2011 13:12:54 Always Learning wrote:
Once your email is out on the Internet is needs a genuine Internet email address. xxx.lan is known only to your internal system and it is not an Internet email address.
Exactly - and my problem is knowing where it is getting this from.
Care to show the headers of your outgoing email ? (cover-up the bits you want to keep private)
Not sure where I can find the outgoing mail but I can give more info that might help.
It's a long story, but probably relevant, so -
We first set up a family LAN around 12 years ago, and called it xxx.net. Eventually I realised that xxx.net was actually a TLD name, so not a good idea. Actually I now own both xxx.org and xxx.net, but I decided that with this install I'd correct what had long been our practice.
On the new CentOS install I set the server name to borg.xxx.lan. I then changed every reference to xxx.net in /etc/hosts, and set about changing the Postfix config files. (I know now that it is using postfix.sendmail,) Somewhere I either have some other file still pointing to the old name, or, more likely, postfix is still using it somewhere.
I have been careful to re-map transports and restart postfix after any changes to the config files. The following from maillog may shed some light as to where the problem lies:
Aug 22 14:02:11 borg sendmail[1711]: p7MD29Lf001711: from=anne, size=6877, class=0, nrcpts=1, msgid=201108221302.p7MD29Lf001711@borg.xxx.lan, relay=root@localhost Aug 22 14:02:11 borg postfix/smtpd[2067]: connect from borg.xxx.net[127.0.0.1] Aug 22 14:02:11 borg postfix/smtpd[2067]: B4693A377C: client=borg.xxx.net[127.0.0.1] Aug 22 14:02:11 borg postfix/cleanup[2070]: B4693A377C: message- id=201108221302.p7MD29Lf001711@borg.xxx.lan Aug 22 14:02:11 borg postfix/qmgr[1499]: B4693A377C: from=anne@borg.xxx.lan, size=7487, nrcpt=1 (queue active) Aug 22 14:02:11 borg sendmail[1711]: p7MD29Lf001711: to=anne@xxx.org, ctladdr=anne (500/100), delay=00:00:02, xdelay=00:00:00, mailer=relay, pri=36877, relay=[127.0.0.1] [127.0.0.1], dsn=2.0.0, stat=Sent (Ok: queued as B4693A377C) Aug 22 14:02:11 borg postfix/smtpd[2067]: disconnect from borg.xxx.net[127.0.0.1] Aug 22 14:02:12 borg postfix/smtp[2071]: B4693A377C: to=anne@xxx.org, relay=mailhost.zen.co.uk[212.23.3.98]:25, delay=0.43, delays=0.09/0.02/0.15/0.18, dsn=5.0.0, status=bounced (host mailhost.zen.co.uk[212.23.3.98] said: 550-Verification failed for anne@borg.xxx.lan 550-Unrouteable address 550 Envelope Sender: Domain must resolve in DNS! (in reply to RCPT TO command)) Aug 22 14:02:12 borg postfix/cleanup[2070]: 2EB94A371B: message- id=20110822130212.2EB94A371B@borg Aug 22 14:02:12 borg postfix/bounce[2072]: B4693A377C: sender non-delivery notification: 2EB94A371B Aug 22 14:02:12 borg postfix/qmgr[1499]: 2EB94A371B: from=<>, size=9481, nrcpt=1 (queue active) Aug 22 14:02:12 borg postfix/qmgr[1499]: B4693A377C: removed Aug 22 14:02:12 borg postfix/smtp[2071]: 2EB94A371B: to=anne@borg.xxx.lan, relay=mailhost.zen.co.uk[212.23.3.98]:25, delay=0.75, delays=0.04/0/0.16/0.54, dsn=2.0.0, status=sent (250 OK id=1QvU8e-00085m-9o) Aug 22 14:02:12 borg postfix/qmgr[1499]: 2EB94A371B: removed
The old transport read:
xxx.net local: .xxx.net local:
- smtp:[mailhost.zen.co.uk]
.* smtp:[mailhost.zen.co.uk]
Those lines have been left and
xxx.lan local: .xxx.lan local:
added in.
A more likely suspect, I think, is main.cf - the relevant lines in the old one being
mydestination = $myhostname, localhost.$mydomain, $mydomain, xxx.lan myhostname = borg.xxx.net mydomain = xxx.net myorigin = $mydomain masquerade_domains = $mydomain #masquerade_domains = |borg.xxx.net mynetworks = 192.168.0.0/24,127.0.0.0/8
and the latest version (after many thrashing edits)
mydestination = $myhostname, localhost.$mydomain, $mydomain, xxx.lan myhostname = borg mydomain = xxx.org myorigin = $mydomain # masquerade_domains = $mydomain mynetworks = 192.168.0.0/24,127.0.0.0/8
Sorry this was so long, but it seemed important to give you as much info as I could.
