Who chopped my lines! - VMS

This is a discussion on Who chopped my lines! - VMS ; If I mail a file which has long lines (longer than 255 characters, probably) from a VMS system to a VMS system (over the internet; local VMS MAIL, i.e. within a cluster, is not affected), then the long lines are ...

+ Reply to Thread
Page 1 of 2 1 2 LastLast
Results 1 to 20 of 25

Thread: Who chopped my lines!

  1. Who chopped my lines!

    If I mail a file which has long lines (longer than 255 characters,
    probably) from a VMS system to a VMS system (over the internet; local
    VMS MAIL, i.e. within a cluster, is not affected), then the long lines
    are reformatted into 255-character lines (except that the last of these
    new lines, of course, might be less than 255 characters long). Only
    "newline" markers are inserted. Thus, context is lost if the line is
    supposed to be long. (With usenet headers, for example, the convention
    is to insert "newline" AND A SPACE to indicate that one line logically
    belongs to the previous one.

    Question: What is doing the chopping, the receiver or the sender? Or
    perhaps some mail gateway in-between?

    Can someone first verify that they can send a, say, 1000-character line
    via email and have it arrive (somewhere) intact, then send it to me from
    a non-VMS system?


  2. Re: Who chopped my lines!

    In article , helbig@astro.multiCLOTHESvax.de
    (Phillip Helbig---remove CLOTHES to reply) writes:

    > Question: What is doing the chopping, the receiver or the sender? Or
    > perhaps some mail gateway in-between?


    I suspect that it is the VMS TCPIP SMTP receiver, but I would like to
    confirm this.

    Assuming this is the case, is it RFC-conform?


  3. Re: Who chopped my lines!

    In article , helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:
    >In article , helbig@astro.multiCLOTHESvax.de
    >(Phillip Helbig---remove CLOTHES to reply) writes:
    >
    >> Question: What is doing the chopping, the receiver or the sender? Or
    >> perhaps some mail gateway in-between?

    >
    >I suspect that it is the VMS TCPIP SMTP receiver, but I would like to
    >confirm this.
    >
    >Assuming this is the case, is it RFC-conform?
    >

    The RFCs specify that an SMTP MTA must accept messages with lines upto 1000
    characters (including the terminating CRLF).

    See

    http://www.rfc-editor.org/rfc/rfc2822.txt

    Section 2.1.1. Line Length Limits


    A mail system wanting to send messages with longer lines should nowadays MIME
    encode the mail message eg Mime encoding it as quoted-printable or Base64 will
    convert it into an encoded message with a maximum linelength of 76 characters
    (plus CRLF).


    David Webb
    Security team leader
    CCSS
    Middlesex University




  4. Re: Who chopped my lines!

    david20@alpha2.mdx.ac.uk wrote:
    > In article , helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:
    >> In article , helbig@astro.multiCLOTHESvax.de
    >> (Phillip Helbig---remove CLOTHES to reply) writes:
    >>
    >>> Question: What is doing the chopping, the receiver or the sender? Or
    >>> perhaps some mail gateway in-between?

    >> I suspect that it is the VMS TCPIP SMTP receiver, but I would like to
    >> confirm this.
    >>
    >> Assuming this is the case, is it RFC-conform?
    >>

    > The RFCs specify that an SMTP MTA must accept messages with lines upto 1000
    > characters (including the terminating CRLF).
    >
    > See
    >
    > http://www.rfc-editor.org/rfc/rfc2822.txt
    >
    > Section 2.1.1. Line Length Limits
    >
    >
    > A mail system wanting to send messages with longer lines should nowadays MIME
    > encode the mail message eg Mime encoding it as quoted-printable or Base64 will
    > convert it into an encoded message with a maximum linelength of 76 characters
    > (plus CRLF).
    >
    >
    > David Webb
    > Security team leader
    > CCSS
    > Middlesex University
    >
    >
    >


    For VMS-to-VMS I usualy do zip->uuencode->mail->uudecode->unzip.
    That works for *any* file, test or binary.

    Jan-Erik.

