Re: Messages not marked new
On Sun, Mar 26, 2006 at 08:22:20PM -0500, Kyle Wheeler wrote:
> When you read the messages with mutt, do they show up as read in the
> other client?
No, and that's the problem. I'll fire up mutt on my laptop, read a
mailbox and then close mutt. Then later I open mutt again and message
I know I read are marked "N".
My wild guess has been that I've got mutt open on that same mailbox
on another machine and somehow so the messages are not updated as
"read".
But, I have not been able to prove that. Often when I open the mail
and find something that's marked "N" when I *know* that I read that
message before and then I'll close mutt and reopen just to check then
the message isn't marked "N". It's messing with me, I think.
> What sort of backing mail store do you use?mbox or maildir?
Maildir on Debian Stable on a hardware raid drive (two disks on a
3-ware card). Nothing too exotic.
>
> It sounds like a problem with either Courier or your other mail
> reader, but just for grins, you can try checking out unsetting
> imap_peek in your muttrc. Ordinarily (when imap_peek is set, which is
> the default) mutt will avoid telling the imap server that messages
> have been read until you close or sync the folder with the server.
> Unsetting that setting will make it stop doing that, and messages will
> become read as soon as you view them in mutt.
Seems like a good way to go. I'll try that.
>
> My first-blush theory would be that mutt is explicitly leaving those
> messages read until you sync the flags to the server (as it does by
> default). Because you haven?t sync?d flags with the server before
> starting up your other client, in that other client the messages show
> up as unread. When you quit that other client, rather than ignoring
> those messages because it hasn?t touched them, it tells Courier that
> those messages should be unread. Courier takes this to be a
> flag-change, and so when mutt goes to sync it?s flags (i.e. you quit
> your first mutt), it is informed that it doesn?t have the most
> up-to-date information about the flag state, and so mutt?s changes are
> ignored. (note: that?s a VERY hand-wavey guess)
As good as my guesses, though.
Thanks. I'll report back if I can figure anything out.
--
Bill Moseley
moseley@xxxxxxxx