Anne _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
ls -l /usr/sbin/sendmail lrwxrwxrwx 1 root root 21 Aug 3 10:20 /usr/sbin/sendmail -> /etc/alternatives/mta
ls -l /etc/alternatives/mta lrwxrwxrwx 1 root root 27 Aug 3 10:20 /etc/alternatives/mta -> /usr/sbin/sendmail.sendmail
ls -l /usr/sbin/sendmail.sendmail -rwxr-sr-x 1 root smmsp 833512 Jun 17 2010 /usr/sbin/sendmail.sendmail
We fist should make sure you are using postfix instead of sendmail, could you please check that Anne?
Regards
Le lun 22 aoû 2011 14:29:29 CEST, Anne Wilson a écrit:
... Aug 22 14:02:12 borg postfix/smtp[2071]: B4693A377C: to=anne@xxx.org, relay=mailhost.zen.co.uk[212.23.3.98]:25, delay=0.43, delays=0.09/0.02/0.15/0.18, dsn=5.0.0, status=bounced (host mailhost.zen.co.uk[212.23.3.98] said: 550-Verification failed for anne@borg.xxx.lan 550-Unrouteable address 550 Envelope Sender: Domain must resolve in DNS! (in reply to RCPT TO command)) ...
I think you also need to rewrite your envelop address.
I had the same problem years ago, so the way to solve it has probably changed, hopefully not too much :
- edit (create if necessary) /etc/postfix/sender_cannonical anne@borg.xxx.lan some.address@some.real.server.for.the.bounces
- edit main.cf : sender_canonical_maps = hash:/etc/postfix/sender_cannonical
- rehash /etc/postfix/sender_cannonical and reload postfix
Good luck,
Actually, the "main suspect" is the program or person that is sending out mail with an unqualified sender, e.g. just "user". Change it to a qualified sender (e.g. with a resolvable FQDN).
Kai
On 8/22/2011 9:26 AM, Kai Schaetzl wrote:
Actually, the "main suspect" is the program or person that is sending out mail with an unqualified sender, e.g. just "user". Change it to a qualified sender (e.g. with a resolvable FQDN).
With sendmail you can set 'MASQUERADE_AS' in sendmail.mc to the FQDN that you would like to have added to unqualified senders. And if you want mail from root to be included in this treatment you have to change the default 'EXPOSED_USER' setting which assumes that root mail will be kept local or sent to an internal hub where you will want to see the real sending host.
It is obvious you run both Sendmail and Postfix, ending in trouble. The sendmail binary (which is a symlink) still points to Sendmail, though you seem to have configured Postfix as you system's MTA.
I have been careful to re-map transports and restart postfix after any changes to the config files. The following from maillog may shed some light as to where the problem lies:
Aug 22 14:02:11 borg sendmail[1711]: p7MD29Lf001711: from=anne, size=6877, class=0, nrcpts=1, msgid=201108221302.p7MD29Lf001711@borg.xxx.lan, relay=root@localhost Aug 22 14:02:11 borg postfix/smtpd[2067]: connect from borg.xxx.net[127.0.0.1] Aug 22 14:02:11 borg postfix/smtpd[2067]: B4693A377C: client=borg.xxx.net[127.0.0.1] Aug 22 14:02:11 borg postfix/cleanup[2070]: B4693A377C: message- id=201108221302.p7MD29Lf001711@borg.xxx.lan Aug 22 14:02:11 borg postfix/qmgr[1499]: B4693A377C: from=anne@borg.xxx.lan, size=7487, nrcpt=1 (queue active) Aug 22 14:02:11 borg sendmail[1711]: p7MD29Lf001711: to=anne@xxx.org, ctladdr=anne (500/100), delay=00:00:02, xdelay=00:00:00, mailer=relay, pri=36877, relay=[127.0.0.1] [127.0.0.1], dsn=2.0.0, stat=Sent (Ok: queued as B4693A377C) Aug 22 14:02:11 borg postfix/smtpd[2067]: disconnect from borg.xxx.net[127.0.0.1] Aug 22 14:02:12 borg postfix/smtp[2071]: B4693A377C: to=anne@xxx.org, relay=mailhost.zen.co.uk[212.23.3.98]:25, delay=0.43, delays=0.09/0.02/0.15/0.18, dsn=5.0.0, status=bounced (host mailhost.zen.co.uk[212.23.3.98] said: 550-Verification failed for anne@borg.xxx.lan 550-Unrouteable address 550 Envelope Sender: Domain must resolve in DNS! (in reply to RCPT TO command)) Aug 22 14:02:12 borg postfix/cleanup[2070]: 2EB94A371B: message- id=20110822130212.2EB94A371B@borg Aug 22 14:02:12 borg postfix/bounce[2072]: B4693A377C: sender non-delivery notification: 2EB94A371B Aug 22 14:02:12 borg postfix/qmgr[1499]: 2EB94A371B: from=<>, size=9481, nrcpt=1 (queue active) Aug 22 14:02:12 borg postfix/qmgr[1499]: B4693A377C: removed Aug 22 14:02:12 borg postfix/smtp[2071]: 2EB94A371B: to=anne@borg.xxx.lan, relay=mailhost.zen.co.uk[212.23.3.98]:25, delay=0.75, delays=0.04/0/0.16/0.54, dsn=2.0.0, status=sent (250 OK id=1QvU8e-00085m-9o) Aug 22 14:02:12 borg postfix/qmgr[1499]: 2EB94A371B: removed
Anne
alternatives --config mta
Switch to Postfix. Validate by
alternatives --display mta
then remove Sendmail if you do not use it.