  5. Re: Who chopped my lines!

    Phillip Helbig---remove CLOTHES to reply wrote:
    > If I mail a file which has long lines (longer than 255 characters,
    > probably) from a VMS system to a VMS system (over the internet; local
    > VMS MAIL, i.e. within a cluster, is not affected), then the long lines
    > are reformatted into 255-character lines (except that the last of these
    > new lines, of course, might be less than 255 characters long). Only
    > "newline" markers are inserted. Thus, context is lost if the line is
    > supposed to be long. (With usenet headers, for example, the convention
    > is to insert "newline" AND A SPACE to indicate that one line logically
    > belongs to the previous one.
    >
    > Question: What is doing the chopping, the receiver or the sender? Or
    > perhaps some mail gateway in-between?
    >
    > Can someone first verify that they can send a, say, 1000-character line
    > via email and have it arrive (somewhere) intact, then send it to me from
    > a non-VMS system?
    >


    Mail was meant for messages! Complaining that it doesn't handle freight
    very well is pointless!

    Try FTP.

  6. Re: Who chopped my lines!

    In article , david20@alpha2.mdx.ac.uk
    writes:

    > >> Question: What is doing the chopping, the receiver or the sender? Or
    > >> perhaps some mail gateway in-between?

    > >
    > >I suspect that it is the VMS TCPIP SMTP receiver, but I would like to
    > >confirm this.


    Actually, I suppose the sender is more likely.

    > >Assuming this is the case, is it RFC-conform?
    > >

    > The RFCs specify that an SMTP MTA must accept messages with lines upto 1000
    > characters (including the terminating CRLF).
    >
    > See
    >
    > http://www.rfc-editor.org/rfc/rfc2822.txt
    >
    > Section 2.1.1. Line Length Limits


    I still don't know if it is the receiver or the sender, but it seems to
    me that sending a message from VMS MAIL to VMS MAIL over the internet
    with HP TCPIP 5.4 with lines between 255 and 1000 characters in length
    has the long lines reformatted---apparently in violation of the RFC.
    How do newer versions of TCPIP do?


  7. Re: Who chopped my lines!

    In article ,
    =?ISO-8859-1?Q?Jan-Erik_S=F6derholm?=
    writes:

    > For VMS-to-VMS I usualy do zip->uuencode->mail->uudecode->unzip.
    > That works for *any* file, test or binary.


    Yes, that should always work. (Sending such a uuencoded ZIP file to a
    non-VMS system often gets processed correctly, even if there is no
    MIME-encoding sent.

    In my case, I am concerned with usenet headers in connection with
    moderated newsgroups. This is non-zipped, non-encoded text-only stuff
    (which on the whole is a GOOD thing). Knowing the problem, I can repair
    the lines at my end. (Contrary to my previous post, I now think it is
    the receiver doing the reformatting, since I see it on many incoming
    messages which don't come from VMS; maybe the sender reformats as well,
    I don't know.) However, if I redirect a message to another moderator,
    say, it is a shame that long lines get reformatted (which is how I
    noticed the problem in the first place).


  8. Re: Who chopped my lines!

    > sending a message from VMS MAIL to VMS MAIL over the internet
    > with HP TCPIP 5.4 with lines between 255 and 1000 characters...



    But, who on earth would want to do that anyway ?
    Isn't this a non-issue, for all practical purposes ?


    Jan-Erik.


  9. Re: Who chopped my lines!

    In article , "Richard B.
    Gilbert" writes:

    > Mail was meant for messages! Complaining that it doesn't handle freight
    > very well is pointless!


    I agree, in general. In my case, I am concerned with mail-transmitted
    usenet posts containing headers.

    In any case, though, a several-years-old RFC should be respected.


  10. Re: Who chopped my lines!

    In article ,
    =?ISO-8859-1?Q?Jan-Erik_S=F6derholm?=
    writes:

    > > sending a message from VMS MAIL to VMS MAIL over the internet
    > > with HP TCPIP 5.4 with lines between 255 and 1000 characters...

    >
    > But, who on earth would want to do that anyway ?
    > Isn't this a non-issue, for all practical purposes ?


    As I mentioned in passing in another post, as the moderator of a
    newsgroup, I receive usenet posts via email. For what it's worth, some
    headers, especially the References: header, can be quite long. And it
    is important, so that threads can be displayed properly. A usenet
    client CAN, but doesn't HAVE TO, seperate long headers into multiple
    lines, continuation lines being denoted by a beginning space. In fact,
    I reformat the posts I feed back into the system this way. However,
    most do not, and I get emails with long lines.

    I know about the problem and since nothing is lost, I can deal with it,
    but it is a pain when redirecting posts to another moderator, say.
    (Normally, if I'm not active as a moderator, posts don't pass through me
    at all, but there are exceptional cases.)


