procmail
[Top] [All Lists]

Re: Mailboxes marked as new

2001-06-23 18:36:15
Robert Hough <rch(_at_)solveinteractive(_dot_)com> writes:
On Sat, Jun 23, 2001, Philip Guenther wrote:

Um, if new mail came in, then the mailbox *should* be marked as "new"!

Someone bored this weekend? :) I thought this was covered in my original
message?

Well, perhaps I see several ways to interpret what you said.  Thus my
request for a precise description.


For the sake of correcting a possible bug in Procmail, I can get more
descriptive if you so wish. However, considering that I have resolved my
problem, I don't see the need to bore the list with such details. I'll
leave the choice up to you though.


Looking at the mutt source (version 1.2.5), the only way I can see for
mutt to claim a mailbox contains new mail, is if it does.  Which makes
your original thought that locking might be a problem is probably correct
(what's the output of "mutt -v" and "procmail -v"?) but that's a matter of
configuration more so than usage.  Switching to maildir eliminates locking
as a source of problems, so we can probably write it off to that cause.



Advances in file system technology, such as soft updates for Berkeley
FFS, can mitigate this effect, but such advances are not widely
deployed.

Just an FYI for those interested. Solaris 7+ also has a similiar feature
that can be enabled using the mount option 'logging'.

Actually, journaling filesystems, such as the 'logging' option in
Solaris 7, are quite different from soft updates, both in the consistency
guarantees that are made and the performance results.  Those interested
in the details should check out:

http://www.usenix.org/publications/library/proceedings/usenix2000/general/seltzer.html

(That's just the abstract.  You have to be a USENIX member to access
the full paper.)


Philip Guenther
_______________________________________________
procmail mailing list
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail

<Prev in Thread] Current Thread [Next in Thread>