Alexander
One of the first things I do after the installation of a system is a yum install postfix followed by a yum remove sendmail. No need to deal with alternatives if you don't intend to deal with sendmail anyway.
Regards. Dennis
On 08/22/2011 08:45 PM, Alexander Dalloz wrote:
It is obvious you run both Sendmail and Postfix, ending in trouble. The sendmail binary (which is a symlink) still points to Sendmail, though you seem to have configured Postfix as you system's MTA.
I have been careful to re-map transports and restart postfix after any changes to the config files. The following from maillog may shed some light as to where the problem lies:
Aug 22 14:02:11 borg sendmail[1711]: p7MD29Lf001711: from=anne, size=6877, class=0, nrcpts=1, msgid=201108221302.p7MD29Lf001711@borg.xxx.lan, relay=root@localhost Aug 22 14:02:11 borg postfix/smtpd[2067]: connect from borg.xxx.net[127.0.0.1] Aug 22 14:02:11 borg postfix/smtpd[2067]: B4693A377C: client=borg.xxx.net[127.0.0.1] Aug 22 14:02:11 borg postfix/cleanup[2070]: B4693A377C: message- id=201108221302.p7MD29Lf001711@borg.xxx.lan Aug 22 14:02:11 borg postfix/qmgr[1499]: B4693A377C: from=anne@borg.xxx.lan, size=7487, nrcpt=1 (queue active) Aug 22 14:02:11 borg sendmail[1711]: p7MD29Lf001711: to=anne@xxx.org, ctladdr=anne (500/100), delay=00:00:02, xdelay=00:00:00, mailer=relay, pri=36877, relay=[127.0.0.1] [127.0.0.1], dsn=2.0.0, stat=Sent (Ok: queued as B4693A377C) Aug 22 14:02:11 borg postfix/smtpd[2067]: disconnect from borg.xxx.net[127.0.0.1] Aug 22 14:02:12 borg postfix/smtp[2071]: B4693A377C: to=anne@xxx.org, relay=mailhost.zen.co.uk[212.23.3.98]:25, delay=0.43, delays=0.09/0.02/0.15/0.18, dsn=5.0.0, status=bounced (host mailhost.zen.co.uk[212.23.3.98] said: 550-Verification failed for anne@borg.xxx.lan 550-Unrouteable address 550 Envelope Sender: Domain must resolve in DNS! (in reply to RCPT TO command)) Aug 22 14:02:12 borg postfix/cleanup[2070]: 2EB94A371B: message- id=20110822130212.2EB94A371B@borg Aug 22 14:02:12 borg postfix/bounce[2072]: B4693A377C: sender non-delivery notification: 2EB94A371B Aug 22 14:02:12 borg postfix/qmgr[1499]: 2EB94A371B: from=<>, size=9481, nrcpt=1 (queue active) Aug 22 14:02:12 borg postfix/qmgr[1499]: B4693A377C: removed Aug 22 14:02:12 borg postfix/smtp[2071]: 2EB94A371B: to=anne@borg.xxx.lan, relay=mailhost.zen.co.uk[212.23.3.98]:25, delay=0.75, delays=0.04/0/0.16/0.54, dsn=2.0.0, status=sent (250 OK id=1QvU8e-00085m-9o) Aug 22 14:02:12 borg postfix/qmgr[1499]: 2EB94A371B: removed
Anne
alternatives --config mta
Switch to Postfix. Validate by
alternatives --display mta
then remove Sendmail if you do not use it.
Alexander
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
On 22 August 2011 20:50, Dennis Jacobfeuerborn dennisml@conversis.de wrote:
One of the first things I do after the installation of a system is a yum install postfix followed by a yum remove sendmail. No need to deal with alternatives if you don't intend to deal with sendmail anyway.
I removed sendmail - but it also removed clamav-milter. When I tried to re-install clamav-milter it would have pulled sendmail back in so I aborted. Advice?
In article CAJfU-f7nCz+iamXzg5rQ0VZAELctBCNcNFySYykAapXw5CeG6g@mail.gmail.com, Anne Wilson cannewilson@googlemail.com wrote:
On 22 August 2011 20:50, Dennis Jacobfeuerborn dennisml@conversis.de wrote:
One of the first things I do after the installation of a system is a yum install postfix followed by a yum remove sendmail. No need to deal with alternatives if you don't intend to deal with sendmail anyway.