  11. Re: Who chopped my lines!

    Phillip Helbig---remove CLOTHES to reply wrote:
    > In article ,
    > =?ISO-8859-1?Q?Jan-Erik_S=F6derholm?=
    > writes:
    >
    >> > sending a message from VMS MAIL to VMS MAIL over the internet
    >>> with HP TCPIP 5.4 with lines between 255 and 1000 characters...

    >> But, who on earth would want to do that anyway ?
    >> Isn't this a non-issue, for all practical purposes ?

    >
    > As I mentioned in passing in another post,


    And my reply above was sent before I had got that
    "other post" from you about mail-headers... :-) :-)

    If I'd known then about the actual "problem" I had
    never posted the comment above.

    Your specific issue might not fall within
    "all practical purposes"... :-) :-)

    Jan-Erik.

  12. Re: Who chopped my lines!

    In article , helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:
    >In article ,
    >=?ISO-8859-1?Q?Jan-Erik_S=F6derholm?=
    >writes:
    >
    >> For VMS-to-VMS I usualy do zip->uuencode->mail->uudecode->unzip.
    >> That works for *any* file, test or binary.

    >
    >Yes, that should always work. (Sending such a uuencoded ZIP file to a
    >non-VMS system often gets processed correctly, even if there is no
    >MIME-encoding sent.
    >
    >In my case, I am concerned with usenet headers in connection with
    >moderated newsgroups. This is non-zipped, non-encoded text-only stuff
    >(which on the whole is a GOOD thing). Knowing the problem, I can repair
    >the lines at my end. (Contrary to my previous post, I now think it is
    >the receiver doing the reformatting, since I see it on many incoming
    >messages which don't come from VMS; maybe the sender reformats as well,
    >I don't know.) However, if I redirect a message to another moderator,
    >say, it is a shame that long lines get reformatted (which is how I
    >noticed the problem in the first place).
    >

    Does the mail pass through DECNET at any point rather than just being straight
    SMTP-to-SMTP ? I seem to remember discussions on the PMDF newsgroup about a 255
    character limit when mail was passed over DECNET.

    David Webb
    Security team leader
    CCSS
    Middlesex University



  13. Re: Who chopped my lines!

    In article , david20@alpha2.mdx.ac.uk
    writes:

    > Does the mail pass through DECNET at any point rather than just being
    > straight SMTP-to-SMTP ? I seem to remember discussions on the PMDF
    > newsgroup about a 255 character limit when mail was passed over DECNET.


    Shouldn't be the problem. I'm not even running DECnet on my cluster (no
    need), and I can't think of any other place DECnet would be involved.


  14. Re: Who chopped my lines!

    On Aug 10, 5:45*am, hel...@astro.multiCLOTHESvax.de (Phillip Helbig---
    remove CLOTHES to reply) wrote:
    > If I mail a file which has long lines (longer than 255 characters,
    > probably) from a VMS system to a VMS system (over the internet; local
    > VMS MAIL, i.e. within a cluster, is not affected),


    Just use SEND/FOREIGN ?

    Hein.


  15. Re: Who chopped my lines!

    In article
    <39d7ff56-f2cc-42ce-9522-005021a4ac9c@w7g2000hsa.googlegroups.com>, Hein
    RMS van den Heuvel writes:

    > On Aug 10, 5:45=A0am, hel...@astro.multiCLOTHESvax.de (Phillip Helbig---
    > remove CLOTHES to reply) wrote:
    > > If I mail a file which has long lines (longer than 255 characters,
    > > probably) from a VMS system to a VMS system (over the internet; local
    > > VMS MAIL, i.e. within a cluster, is not affected),

    >
    > Just use SEND/FOREIGN ?


    Might be an option. Will have to look up the documentation. I've heard
    of this, but never used it. Obvious commands give a syntax error.
    Again, will have to look it up in the documentation on undocumented
    features (unless someone is quicker and posts here).


  16. Re: Who chopped my lines!

    Not sure in your specific case but:

    VMSmail, character cell, will only output the first 256 characters of a
    line. They will wrap around a few lines. The rest of a line is in the
    mail file but not displayed.

    VMSmail will truncate lines greater than 256 characters when transported
    via DECnet.

  17. Re: Who chopped my lines!

    Phillip Helbig---remove CLOTHES to reply wrote:
    > In article
    > <39d7ff56-f2cc-42ce-9522-005021a4ac9c@w7g2000hsa.googlegroups.com>, Hein
    > RMS van den Heuvel writes:
    >
    >> On Aug 10, 5:45=A0am, hel...@astro.multiCLOTHESvax.de (Phillip Helbig---
    >> remove CLOTHES to reply) wrote:
    >>> If I mail a file which has long lines (longer than 255 characters,
    >>> probably) from a VMS system to a VMS system (over the internet; local
    >>> VMS MAIL, i.e. within a cluster, is not affected),

    >> Just use SEND/FOREIGN ?

