On Sun, 18 May 2025 at 21:35, Helge Kreutzmann <[email protected]> wrote:
>
> Hello Richard,
> Am Sun, May 18, 2025 at 07:01:34PM +0000 schrieb Helge Kreutzmann:
> > Hello Richard,
> > Am Sun, May 18, 2025 at 07:12:15PM +0100 schrieb Richard Lewis:
> > > i think what i need to know is - how do i configure exim to match your set
> > > up. .probably thay is just what procmail package to install, and any
> > > settings for that? (i know nothing about procmail)
> > >
> > > im starting in a systemd-nspawn container and am not seeing the issue yet)
> >
> > I cannot tell you how I installed it initially. I started with Debian
> > around 2000, and the systems evolved, i.e. when installing a new
> > machine I usually copied over important configuration files.
> >
> > I noticed that exim contains some rules regarding procmail, but I'm
> > sure that I did not touch them. I have no idea how to configure exim
> > except by Debconf.
> >
> > The procmail recipie which is in place is neglible, so I just
> > deinstalled procmail.
> >
> > Let's see the next run …
>
> Ok, some mails no longer get properly sorted, but on this machine this
> is tolerable. But procmail is gone now.
>
> Now I get only single e-mails (with all the changes in the settings
> shown earlier).
>
> However, my server extensivly uses procmail (and logcheck), so this
> combo should really work. I have no idea how we can debug this in the
> hard freeze now. (My server is running stable, and does so for a
> reason).
>
> From my side, I'm now back to work, so before Friday I won't have much
> time for debugging.
>
> I would suggest to install a machine with local exim (per debconf) and
> some simple procmail rules, maybe just installing procmail is
> sufficient, I don't know.

this is exactly what i am trying to do!

so far, i could not reproduce the original bug. I have found one
oddity which may be related, but no lost mails

i have installed procmail, and it made no difference. but i suspect
it's not doing anything, but i dont know procmail, so i need some
help.

> And maybe for postponing the changes in 1.4.4 for the next development
> cycle?

that's not my decision to make.  without a reproducible bug, it seems unlikely?

Reply via email to