[prev in list] [next in list] [prev in thread] [next in thread] 

List:       exmh-users
Subject:    persistent unseen messages
From:       "Brent B. Welch" <Brent.Welch () eng ! sun ! com>
Date:       1995-08-21 9:54:57
[Download RAW message or body]

Return-Path: glen@popcbr.rockefeller.edu 
Return-Path: <glen@popcbr.rockefeller.edu>
Received: from Eng.Sun.COM (engmail1.Eng.Sun.COM) by sage.Eng.Sun.COM (5.x/SMI-SVR4)
	id AA16725; Wed, 16 Aug 1995 13:38:04 -0700
Received: from mercury.Sun.COM (mercury.EBay.Sun.COM) by Eng.Sun.COM (5.x/SMI-5.3)
	id AA22925; Wed, 16 Aug 1995 13:37:52 -0700
Received: from popcbr.rockefeller.edu by mercury.Sun.COM (Sun.COM)
	id NAA05782; Wed, 16 Aug 1995 13:37:41 -0700
Received: from cbr6.rockefeller.edu ([129.85.4.6]) by popcbr.rockefeller.edu (4.1/SMI-4.1)
	id AA20151; Wed, 16 Aug 95 16:37:36 EDT
Received: from localhost (4.1/SMI-4.1)
	id AA01685; Wed, 16 Aug 95 16:37:34 EDT
Message-Id: <9508162037.AA01685@popcbr.rockefeller.edu>
To: "Brent B. Welch" <Brent.Welch@Eng>
Subject: Re: 1) persistent unseen messages, 2) reply is confused 
In-Reply-To: Your message of "Thu, 10 Aug 1995 10:46:15 PDT."
             <9508101746.AA27307@sage.Eng.Sun.COM> 
Date: Wed, 16 Aug 1995 16:37:32 EDT
From: Glen Gunsalus <glen@popcbr.rockefeller.edu>
Content-Type: text
Content-Length: 2815

I left this hanging so thought I'd tell you how much (or little) I'd
progressed.

Also wanted to thank you for your patience in moving my education forward
wrt private and public sequences.

The short answer is I deleted all .mh_sequences files and all problems
ceased.  None have yet reappeared so it's put my need to understand the
ramifications of private/public sequences and the way MH was compiled into
background.

Thanks again for your help.


>>>"Brent B. Welch" said:
 > 
 > glen@popcbr.rockefeller.edu said:
 > > options: [ATZ] [BIND] [BSD42] [BSD43] [DUMB] [FOLDPROT='"0700"'] 
 > > [ISI]          [MHE] [MHRC] [MIME] [MSGPROT='"0600"'] [NFS] [NOMHSEQ] 
 > >          [OVERHEAD] [POP] [POP2] [POPSERVICE='"pop3"'] [RPATHS] 
 > > [RPOP]          [SENDMTS] [SMTP] [SUN40] [TTYD] [TYPESIG=void] 
 > > [TYPESIG=void]          [UCI] [WHATNOW] [ZONEINFO] 
 > 
 > THe NOMHSEQ means that MH is compiled to use only private sequences,
 > and exmh is probably goofing in this case.
 > 
 > The difference between private and public sequences is this:
 > A public sequence is kept in the .mh_sequences file in the folder
 > directory.
 > A private sequence is kept in the context file, which is .exmhcontext
 > when you use exmh, and just plain old "context" when using regular MH.
 > 
 > While exmh tries to maintain private sequences, it is apparent that it
 > sometimes gets confused and updates the .mh_sequences file instead.
 > With exmh 1.6*, exmh does more direct manipulation of the MH-related files
 > as opposed to using the MH mark program.
 > Looking at the code, I can see that if you have stray copies of
 > .mh_sequences file (perhaps from earlier versions of exmh),
 > then exmh may get confused and update those files instead of
 > your context file.  I cannot see how new .mh_sequences files would
 > be created by exmh if there were not already one.
 > 
 > The summary is that exmh does better when you use public sequences.
 > For example, command line usage of MH programs will update the same
 > .mh_sequences file, and you'll avoid the apparent bug mentioned above.
 > 
 > It looks like a global rm of your .mh_sequences file will get you
 > into a good state with respect to exmh 1.6.2
 > IN THE CASE YOU USE PRIVATE SEQUENCES.
 > 
 > You force MH to use private sequences in two ways:
 > either put this into your .mh_profile ->
 > mh-sequences:
 > 
 > or compile MH with the NOMHSEQ option.  Verify this by
 > runing "inc -help"
 > 
 > --	Brent Welch	Sun Labs	brent.welch@eng.sun.com
 > 
 > 
 > 

--------
Glen L. Gunsalus, Ph.D.   Internet:    glen@popcbr.rockefeller.edu
The Population Council           g-gunsalus@popcbr.rockefeller.edu
1230 York Avenue          Bitnet:      glen@rockvax.bitnet
New York, NY 10021        
(212) 327-8757                Facsimile: (212) 327-8839, 7678

[prev in list] [next in list] [prev in thread] [next in thread]