Re: exim4 doesn't connect to smarthost

From: Robin Battey (zanfur_at_zanfur.com)
Date: 12/27/04

  • Next message: Alex Malinovich: "Re: OT: Re: Why do people use 1280x1024? (was Re: custom gdm screen resolution? [SOLVED - work around])"
    To: michf@post.tau.ac.il
    Date: Mon, 27 Dec 2004 11:46:09 -0800
    
    

    I just struggled with this for a few hours, found and solved the problem,
    and thought I'd post a response to explain what (I think) is going on in
    your situation.

    Exim (at least, exim4) in its infinite wisdom, caches transport "hints" in
    its retry and wait_transport databases. When my smarthost went down for a
    decently long period of time, my exim installation cached that information
    and didn't even bother checking to see if it was up, which is why you see
    no connections using tcpdump.

    The solution: Exim comes with three utilities for managing the hints
    caches, called exim_dumpdb, exim_tidydb, and exim_fixdb. The exim_dumpdb
    program displays the contents of the cache, exim_tidydb removes old
    entries, and exim_fixdb recovers corrupt caches. Check the man page for
    more info. The one you'll need is most likely exim_tidydb, to remove the
    out-of-date entries. I ran it with an argument of one second, to
    effectively remove all of them (they'll be recreated on demand). After
    that, everything magically started working.

    Cheers!
    -robin

    -- 
    Using Opera's revolutionary e-mail client: http://www.opera.com/m2/
    -- 
    To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org 
    with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
    

  • Next message: Alex Malinovich: "Re: OT: Re: Why do people use 1280x1024? (was Re: custom gdm screen resolution? [SOLVED - work around])"

    Relevant Pages

    • RE: Enterprise Manager
      ... You probably just needed to do a refresh on the appropiate collection in enterprise manager - it caches the entries. ... "mike" wrote: ...
      (microsoft.public.sqlserver.programming)
    • Automatic `nodump flag?
      ... to be excluded from the regular filesystem dumps. ... new entries appear in the caches, ... nodump flag set. ...
      (freebsd-questions)
    • Re: Thread-safe caches
      ... PC> new entries in the caches are actually not written every now and then, ... i think plists are thread-safe when setq, rplaca and rplacd are atomic, ... concurrent incompatible updates possible and what to do in this case. ...
      (comp.lang.lisp)