I removed sendmail - but it also removed clamav-milter. When I tried to re-install clamav-milter it would have pulled sendmail back in so I aborted. Advice?
milters are specific to sendmail. clamav-milter is a linkage between ClamAV and sendmail, so you don't need it with postfix. If you want to use ClamAV, you will need to find out how to link it with postfix, but it won't be by using clamav-milter.
Cheers Tony
On 23/08/2011 11:46, Tony Mountifield wrote:
In article CAJfU-f7nCz+iamXzg5rQ0VZAELctBCNcNFySYykAapXw5CeG6g@mail.gmail.com, Anne Wilson cannewilson@googlemail.com wrote:
On 22 August 2011 20:50, Dennis Jacobfeuerborn dennisml@conversis.de wrote:
One of the first things I do after the installation of a system is a yum install postfix followed by a yum remove sendmail. No need to deal with alternatives if you don't intend to deal with sendmail anyway.
I removed sendmail - but it also removed clamav-milter. When I tried to re-install clamav-milter it would have pulled sendmail back in so I aborted. Advice?
milters are specific to sendmail. clamav-milter is a linkage between ClamAV and sendmail, so you don't need it with postfix. If you want to use ClamAV, you will need to find out how to link it with postfix, but it won't be by using clamav-milter.
Cheers Tony
You can get src.rpm files for clamav from here:
http://pkgs.repoforge.org/clamav/
Regards
On Tue, Aug 23, 2011 at 11:59:08AM +0200, Marc Deop i Argemí wrote:
On 23/08/2011 11:46, Tony Mountifield wrote:
Anne Wilson cannewilson@googlemail.com wrote:
I removed sendmail - but it also removed clamav-milter. When I tried to re-install clamav-milter it would have pulled sendmail back in so I aborted. Advice?
milters are specific to sendmail. clamav-milter is a linkage between ClamAV and sendmail, so you don't need it with postfix. If you want to use ClamAV, you will need to find out how to link it with postfix, but it won't be by using clamav-milter.
Cheers Tony
You can get src.rpm files for clamav from here:
Not even necessary to use the src.rpm. A simple yum install clamav will install without pulling sendmail. However, as Tony wrote, Anne will have to figure out how to use clamav with postfix.
On 23 August 2011 10:46, Tony Mountifield tony@softins.co.uk wrote:
In article CAJfU-f7nCz+iamXzg5rQ0VZAELctBCNcNFySYykAapXw5CeG6g@mail.gmail.com, Anne Wilson cannewilson@googlemail.com wrote:
On 22 August 2011 20:50, Dennis Jacobfeuerborn dennisml@conversis.de wrote:
One of the first things I do after the installation of a system is a yum install postfix followed by a yum remove sendmail. No need to deal with alternatives if you don't intend to deal with sendmail anyway.
I removed sendmail - but it also removed clamav-milter. When I tried to re-install clamav-milter it would have pulled sendmail back in so I aborted. Advice?
milters are specific to sendmail. clamav-milter is a linkage between ClamAV and sendmail, so you don't need it with postfix. If you want to use ClamAV, you will need to find out how to link it with postfix, but it won't be by using clamav-milter.
OK, that's fine, then. In the past I ran scans via cron - which I've set up again. However, I'm back with the original problem of not receiving root mail - and not receiving any local messages that should be routed by the /etc/postfix/transport.db
Anne
Anne Wilson wrote:
On 23 August 2011 10:46, Tony Mountifield tony@softins.co.uk wrote:
In article CAJfU-f7nCz+iamXzg5rQ0VZAELctBCNcNFySYykAapXw5CeG6g@mail.gmail.com, Anne Wilson cannewilson@googlemail.com wrote:
On 22 August 2011 20:50, Dennis Jacobfeuerborn dennisml@conversis.de wrote:
One of the first things I do after the installation of a system is a yum install postfix followed by a yum remove sendmail. No need to deal with alternatives if you don't intend to deal with sendmail anyway.
I removed sendmail - but it also removed clamav-milter. When I tried to re-install clamav-milter it would have pulled sendmail back in so I aborted. Advice?
milters are specific to sendmail. clamav-milter is a linkage between ClamAV and sendmail, so you don't need it with postfix. If you want to use ClamAV, you will need to find out how to link it with postfix, but it won't be by using clamav-milter.
OK, that's fine, then. In the past I ran scans via cron - which I've set up again. However, I'm back with the original problem of not receiving root mail - and not receiving any local messages that should be routed by the /etc/postfix/transport.db
I use amavisd and clamav/clamd from rf repo - instructions on setup are on the CentOS wiki - has always just worked for me with 5.x HTH
Anne _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
On 8/23/2011 4:46 AM, Tony Mountifield wrote:
I removed sendmail - but it also removed clamav-milter. When I tried to re-install clamav-milter it would have pulled sendmail back in so I aborted. Advice?
milters are specific to sendmail. clamav-milter is a linkage between ClamAV and sendmail, so you don't need it with postfix. If you want to use ClamAV, you will need to find out how to link it with postfix, but it won't be by using clamav-milter.
