home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!olivea!decwrl!decwrl!infopiz!mccall!ipmdf-newsgate!list
- From: dan@innosoft.com (Daniel C. Newman)
- Newsgroups: vmsnet.mail.pmdf
- Subject: RE: Problem with printer channel?
- Message-ID: <01GNN6E8XQ3694DZVH@INNOSOFT.COM>
- Date: 16 Aug 92 16:52:03 GMT
- Organization: The Internet
- Lines: 137
- Return-Path: <epmdf@YMIR.CLAREMONT.EDU>
- Resent-Date: 16 Aug 1992 08:52:03 -0800 (PST)
- Resent-From: epmdf@YMIR.CLAREMONT.EDU
- Errors-To: epmdf@YMIR.CLAREMONT.EDU
- Resent-Message-ID: <01GNN6FFPZJM95NC68@YMIR.CLAREMONT.EDU>
- X-Vms-To: IN%"C.Chaundy@its.unimelb.EDU.AU"
- X-Vms-Cc: IPMDF
- Mime-Version: 1.0
- Content-Type: TEXT/PLAIN; CHARSET=US-ASCII
- Content-Transfer-Encoding: 7BIT
-
- > With V4.1, I want to upgrade to use the standard PMDF printer channel and
- > wondered how it reacted if the mailbox part of the 'to' address was not a
- > AVPL to avoid having to re-educate users and make bulk changes to a huge
- > number of directory channel files - I got the answer - *badly*. But what
- > was worse was instead of generating a bounce message to the postmaster and
- > the originator, it generated a message to the postmaster and the recipient!
- > Of course, the recipient address was illegal (no AVPL) so a groovy little
- > mail loop started up :-).
-
- A patch for this (in the 4.1-8 timeframe) was posted to info-pmdf on 26 June.
- I've included the patch below.
-
- > Apart from the fact that I will have to move to AVPL pairs (it would have
- > been nice to default mailboxes that are not AVPL pairs to just 'Attention:'),
-
- No, the mail should just bounce. Doing otherwiise makes an assumption: it
- assumes that the rest of the address is correct. For instance, maybe I just
- sent mail to /.../@printer.innosoft.com and now I'm sending mail to Ned which
- and I accidentally send it to ned@printer.innosoft.com. Now it prints over
- on the admin. side of the Innosoft office and the wrong people have a chance
- to read it.... Anyhow, maybe this sounds like a weak argument but my point is
- that making assumptions about mailboxes is always fraught with peril and best
- avoided whenever possible.
-
- > I assume this other behaviour is most definitely a bug in the printer channel.
-
- It was.
-
- Dan
- ------------
-
- Date: 26 Jun 1992 09:36 -0800 (PST)
- From: "Daniel C. Newman" <DAN@HMCVAX.CLAREMONT.EDU>
- Subject: RE: Delivery failure on the Printer channel....
- To: SMITH%NYUMED.BITNET@ymir.claremont.edu, INFO-PMDF@YMIR.CLAREMONT.EDU
- Errors-to: epmdf@YMIR.CLAREMONT.EDU
- Resent-message-id: <01GLNZ343T428ZDYVD@YMIR.CLAREMONT.EDU>
- Message-id: <01GLNZ2EFH688WVYOD@HMCVAX.CLAREMONT.EDU>
- X-Envelope-to: ned+info-pmdf, portia
- X-VMS-To: IN%"SMITH@NYUMED.BITNET"
- X-VMS-Cc: IPMDF
- Content-type: TEXT/PLAIN; CHARSET=US-ASCII
- Content-transfer-encoding: 7BIT
-
- > I tried to use All-in-fun mail to create a message to send to the printer
- > channel. To make it more interesting, I put an attachment on it (Ultimately
- > it would be good to have PostScript attachments, since this is a LaserWriter
- > being driven by PacerPrint).
-
- Attachments won't buy you anything special. The printer channel merely prints
- verbatim the contents of the message body. If there's PostScript in it, it
- prints as text the PostScript. If there's an encoded binary attachment, it
- prints the encoded data.
-
- > What happened was that PMDF-MR and the pronter channel kept kicking the thing
- > back and forth: no printing though.... I then tried to kill it, whcih was
- > not at all easy (wait until the printer channel was working, stop the queue,
- > delete the job, delete the queued files).
-
- Well, if there is some sort of error ($SNDJBC call fails, attribute-value pairs
- in the printer address are malformed, etc.) the printer channel will bounce
- the message back to the address specified by the Errors-To: header. If no
- such header is present, the message is bounced to the envelope From: address
-
- Now, the problems turns out to be that the envelope To: address is accidentally
- being used in place of the envelope From: address and thereby generating the
- loop you're seeing.
-
- Dan
-
- Here's the VMS DIFFerences
-
- ************
- File PMDF_ROOT:[SRC]PRINTER.PAS;60
- 414 procedure add_bad (var env_to, why : vstring);
- 415
- ******
- File PMDF_ROOT:[SRC]PRINTER.PAS;59
- 414 procedure add_bad (var from, why : vstring);
- 415
- ************
- ************
- File PMDF_ROOT:[SRC]PRINTER.PAS;60
- 425 envelope_to := env_to;
- 426 reason := why;
- 427 next := nil;
- 428 end; (* with *)
- ******
- File PMDF_ROOT:[SRC]PRINTER.PAS;59
- 425 envelope_from := from;
- 426 reason := why;
- 427 next := nil;
- 428 end; (* with *)
- ************
-
-
- and here's the context differences
-
- *** PRINTER.PAS;60
- --- PRINTER.PAS;59
- **************
- *** 411,417
- printer_file : text;
- to_address_list, to_address_list_root : address_list_ptr;
-
- ! procedure add_bad (var env_to, why : vstring);
-
- begin (* add_bad *)
- if bad_addresses = nil then begin
- --- 411,417 -----
- printer_file : text;
- to_address_list, to_address_list_root : address_list_ptr;
-
- ! procedure add_bad (var from, why : vstring);
-
- begin (* add_bad *)
- if bad_addresses = nil then begin
- **************
- *** 422,430
- bad_addresses := bad_addresses^.next;
- end; (* if *)
- with bad_addresses^ do begin
- ! envelope_to := env_to;
- ! reason := why;
- ! next := nil;
- end; (* with *)
- end; (* add_bad *)
-
- --- 422,430 -----
- bad_addresses := bad_addresses^.next;
- end; (* if *)
- with bad_addresses^ do begin
- ! envelope_from := from;
- ! reason := why;
- ! next := nil;
- end; (* with *)
- end; (* add_bad *)
-