Hi ,Wietse
My system is centos6.2/64bit

version2.6.6
postconf >postconf.feel

version2.9.3
postconf >postconf.feels
diff postconf.feel postconf.feels >postconf.patch

cat postconf.patch
    3a4
    > address_verify_cache_cleanup_interval = 12h
    6c7
    < address_verify_map =
    ---
    > address_verify_map = btree:$data_directory/verify_cache
    16a18
    > address_verify_sender_dependent_default_transport_maps =
$sender_dependent_default_transport_maps
    ......


Different type with that got from yours, How can I patch?
Need your help, Thank you
Tom



2012/6/12 Feel Zhou <feelz...@gmail.com>

> /usr/local/src/postfix-2.9.3/src/smtp/smtp_troble.c
> I find this file,and check it,got the different with your patch
> Maybe I know how to patch old file to the new file. Let me try it.
> When patch finish, I need to make install again. Am I right?
>
>
> 2012/6/12 Feel Zhou <feelz...@gmail.com>
>
>>
>> Hello, wietse
>>
>> My operating system is centos6.2/64bit, It has postfix 2.6.6 itself, I
>> tar the new postfix 2.9.3
>> Does this patch work in the 2.6.6 and 2.9.3 both?
>> Does this patch work in centos 6.2/32bits?
>> Would you show me the details about how to use this patch? Thank you very
>> much
>>
>>
>>
>> 2012/6/12 Viktor Dukhovni <postfix-us...@dukhovni.org>
>>
>>> On Mon, Jun 11, 2012 at 11:33:01AM -0400, Wietse Venema wrote:
>>>
>>> > Viktor Dukhovni:
>>> > > >         -o soft_bounce=yes
>>> > > >         -o smtp_fallback_relay=relayusa.example.com
>>> > >
>>> > > That said, neither form will forward rejected mail to the fallback
>>> > > relay. The soft-bounce feature operates at the wrong layer for that
>>> > > to happen. First the SMTP delivery agent gives up on the rejected
>>> >
>>> > That seems to violate the principle of least astonishment.
>>> Conceptually,
>>> > soft-bounce means treat 5xx replies as 4xx replies.
>>> >
>>> > [I also recall that this came up a few years ago, but then there
>>> > was no time to address this properly].
>>>
>>> Thanks, yes I think this came up before. While the patch crosses
>>> abstraction boundaries by exposing soft_bounce above the bounce
>>> routines, the result is less surprising to users and the new
>>> code fits reasonably cleanly into smtp(8).
>>>
>>> The OP will need to apply the patch or wait for an upgrade to the
>>> next snapshot release.
>>>
>>> --
>>>        Viktor.
>>>
>>
>>
>

Reply via email to