The weirdness is all my fault. I'm running debian, but compiled 4.12
from source, installed to '/usr/local/exim/', installed an equivs
package, created a symlink for the log directory, but didn't create
one for the spool. So I removed the retry file for an old
installation.
I'm sorry to have wasted your time in an issue that should have
been known to me. I have learned a lot from this experience, though,
and your assistance is very appreciated.
thanks again,
hank
On Fri, May 02, 2003 at 05:15:17PM +0100, Philip Hazel wrote:
> On Fri, 2 May 2003, Adam Henry wrote:
>
> > The retry file I had deleted was fairly large, and this site does
> > handle a lot of mail. I would guess that this file would have been
> > recreated after two days, but that is not the case.
>
> Do you mean there is no file in the db directory? That's odd.
>
> > While not vital,
> > doesn't it serve an important function?
>
> Yes.
>
> > Additionally, 'exinext'
> > uses 'exim_dumpdb' to gather information about the retry times. It
> > would seem that it uses 'exim_dumpdb /var/spool/exim retry' in
> > some way to gather information.
>
> Correct.
>
> > However, when I run this command I
> > get this result:
> >
> > # exim_dumpdb /var/spool/exim retry
> > ** Failed to open database lock file
> > /var/spool/exim/db/retry.lockfile: No such file or directory
>
> That is odd.
>
> > However, as you see above, I can use 'exinext' just fine.
>
> That is even odder, because as you say, it is doing just that. Are you
> sure you are running the same exim_dumpdb?
>
> > So that
> > would suggest the retry file is only for "hints", and that there
> > is another location for a more authoritative listing of retry times.
>
> No. There is no other location. The hints are all there is.
>
> > When are the contents of this "master" database propagated into
> > the "hints" file?
>
> No, there really is no other database. Honest! Something really weird
> must be going on. This feels suspiciously as though there are two
> different builds of Exim around, and they are getting confused. But
> that's only a very wild guess.
>
> --
> Philip Hazel University of Cambridge Computing Service,
> ph10@??? Cambridge, England. Phone: +44 1223 334714.
