Synth-DIY Yahoo! Groups Archives

milter-greylist rejecting all mail

2009-01-20 by Greg Troxel

I have been running milter-greylist from pkgsrc successfully on
NetBSD/i386 4-stable with postfix 2.5.x for a long time. I also use
spamass-milter. Recently I started getting the following (private data
redacted; it all looked fine).

Jan 19 11:47:38 hostname postfix/smtpd[16975]: warning: milter unix:/var/milter-greylist/milter-greylist.sock: can't read SMFIC_RCPT reply packet header: Connection timed out
Jan 19 11:47:38 hostname postfix/smtpd[16975]: NOQUEUE: milter-reject: RCPT from sending.host[ip.addr.c.d]: 451 4.7.1 Service unavailable - try again later; from=<fromuser@them> to=<user@hostname> proto=ESMTP helo=<fqdn.org>

The only thing that I think changed is that my box went from having a v6
address that didn't work to one that did, but that happened several days
before the problem.

It sort of looks like milter-greylist has gone out to lunch and doesn't
answer in time. Could this happen with bad RBL servers? - I do have a
lot of rules to delay longer for hosts on blocklists.

I will look further, but thought I'd mention this in case anyone else
has seen it.

Re: [milter-greylist] milter-greylist rejecting all mail

2009-01-21 by Petar Bogdanovic

On Tue, Jan 20, 2009 at 08:25:31AM -0500, Greg Troxel wrote:
> 
> I have been running milter-greylist from pkgsrc successfully on
> NetBSD/i386 4-stable with postfix 2.5.x for a long time.  I also use
> spamass-milter.  Recently I started getting the following (private data
> redacted; it all looked fine).
> 
> Jan 19 11:47:38 hostname postfix/smtpd[16975]: warning: milter unix:/var/milter-greylist/milter-greylist.sock: can't read SMFIC_RCPT reply packet header: Connection timed out
> Jan 19 11:47:38 hostname postfix/smtpd[16975]: NOQUEUE: milter-reject: RCPT from sending.host[ip.addr.c.d]: 451 4.7.1 Service unavailable - try again later; from=<fromuser@them> to=<user@hostname> proto=ESMTP helo=<fqdn.org>

Similar setup, same problem..


> It sort of looks like milter-greylist has gone out to lunch and doesn't
> answer in time.  Could this happen with bad RBL servers?

Take a look at this session:

   postfix/smtpd[pid]: connect from unknown[ip]
   postfix/smtpd[pid]: NOQUEUE: reject: RCPT from unknown[ip]: 450 4.1.1 <ADDR>: Recipient address rejected: (...)
   postfix/smtpd[pid]: NOQUEUE: reject: RCPT from unknown[ip]: 550 5.1.1 <ADDR>: Recipient address rejected: (...)
   postfix/smtpd[pid]: NOQUEUE: reject: RCPT from unknown[ip]: 550 5.1.1 <ADDR>: Recipient address rejected: (...)
   postfix/smtpd[pid]: NOQUEUE: reject: RCPT from unknown[ip]: 550 5.1.1 <ADDR>: Recipient address rejected: (...)
   postfix/smtpd[pid]: NOQUEUE: reject: RCPT from unknown[ip]: 550 5.1.1 <ADDR>: Recipient address rejected: (...)
   postfix/smtpd[pid]: warning: milter unix:/var/milter/mgrey.sock: can't read SMFIC_RCPT reply packet header: Connection timed out
   postfix/smtpd[pid]: NOQUEUE: milter-reject: unknown[ip]: 451 4.7.1 Service unavailable - try again later; from=<ADDR>to=<ADDR> proto=SMTP helo=<ADDR>
   postfix/smtpd[pid]: NOQUEUE: reject: RCPT from unknown[ip]: 550 5.1.1 <ADDR>: Recipient address rejected: (...)
   postfix/smtpd[pid]: NOQUEUE: reject: RCPT from unknown[ip]: 550 5.1.1 <ADDR>: Recipient address rejected: (...)
   postfix/smtpd[pid]: NOQUEUE: reject: RCPT from unknown[ip]: 550 5.1.1 <ADDR>: Recipient address rejected: (...)
   postfix/smtpd[pid]: NOQUEUE: reject: RCPT from unknown[ip]: 550 5.1.1 <ADDR>: Recipient address rejected: (...)
   postfix/smtpd[pid]: too many errors after RCPT from unknown[ip]
   postfix/smtpd[pid]: disconnect from unknown[ip]


It occurs in the middle of a conversation with one single client and
since we run local caching nameservers, it's unlikely that lame RBLs
are the reason.


   Petar Bogdanovic

Milter-greylist

milter-greylist rejecting all mail

milter-greylist rejecting all mail

Re: [milter-greylist] milter-greylist rejecting all mail

Move to quarantaine