I only use sendmail with MimeDefang as a milter (which can coordinate everything else), but I thought that postfix has had milter support for a few years now.
In article CAJfU-f7nCz+iamXzg5rQ0VZAELctBCNcNFySYykAapXw5CeG6g@mail.gmail.com, Anne Wilson cannewilson@googlemail.com wrote:
I removed sendmail - but it also removed clamav-milter. When I tried to re-install clamav-milter it would have pulled sendmail back in so I aborted. Advice?
milters are specific to sendmail. clamav-milter is a linkage between ClamAV and sendmail, so you don't need it with postfix. If you want to use ClamAV, you will need to find out how to link it with postfix, but it won't be by using clamav-milter.
Cheers Tony
Sorry Tony, that's not correct. Though the MILTER interface (Send_M_ail f_ILTER) originally was an enhancement in Sendmail, meanwhile since quite some time Postfix supports milter applications as well.
http://www.postfix.org/MILTER_README.html
So of course Postfix users can implement milter based solutions. And in contrast to an AMaViS setup a milter based binding of ClamAV is lightweight.
Regards
Alexander
On Aug 23, 2011, at 9:41 AM, Alexander Dalloz wrote:
In article CAJfU-f7nCz+iamXzg5rQ0VZAELctBCNcNFySYykAapXw5CeG6g@mail.gmail.com, Anne Wilson cannewilson@googlemail.com wrote:
I removed sendmail - but it also removed clamav-milter. When I tried to re-install clamav-milter it would have pulled sendmail back in so I aborted. Advice?
milters are specific to sendmail. clamav-milter is a linkage between ClamAV and sendmail, so you don't need it with postfix. If you want to use ClamAV, you will need to find out how to link it with postfix, but it won't be by using clamav-milter.
Cheers Tony
Sorry Tony, that's not correct. Though the MILTER interface (Send_M_ail f_ILTER) originally was an enhancement in Sendmail, meanwhile since quite some time Postfix supports milter applications as well.
http://www.postfix.org/MILTER_README.html
So of course Postfix users can implement milter based solutions. And in contrast to an AMaViS setup a milter based binding of ClamAV is lightweight.
---- of course but unless the clamav-milter packager removes the dependency on sendmail, it's out with the bathwater.
Craig
In article 50117.213.157.4.156.1314117697.squirrel@nimrod.dscd.de, Alexander Dalloz ad+lists@uni-x.org wrote:
In article CAJfU-f7nCz+iamXzg5rQ0VZAELctBCNcNFySYykAapXw5CeG6g@mail.gmail.com, Anne Wilson cannewilson@googlemail.com wrote:
I removed sendmail - but it also removed clamav-milter. When I tried to re-install clamav-milter it would have pulled sendmail back in so I aborted. Advice?
milters are specific to sendmail. clamav-milter is a linkage between ClamAV and sendmail, so you don't need it with postfix. If you want to use ClamAV, you will need to find out how to link it with postfix, but it won't be by using clamav-milter.
Sorry Tony, that's not correct. Though the MILTER interface (Send_M_ail f_ILTER) originally was an enhancement in Sendmail, meanwhile since quite some time Postfix supports milter applications as well.
http://www.postfix.org/MILTER_README.html
So of course Postfix users can implement milter based solutions. And in contrast to an AMaViS setup a milter based binding of ClamAV is lightweight.
OK, thanks, I stand corrected, and educated! Must admit I'm not very familiar with postfix - only ever tried to do some debugging on it for someone else.
Of course, that begs the question of whether clamav-milter would work with postfix. If so, it should depend on (sendmail OR postfix). If not, perhaps it should be called clamav-milter-sendmail instead, with another package for clamav-milter-postfix.
Cheers Tony
On 22 August 2011 19:45, Alexander Dalloz ad+lists@uni-x.org wrote:
Anne
alternatives --config mta
Switch to Postfix. Validate by
alternatives --display mta
then remove Sendmail if you do not use it.
