Re: [lemonade] Regarding CONDSTORE IMAP extension

Just to remind folk, the update to RFC 4551 - now also documenting QRESYNC
as well - is currently in Working Group Last Call, which ends on the 6th of
December.

https://ietf.org/doc/draft-ietf-qresync-rfc5162bis/

If you think this needs to be clearer in the new document, please let the
document authors, and the mailing list at imapext@ietf.org know.

On Tue, Dec 3, 2013 at 12:02 AM, Samsung SVL <samsungsvl@yahoo.com> wrote:

> Hi Chris,
>
> Thanks for your clarification.
>
> Regards,
> Jay
>
>   ------------------------------
>  *From:* Chris Newman <chris.newman@oracle.com>
> *To:* Samsung SVL <samsungsvl@yahoo.com>; lemonade@ietf.org
> *Sent:* Monday, 2 December 2013, 18:10
>
> *Subject:* Re: [lemonade] Regarding CONDSTORE IMAP extension
>
> --On December 2, 2013 13:51:24 -0800 Samsung SVL <samsungsvl@yahoo.com>
>
> wrote:
>
> > Hi All,
> >
> > Kindly clarify my below query regarding "CONDSTORE" IMAP Extension.
> >
> > When IMAP server advertises the "CONDSTORE" support as part of
> > CAPABILITIES, client assumes that server support IMAP CONSTORE
> > extension.
> >
> > As per RFC 4551,  client issues SELECT command  for a mailbox in the
> > below format and receives the response with HIGHESTMODSEQ value as  "1"
> >
> > C: A142 SELECT INBOX
> >
> > During the subsequent SELECT command, client receives "HIGHESTMODSEQ"
> > value always set to "1". Its not getting incremented with respect to the
> > changes in the mailbox.
> >
> > Some IMAP server changes the HIGHESTMODSEQ value only when client issues
> > SELECT command in the below format.
> >
> > C:A142 SELECT "INBOX" (QRESYNC ("UIDVALIDITY" "HIGHESTMODSEQ"))
> >
> > Kindly clarify which format is right one, because most of the IMAP server
> > behaves differently.
>
>
> Technically, the former command is correct only if you've previously
> issued
> a condstore-enabling command (or don't want to use condstore) and the
> latter is correct only if the server advertises QRESYNC and you have
> issued
> "ENABLE QRESYNC" previously. For more details see RFC 4551 and RFC 5162.
>
> Have you tried using:
>
> C: A142 SELECT INBOX (CONDSTORE)
>
> That's a condstore-enabling command that servers advertising the CONDSTORE
> capability are required to support. Servers are free to not return
> HIGHESTMODSEQ at all until a condstore-enabling command occurs, and then
> they MUST return either HIGHESTMODSEQ or NOMODSEQ.
>
> Of course, if a server is returning HIGHESTMODSEQ and not updating it
> after
> flag changes, that server may have a bug, so you may have to fallback to
> not using CONDSTORE with that server and just do a full resync every time.
> You might want to report that bug to the server vendor so they have the
> opportunity to fix it in a subsequent release.
>
>         - Chris
>
>
>
>
> _______________________________________________
> lemonade mailing list
> lemonade@ietf.org
> https://www.ietf.org/mailman/listinfo/lemonade
> Supplemental Web Site:
> http://www.standardstrack.com/ietf/lemonade
>