    >
    > Might be an option. Will have to look up the documentation. I've heard
    > of this, but never used it. Obvious commands give a syntax error.
    > Again, will have to look it up in the documentation on undocumented
    > features (unless someone is quicker and posts here).
    >


    I thought that was an DECnet-to-DECnet option only.
    I might be wrong, it's 15+ years since I last used it... :-)

    Jan-Erik.

  18. Re: Who chopped my lines!


    "Jan-Erik Söderholm" wrote in message
    news:nUHnk.1652$U5.971@newsb.telia.net...
    > Phillip Helbig---remove CLOTHES to reply wrote:
    > > In article
    > > <39d7ff56-f2cc-42ce-9522-005021a4ac9c@w7g2000hsa.googlegroups.com>, Hein
    > > RMS van den Heuvel writes:
    > >
    > >> On Aug 10, 5:45=A0am, hel...@astro.multiCLOTHESvax.de (Phillip

    Helbig---
    > >> remove CLOTHES to reply) wrote:
    > >>> If I mail a file which has long lines (longer than 255 characters,
    > >>> probably) from a VMS system to a VMS system (over the internet; local
    > >>> VMS MAIL, i.e. within a cluster, is not affected),
    > >> Just use SEND/FOREIGN ?

    > >
    > > Might be an option. Will have to look up the documentation. I've heard
    > > of this, but never used it. Obvious commands give a syntax error.
    > > Again, will have to look it up in the documentation on undocumented
    > > features (unless someone is quicker and posts here).
    > >

    >
    > I thought that was an DECnet-to-DECnet option only.
    > I might be wrong, it's 15+ years since I last used it... :-)
    >
    > Jan-Erik.


    I used to use SEND/FOREIGN to transport files across an X.25 link, so it's
    not limited purely to DECNET to DECNET transports.
    --
    Lee K. Gleason N5ZMR
    Control-G Consultants
    lee.gleason@comcast.net



  19. Re: Who chopped my lines!

    Hein RMS van den Heuvel wrote:
    > On Aug 10, 5:45 am, hel...@astro.multiCLOTHESvax.de (Phillip Helbig---
    > remove CLOTHES to reply) wrote:
    >> If I mail a file which has long lines (longer than 255 characters,
    >> probably) from a VMS system to a VMS system (over the internet; local
    >> VMS MAIL, i.e. within a cluster, is not affected),

    >
    > Just use SEND/FOREIGN ?


    Not only will it work VMS-VMS. Some SMTP implementations
    will actually be able to convert to Base64 encoded MIME
    when going out on the internet.

    Arne

  20. Re: Who chopped my lines!

    Lee K. Gleason wrote:
    > "Jan-Erik Söderholm" wrote in message
    > news:nUHnk.1652$U5.971@newsb.telia.net...
    >> Phillip Helbig---remove CLOTHES to reply wrote:
    >>> In article
    >>> <39d7ff56-f2cc-42ce-9522-005021a4ac9c@w7g2000hsa.googlegroups.com>, Hein
    >>> RMS van den Heuvel writes:
    >>>> On Aug 10, 5:45=A0am, hel...@astro.multiCLOTHESvax.de (Phillip Helbig---
    >>>> remove CLOTHES to reply) wrote:
    >>>>> If I mail a file which has long lines (longer than 255 characters,
    >>>>> probably) from a VMS system to a VMS system (over the internet; local
    >>>>> VMS MAIL, i.e. within a cluster, is not affected),
    >>>> Just use SEND/FOREIGN ?
    >>> Might be an option. Will have to look up the documentation. I've heard
    >>> of this, but never used it. Obvious commands give a syntax error.
    >>> Again, will have to look it up in the documentation on undocumented
    >>> features (unless someone is quicker and posts here).
    >>>

    >> I thought that was an DECnet-to-DECnet option only.
    >> I might be wrong, it's 15+ years since I last used it... :-)

    >
    > I used to use SEND/FOREIGN to transport files across an X.25 link, so it's
    > not limited purely to DECNET to DECNET transports.


    I believe that it was even documented for PSI (X25) !

    Arne

+ Reply to Thread
Page 1 of 2 1 2 LastLast