Thanks - that's what I was looking for.
alternatives --display mta mta - status is manual. link currently points to /usr/sbin/sendmail.postfix /usr/sbin/sendmail.postfix - priority 30 slave mta-pam: /etc/pam.d/smtp.postfix slave mta-mailq: /usr/bin/mailq.postfix slave mta-newaliases: /usr/bin/newaliases.postfix slave mta-rmail: /usr/bin/rmail.postfix slave mta-sendmail: /usr/lib/sendmail.postfix slave mta-mailqman: /usr/share/man/man1/mailq.postfix.1.gz slave mta-newaliasesman: /usr/share/man/man1/newaliases.postfix.1.gz slave mta-aliasesman: /usr/share/man/man5/aliases.postfix.5.gz slave mta-sendmailman: /usr/share/man/man1/sendmail.postfix.1.gz /usr/sbin/sendmail.sendmail - priority 90 slave mta-pam: /etc/pam.d/smtp.sendmail slave mta-mailq: /usr/bin/mailq.sendmail slave mta-newaliases: /usr/bin/newaliases.sendmail slave mta-rmail: /usr/bin/rmail.sendmail slave mta-sendmail: /usr/lib/sendmail.sendmail slave mta-mailqman: /usr/share/man/man1/mailq.sendmail.1.gz slave mta-newaliasesman: /usr/share/man/man1/newaliases.sendmail.1.gz slave mta-aliasesman: /usr/share/man/man5/aliases.sendmail.5.gz slave mta-sendmailman: /usr/share/man/man8/sendmail.sendmail.8.gz Current `best' version is /usr/sbin/sendmail.sendmail.
I assume this means that sendmail.postfix will be used as preference, but that last line confuses me. Is it safe to remove sendmail now, or does that imply that some things will still need it?
Anne
On 23 August 2011 07:47, Anne Wilson cannewilson@googlemail.com wrote:
On 22 August 2011 19:45, Alexander Dalloz ad+lists@uni-x.org wrote:
Anne
alternatives --config mta
Switch to Postfix. Validate by
alternatives --display mta
then remove Sendmail if you do not use it.
Thanks - that's what I was looking for.
alternatives --display mta mta - status is manual. link currently points to /usr/sbin/sendmail.postfix /usr/sbin/sendmail.postfix - priority 30 slave mta-pam: /etc/pam.d/smtp.postfix slave mta-mailq: /usr/bin/mailq.postfix slave mta-newaliases: /usr/bin/newaliases.postfix slave mta-rmail: /usr/bin/rmail.postfix slave mta-sendmail: /usr/lib/sendmail.postfix slave mta-mailqman: /usr/share/man/man1/mailq.postfix.1.gz slave mta-newaliasesman: /usr/share/man/man1/newaliases.postfix.1.gz slave mta-aliasesman: /usr/share/man/man5/aliases.postfix.5.gz slave mta-sendmailman: /usr/share/man/man1/sendmail.postfix.1.gz /usr/sbin/sendmail.sendmail - priority 90 slave mta-pam: /etc/pam.d/smtp.sendmail slave mta-mailq: /usr/bin/mailq.sendmail slave mta-newaliases: /usr/bin/newaliases.sendmail slave mta-rmail: /usr/bin/rmail.sendmail slave mta-sendmail: /usr/lib/sendmail.sendmail slave mta-mailqman: /usr/share/man/man1/mailq.sendmail.1.gz slave mta-newaliasesman: /usr/share/man/man1/newaliases.sendmail.1.gz slave mta-aliasesman: /usr/share/man/man5/aliases.sendmail.5.gz slave mta-sendmailman: /usr/share/man/man8/sendmail.sendmail.8.gz Current `best' version is /usr/sbin/sendmail.sendmail.
I assume this means that sendmail.postfix will be used as preference, but that last line confuses me. Is it safe to remove sendmail now, or does that imply that some things will still need it?
Anne
A large step nearer. Mail is now being received. Unfortunately I still can't send, so I need to find out why it isn't using the Transport settings.
Anne
On Mon, Aug 22, 2011 at 9:29 PM, Anne Wilson cannewilson@googlemail.com wrote:
Aug 22 14:02:11 borg postfix/qmgr[1499]: B4693A377C: from=anne@borg.xxx.lan, size=7487, nrcpt=1 (queue active) Aug 22 14:02:12 borg postfix/smtp[2071]: B4693A377C: to=anne@xxx.org, relay=mailhost.zen.co.uk[212.23.3.98]:25, delay=0.43,
mydestination = $myhostname, localhost.$mydomain, $mydomain, xxx.lan
Hi Anne :) Found you again from Mandriva list years ago :)
I think you need to put xxx.org also in mydestination. So the mail won't get through internet. As you can see, it's sending to anne@xxx.org.
mydestination = $myhostname, localhost.$mydomain, $mydomain, xxx.lan, xxx.org
On 23 August 2011 09:11, Fajar Priyanto fajarpri@arinet.org wrote:
On Mon, Aug 22, 2011 at 9:29 PM, Anne Wilson cannewilson@googlemail.com wrote:
Aug 22 14:02:11 borg postfix/qmgr[1499]: B4693A377C: from=anne@borg.xxx.lan, size=7487, nrcpt=1 (queue active) Aug 22 14:02:12 borg postfix/smtp[2071]: B4693A377C: to=anne@xxx.org, relay=mailhost.zen.co.uk[212.23.3.98]:25, delay=0.43,
mydestination = $myhostname, localhost.$mydomain, $mydomain, xxx.lan
Hi Anne :) Found you again from Mandriva list years ago :)
Hi. Good to talk to you a gain :-)
I think you need to put xxx.org also in mydestination. So the mail won't get through internet. As you can see, it's sending to anne@xxx.org.
