| Main Archive Page > Month Archives > postfix-users archives |
Re: Logging a more meaningful admin message on message_size_limit bounce
From: Kris Deugau <kdeugau_at_nospam>Date: Thu Oct 20 2011 - 19:59:11 GMT
To: Postfix users Mailing List <postfix-users@postfix.org>
Quanah Gibson-Mount wrote:
> --On Thursday, October 20, 2011 9:42 PM +0200 Reindl Harald
> <h.reindl@thelounge.net> wrote:
>> Am 20.10.2011 21:33, schrieb Quanah Gibson-Mount:
>>> A user gets:
>>> <user@zcs7.qa.qalab.com>: message size 18302670 exceeds size limit
>>> 5024000 of server zcs7-ga.qa.qalab.com[xx.xx.xx.xxx]
>>>
>>> The admin only gets to see:
>>> Sep 19 14:15:43 zcs7-ga postfix/smtpd[23718]: connect from
>>> unknown[xx.xx.xx.xxx]
>>> Sep 19 14:15:43 zcs7-ga postfix/smtpd[23718]: lost connection after EHLO
>>> from unknown[xx.xx.xx.xxx]
>>> Sep 19 14:15:43 zcs7-ga postfix/smtpd[23718]: disconnect from
>>> unknown[xx.xx.xx.xxx]
>> how should this work? the server does NOT know why the
>> CLIENT is closing the connection
>
> The server is the one rejecting the mail because it passed a limit set
> in postfix.
In the log example you show, the client connected, issued EHLO, then
disconnected.
The server has no way of knowing why the client disconnected in this case.
If the client instead continued the SMTP transaction and sent more in
DATA than the server allowed, *then* you should see the appropriate
entry in the log, since the client actively tried to do something the
server disallowed.
-kgd