Use of 'returnaddress' channel keyword on 'inline' Directory channels - VMS

This is a discussion on Use of 'returnaddress' channel keyword on 'inline' Directory channels - VMS ; Hi, %PMDF-I-VERSION, PMDF version is PMDF V6.3 COMPAQ AlphaServer DS20E 666 MH running OpenVMS Alpha V8.3 PMDF_SHARE_LIBRARY version V6.3-x13; linked 00:12:58, Aug 26 2007 I have a number of Directory channels, one for each domain that is hosted here. All ...

+ Reply to Thread
Results 1 to 2 of 2

Thread: Use of 'returnaddress' channel keyword on 'inline' Directory channels

  1. Use of 'returnaddress' channel keyword on 'inline' Directory channels

    Hi,

    %PMDF-I-VERSION, PMDF version is PMDF V6.3
    COMPAQ AlphaServer DS20E 666 MH running OpenVMS Alpha V8.3
    PMDF_SHARE_LIBRARY version V6.3-x13; linked 00:12:58, Aug 26 2007

    I have a number of Directory channels, one for each domain that is hosted
    here. All of them use a PMDF CRDB file for address lookups, and all have
    the 'inline' channel keyword so that the SMTP server process checks the
    specified recipient address as soon as a RCPT TO command is received. I
    also set 'returnaddress postmaster@domain.name' (where 'domain.name'
    corresponds to the domain being served by this channel) so that PMDF-
    generated nondelivery reports would go back with a domain-specific
    Postmaster address.

    However it seems to me that if 'inline' is in use on such channels, PMDF
    won't ever generate any reports: either the address will be valid at the
    time the message is received, or it won't be and the SMTP server will reject
    it immediately with a "550 5.1.1 unknown or illegal user" error. If
    something goes wrong with delivery of the message after it's been accepted
    by the SMTP server process, it won't be in the context of the Directory
    channel and so the 'returnaddress ...' setting won't be in effect.

    Have I understood this correctly, or have I missed something?

    Thanks,

    Jeremy Begg

    +---------------------------------------------------------+
    | VSM Software Services Pty. Ltd. |
    | http://www.vsm.com.au/ |
    | "OpenVMS Systems Management & Programming" |
    |---------------------------------------------------------|
    | P.O.Box 402, Walkerville, | E-Mail: jeremy@vsm.com.au |
    | South Australia 5081 | Phone: +61 8 8221 5188 |
    |---------------------------| Mobile: 0414 422 947 |
    | A.C.N. 068 409 156 | FAX: +61 8 8221 7199 |
    +---------------------------------------------------------+

  2. Re: Use of 'returnaddress' channel keyword on 'inline' Directory channels

    > Hi,

    > %PMDF-I-VERSION, PMDF version is PMDF V6.3
    > COMPAQ AlphaServer DS20E 666 MH running OpenVMS Alpha V8.3
    > PMDF_SHARE_LIBRARY version V6.3-x13; linked 00:12:58, Aug 26 2007


    > I have a number of Directory channels, one for each domain that is hosted
    > here. All of them use a PMDF CRDB file for address lookups, and all have
    > the 'inline' channel keyword so that the SMTP server process checks the
    > specified recipient address as soon as a RCPT TO command is received. I
    > also set 'returnaddress postmaster@domain.name' (where 'domain.name'
    > corresponds to the domain being served by this channel) so that PMDF-
    > generated nondelivery reports would go back with a domain-specific
    > Postmaster address.


    The key point here is "PMDF-generated". Such a keyword obviously can have
    no effect on other agent's behavior when generating DSNs.

    > However it seems to me that if 'inline' is in use on such channels, PMDF
    > won't ever generate any reports: either the address will be valid at the
    > time the message is received, or it won't be and the SMTP server will reject
    > it immediately with a "550 5.1.1 unknown or illegal user" error. If
    > something goes wrong with delivery of the message after it's been accepted
    > by the SMTP server process, it won't be in the context of the Directory
    > channel and so the 'returnaddress ...' setting won't be in effect.


    Yep. In fact one could argue that the entire practice of having
    postmaster copies of DSNs is wrongminded because it violates the
    least astonishment principle.

    > Have I understood this correctly, or have I missed something?


    Sounds right to me.

    Ned

+ Reply to Thread