Re: [Evolution] Search Folders on IMAP+

[Adam Tauno Williams <awilliam whitemice org>]

On Wed, 2011-01-05 at 11:37 -0430, Patrick O'Callaghan wrote:
> On Wed, 2011-01-05 at 09:42 -0500, Philippe LeCavalier wrote:
> > i) I cannot confirm whether or not Evo or the server for that matter
> > is in fact unsubscribing me from certain folders since the checkmark
> > in Evo remains and the folder subscription settings in any webmail
> > client is correct. However, the folder(s) in question are no longer in
> > the folder list. So whether I'm unsubscribed or not the folder isn't
> > visible until I either remove and add the the checkmark in Evo OR I
> > unsubscribe and subscribe to said folder in a webmail client.
> The synchronization points between Evo and the server are something that
> don't seem to be documented in any obvious way. Evo appears to use "lazy
> synch", i.e. it only contacts the server when it has to. This can lead
> to apparent inconsistencies between multiple IMAP clients (including any
> webmail clients of course), but note that the IMAP spec explicitly does
> not define what is supposed to happen in these cases, so it's not
> technically a bug.

Mmm, with IDLE support it should be pretty consistent.  And enabling [if
supported] CONDSTORE extensions on the server may help as well.

But I think that only helps with messages; I don't think it does
anything about subscribe/unsubscribe events.

I'd enable telemetry logging on your IMAP server and induce [or wait
for] the bug. 

> > ii) The search folders function hangs on random folders at 0%
> > complete. When I click the exit to terminate the folder in question
> > disappears from me folder list and the above mentioned proceed needs
> > to be performed before I can see the folder again.
> I used to get this a lot, then with 2.30 it seemed to have been fixed,
> and now with 2.32 it has come back, though less than before. It's hard
> to reproduce and seems to happen essentially at random. I've suggested
> on previous occasions that Evo needs a way to force a reset of its
> network connections, rather than just hanging indefinitely, but since I
> don't know the precise details of what's going on it's not so easy to
> make a case for this, except to say that if you force-shutdown and
> restart than Evo usually recovers.