mydestination = $myhostname, localhost.$mydomain, $mydomain, xxx.lan, xxx.org
This morning I restored the old main.cf, which uses the old network settings of xxx.net. Now mail is being received, but sending mail is being refused by my server. I have two sending profiles - one that sends to the server (CentOS 6), which then should use /etc/postfix/transport to separate local mail and external mail, sending the external mail to the ISP's smtp server. The second profile is googlemail smtp, used when I'm away from home.
Whichever I try to use I get a message that my mail server has refused the connection. I'm still trying to work out why. It sounds like authentication problem, but I can't see why it's happening.
Anne
On 8/23/11 3:51 AM, Anne Wilson wrote:
mydestination = $myhostname, localhost.$mydomain, $mydomain, xxx.lan, xxx.org
This morning I restored the old main.cf, which uses the old network settings of xxx.net. Now mail is being received, but sending mail is being refused by my server. I have two sending profiles - one that sends to the server (CentOS 6), which then should use /etc/postfix/transport to separate local mail and external mail, sending the external mail to the ISP's smtp server. The second profile is googlemail smtp, used when I'm away from home.
Whichever I try to use I get a message that my mail server has refused the connection. I'm still trying to work out why. It sounds like authentication problem, but I can't see why it's happening.
Unless the next hop requires smtp authentication, the likely reason is that the 'From: ' address doesn't have a DNS-resolvable domain name.
On Tuesday 23 Aug 2011 Les Mikesell wrote:
On 8/23/11 3:51 AM, Anne Wilson wrote:
mydestination = $myhostname, localhost.$mydomain, $mydomain, xxx.lan, xxx.org
This morning I restored the old main.cf, which uses the old network settings of xxx.net. Now mail is being received, but sending mail is being refused by my server. I have two sending profiles - one that sends to the server (CentOS 6), which then should use /etc/postfix/transport to separate local mail and external mail, sending the external mail to the ISP's smtp server. The second profile is googlemail smtp, used when I'm away from home.
Whichever I try to use I get a message that my mail server has refused the connection. I'm still trying to work out why. It sounds like authentication problem, but I can't see why it's happening.
Unless the next hop requires smtp authentication, the likely reason is that the 'From: ' address doesn't have a DNS-resolvable domain name.
Hi, Les. Another "old friend" :-)
FWIW, I found that the right combination in main.cf was to set myhost to the local name, borg.xxx.lan and mydomain to one of my owned domains, xxx.net, as it was on the old server. Adding a couple of lines to transport (re-hashed, of course) now gives me internal mail when addressed to anne@xxx.net or anne@xxx.lan, and external mail when addressed any other way. Perfect.
Thanks to all who tried to help.
Anne
The computer is an HP Proliant DC7600S. O/S: Centos 5.7
I try to connect minicom or Hylafax to /dev/ttyS0 and I can't reach the modem.
Can somebody had this problem and find a solution?
--- Michel Donais
----- Original Message ----- | The computer is an HP Proliant DC7600S. | O/S: Centos 5.7 | | I try to connect minicom or Hylafax to /dev/ttyS0 and I can't reach | the | modem. | | Can somebody had this problem and find a solution? | | | --- | Michel Donais | | _______________________________________________ | CentOS mailing list | CentOS@centos.org | http://lists.centos.org/mailman/listinfo/centos
It's a permissions issue. Either add the users in question to the group owning /dev/ttyS0 or you need to create a device override script which changes the ownership of the file each time the system boots.
On Sat, 24 Sep 2011, Michel Donais wrote:
To: CentOS mailing list centos@centos.org From: Michel Donais donais@telupton.com Subject: [CentOS] I can't connect to /dev/ttyS0
The computer is an HP Proliant DC7600S. O/S: Centos 5.7
I try to connect minicom or Hylafax to /dev/ttyS0 and I can't reach the modem.
Can somebody had this problem and find a solution?
Hello Michel.
You first need to double-check that /dev/ttyS0 is actually the correct serial COM port you have the modem connected to.
I've know the COM port numbers to mysteriously change for some reason. My APC UPS was connected and working on /dev/ttyS1 and it stopped working recently. The serial COM port has changed for some reason to /dev/ttyS0. So I had to edit the config file and stop and restart apcupsd, and it's working fine again now.
Please look under System->Hardware and then click the 16550A-compatible COM port in the LH pane.
Then in the RH pane click on 'Advanced', and find the serial.device string or linux.device_file.
This will be something like:
linux.device_file strlist /dev/ttySx where x == 0|1|2|3 serial.device strlist /dev/ttySx where x == 0|1|2|3
Make sure this is the same as what you are trying to connect to.
