home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!decwrl!infopiz!mccall!ipmdf-newsgate!list
- Newsgroups: vmsnet.mail.pmdf
- Subject: RE: Defragment channel
- Message-ID: <01GOT6493RIQ9TCNO5@INNOSOFT.COM>
- From: "Daniel C. Newman" <dan@innosoft.com>
- Date: 15 Sep 1992 10:17:16 -0700 (PDT)
- Organization: The Internet
- Return-Path: <epmdf@YMIR.CLAREMONT.EDU>
- Resent-Date: 15 Sep 1992 10:17:16 -0700 (PDT)
- Resent-From: epmdf@YMIR.CLAREMONT.EDU
- Errors-To: epmdf@YMIR.CLAREMONT.EDU
- Resent-Message-ID: <01GOT66MZV3M9851ES@YMIR.CLAREMONT.EDU>
- X-Vms-To: IN%"bob@camb.com"
- X-Vms-Cc: IPMDF
- Mime-Version: 1.0
- Content-Type: TEXT/PLAIN; CHARSET=US-ASCII
- Content-Transfer-Encoding: 7BIT
- Lines: 58
-
- > I am trying to use 4.1 facilities to fragment large messages and then
- > reassemble them between two systems that communicate via PhoneNet.
- > The sending system is breaking the large messages into smaller
- > components with what appear to be reasonable MIME headers indicating
- > the multiple parts. However, the message is not being reassembled.
- > The pieces are delivered individually at the 2nd system. I assume
- > I've misconfigured something at that end.
-
- Does the remote system have a "defragment" channel. These are generated
- by the PMDF V4.1 configuration utility. If the remote system does not have
- such a channel, then add to the system's PMDF configuration rewrite rules of
- the form
-
- defragment.sigurd.innosoft.com $U@defragment.sigurd.innosoft.com
-
- and a channel block of the form
-
- defragment nox_env_to
- defragment.sigurd.innosoft.com
-
- Also, make sure that the local channel is marked "defragment"
-
- Now, if there is a defragment channel and all on the remote system, then
- the problem is probably one of crucial MIME header lines being lost. None
- of the involved channels are marked headeromit or headertrim are they? In
- the case of headertrim, you may wish to review which headers are being removed.
- You must preserve the Content-type: and Content-transfer-encoding: headers.
-
- If you have a defragment channel but the fragments are immediately being passed
- on (unassembled) then lost header lines are undoubtedly the problem.
-
- > I could use a pointer to documentation on how to configure the
- > target system.
-
- No such documentation exists at present.
-
- > Is there an options file that controls things like
- > how long the fragments are allowed to sit around before being
- > forwarded, as is, unreassembled? etc.
-
- Fragments sit around for min (notices[1],notices[2],...,notices[n]) DIV 2
- where notices[1],...,notices[n] are the days (or minutes) on which to return
- warning messages or bounce the message. The default is 3, 6, 9, 12 and so
- by default, the fragments are passed on, unassembled, after they have sat
- around for 3 DIV 2 = 1 day. If a notices keyword is used with the defragment
- channel, then those notices are used; otherwise, the notices associated with
- the local channel are used (which may be the defaults, 3, 6, 9, 12).
-
- > BTW, this appears to be a feature that will help us a lot on this
- > particular link which is used to transfer, via 19.2K PhoneNet, some
- > really large, multimegabyte files.
-
- This is one our main motivations for implementing this facility. Note, that
- message fragmention stuff is an example of MIME usage. The message to be
- fragmented is broken into smaller messages; each smaller message is marked
- as a MIME message/partial.
-
- Dan
-