> Please review the Postfix 2.6 RELEASE_NOTES file, in particular
> the section titled "Major changes - milter support".
>
"milter_protocol = 2" in main.cf." did the trick... Thank you.
I have just upgraded from postfix 2.3.3 (CentOS 5.3 stock package) to
2.6.5-1. The process went quite smoothly and everything seems good,
except I just noticed a log entry pertaining to failed delivery for a
virtual user. I am not familiar enough with milters to know if this
indicates a problem or
> Simon Mudd is the man who makes these RPMs. We mirror his site. You can get
> the RPMs here:
> ftp://mirror.newnanutilities.org/pub/postfix-rpm/official/2.6/
That's great, I'd been googl(dir)ected to Simon's site, could not
locate recent releases without your help...
Does anyone maintain a repo for CentOS/RHE packages for some of the
later builds? On Cent 5.3 we're still looking at postfix 3.3, and I
now see there are some fair advantages to an upgrade. I don't really
want to muddle through creating a spec file if someone has been down
this road.
- Andrew
>> Than what is the preferred method to verify that a transport is being used
>> for the domain it configured for...
>
> Generally, you don't need to verify this. If the transport table is defined
> it is used as advertised. If you absolutely must check:
>
> - Send mail to two users that should
On Thu, Sep 17, 2009 at 2:02 PM, Victor Duchovni
wrote:
> On Thu, Sep 17, 2009 at 01:59:25PM -0400, fursink wrote:
>
>> Would the fact that this is 2.3.3... prevent the transport from working
>> at all?
>
> No, and in fact the transport is almost certainly used, but &qu
On Thu, Sep 17, 2009 at 1:48 PM, Victor Duchovni
wrote:
> On Thu, Sep 17, 2009 at 01:27:45PM -0400, fursink wrote:
>
>> Sep 17 13:25:55 tmail postfix/smtp[27618]: EBC6F87D49:
>> to=, relay=e.mx.mail.yahoo.com[216.39.53.1]:25,
>> delay=3.7, delays=0.02/0/0.64/3, dsn=2.
On Thu, Sep 17, 2009 at 1:17 PM, /dev/rob0 wrote:
> On Thursday 17 September 2009 12:11:22 AndrewLong wrote:
>> Wietse Venema wrote:
>> > You forgot the transport table configuration. Without this,
>> > Postfix will never use the "slow" transport.
>
>> /etc/postfix/transport::
>> yahoo.com s
I apologize if this topic seems basic or I'm missing something obvious
here, but after much googling and looking to Ralph's book for answers,
I am still unclear on this...
Despite creating a slow transport for yahoo and verifying that the
unix socket 'slow' exists, it seems that it's not being use
>> # ==
>> smtp inet n - n - - smtpd
>> slow unix n - n - 1 smtp
>
> This is broken, the "unix" delivery agent needs to be "private" not
> "public".
Bang,
oops, the whole file is rather:
# Postfix master process configuration file. For details on the format
# of the file, see the master(5) manual page (command: "man 5 master").
#
# ==
# service type private unpriv chroot wak
> If the socket does NOT exist, then you mis-configured master.cf.
>
> If the socket does exist, then you mis-configured SELinux, AppArmor,
> Strace, whatever.
>
The socket does not exist, but I see no error in master.cf. Do you
see a problem?
slow unix n - n - 1
> Then you need to shutdown SELinux, AppArmor, etc. "security" software.
SELinux is disabled and I am using nothing else...
Andrew
> You need to add a "slow" transport to master.cf, then "postfix reload".
>
/etc/postfix/master.cf:
# slow transport for "delicate" domains, ie. Yahoo.com
slowunix n - n - 1 smtp
-o syslog_name=postfix-slow
-o smtp_helo_timeout=10
I have reloaded this many
strace qmgr shows:
connect(8, {sa_family=AF_FILE, path="private/slow"}, 110) = -1 ENOENT
(No such file or directory)
But I unsure if I should have touched a file somewhere - I read
nothing about that.
Andrew
Per instructions at
http://linuxnet.ca/postfix/dedicated_transport.html and tips by
several group members, I created a "slow" transport
for sending to Yahoo. I thought everything was working, but now I
notice "warning: connect to transport slow: No such file or directory"
in the logs.
I notice tha
16 matches
Mail list logo