Here's a screenshot of Device Manager showing my COM port settings:
http://oi56.tinypic.com/2a7h1sp.jpg
If nothing shows under Device Manager settings, please check you have your serial COM ports turned on in your BIOS settings at reboot time :)
Kind Regards,
Keith Roberts
----------------------------------------------------------------- Websites: http://www.karsites.net http://www.php-debuggers.net http://www.raised-from-the-dead.org.uk
All email addresses are challenge-response protected with TMDA [http://tmda.net] -----------------------------------------------------------------
You first need to double-check that /dev/ttyS0 is actually the correct serial COM port you have the modem connected to.
Please look under System->Hardware and then click the 16550A-compatible COM port in the LH pane.
Then in the RH pane click on 'Advanced', and find the serial.device string or linux.device_file.
This will be something like:
linux.device_file strlist /dev/ttySx where x == 0|1|2|3 serial.device strlist /dev/ttySx where x == 0|1|2|3
Make sure this is the same as what you are trying to connect to.
Thank's Keith
I got it working
--- Michel Donais
On Sun, 25 Sep 2011, Michel Donais wrote:
To: CentOS mailing list centos@centos.org From: Michel Donais donais@telupton.com Subject: Re: [CentOS] I can't connect to /dev/ttyS0
You first need to double-check that /dev/ttyS0 is actually the correct serial COM port you have the modem connected to.
Please look under System->Hardware and then click the 16550A-compatible COM port in the LH pane.
Then in the RH pane click on 'Advanced', and find the serial.device string or linux.device_file.
This will be something like:
linux.device_file strlist /dev/ttySx where x == 0|1|2|3 serial.device strlist /dev/ttySx where x == 0|1|2|3
Make sure this is the same as what you are trying to connect to.
Thank's Keith
I got it working
Hi Michael.
Please can you share with us what the reason was for the problem, so others with similar problems can be enlighted, including myself?
Kind Regards,
Keith
----------------------------------------------------------------- Websites: http://www.karsites.net http://www.php-debuggers.net http://www.raised-from-the-dead.org.uk
All email addresses are challenge-response protected with TMDA [http://tmda.net] -----------------------------------------------------------------
You first need to double-check that /dev/ttyS0 is actually the correct serial COM port you have the modem connected to.
Please look under System->Hardware and then click the 16550A-compatible COM port in the LH pane.
Then in the RH pane click on 'Advanced', and find the serial.device string or linux.device_file.
This will be something like: linux.device_file strlist /dev/ttySx where x == 0|1|2|3 serial.device strlist /dev/ttySx where x == 0|1|2|3
Please can you share with us what the reason was for the problem, so others with similar problems can be enlighted, including myself?
Well it's a bit ennoying. I've done all the given check list and got nearly no answer for a solution. So I decided to open the box to see that the serial device connector was partly unplug from the mother board; so it was the source of my problem.
I put it back in palce and it was done.
--- Michel Donais
On Sun, 25 Sep 2011, Michel Donais wrote:
To: CentOS mailing list centos@centos.org From: Michel Donais donais@telupton.com Subject: Re: [CentOS] I can't connect to /dev/ttyS0
You first need to double-check that /dev/ttyS0 is actually the correct serial COM port you have the modem connected to.
Please look under System->Hardware and then click the 16550A-compatible COM port in the LH pane.
Then in the RH pane click on 'Advanced', and find the serial.device string or linux.device_file.
This will be something like: linux.device_file strlist /dev/ttySx where x == 0|1|2|3 serial.device strlist /dev/ttySx where x == 0|1|2|3
Please can you share with us what the reason was for the problem, so others with similar problems can be enlighted, including myself?
Well it's a bit ennoying. I've done all the given check list and got nearly no answer for a solution. So I decided to open the box to see that the serial device connector was partly unplug from the mother board; so it was the source of my problem.
I put it back in palce and it was done.
Ah - Thanks for that Michel - very informative. That's probably one of the last tings I would have taken a look at.
So it was actually a hardware problem - not a software one.
I'm Very pleased you were able to identify and fix the problem :)
Kind Regards,
Keith Roberts
----------------------------------------------------------------- Websites: http://www.karsites.net http://www.php-debuggers.net http://www.raised-from-the-dead.org.uk
All email addresses are challenge-response protected with TMDA [http://tmda.net] -----------------------------------------------------------------
-
Alexander Dalloz -
Always Learning -
Anne Wilson -
Barry Brimer -
Brian Mathis -
Craig White -
Dennis Jacobfeuerborn -
Fajar Priyanto -
James A. Peltier -
Kai Schaetzl -
Keith Roberts -
Les Mikesell -
Marc Deop i Argemí -
Michel Donais -
Philippe Naudin -
Rob Kampen -
Scott Robbins -
tony＠softins.co.uk