> "M" == Marco writes:
M> thank you for the info. I trusted an old commit which I found on a
M> SPEC file for Fedora Cyrus IMAP 3.0.x. I'm trying to build an RPM
M> file.
Yes, you shouldn't trust that as I froze the Cassandane version in
Fedora's packaging at an appropriate checkout back w
On Wed, 2020-05-06 at 09:11 -0400, Ken Murchison wrote:
> You may have a bunch of messages that were marked as \Deleted (and
> not
> displayed by your client) but haven't been expunged.
I have my every-day-all-day e-mail client, evolution, set to display
deleted messages (it shows them with a str
Hello,
On 06/05/2020 02:15, ellie timoney has written:
Cassandane is designed to be run from a git clone, and if you're using it, you
should generally keep it up to date. It is kept generally backwards
compatible, because I use it for testing new releases from old branches. You
don't need di
You may have a bunch of messages that were marked as \Deleted (and not
displayed by your client) but haven't been expunged.
How many of the messages shown by mbexamine have the \Deleted flag set?
On 5/6/20 8:32 AM, Brian J. Murrell wrote:
On Wed, 2020-05-06 at 07:18 -0400, Ken Murchison wrot
On Wed, 2020-05-06 at 07:18 -0400, Ken Murchison wrote:
> You can use 'mbexamine' and 'unexpunge -l'
mbexamine looks interesting, but unexpunge -l returns nothing for my
INBOX. I think this is because expunge is immediate in 2.4.17 isn't
it?
mbexamine's output is pretty terse though. Given this
You can use 'mbexamine' and 'unexpunge -l' on a mailbox and compare the
UIDs listed by the 2 commands to the message files in the mailbox directory.
I you think you have orphaned (not expunged) message files, you can try
'reconstruct -G' on the mailbox to restore them into the index.
On 5/6/
On Fri, 2020-05-01 at 14:29 -0400, Brian J. Murrell wrote:
> I'm fairly convinced that I have (lots of) files in my Cyrus mail
> spool
> that are not actually in any index -- orphan files.
>
> How can I verify this and identify the orphan files?
Nobody has any ideas at all about how this can be v
