|
Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com |
Subject: Re: "table has changed -- exiting" in maillog
From: Wietse Venema (wietse
porcupine.org)Date: Fri Oct 06 2000 - 08:53:10 CDT
- Next message: Lars Hecking: "Re: Postfix snapshot 20001005 available"
- Previous message: Wietse Venema: "Re: alternate hashing scheme (was: Re: Feature ... more generic $mail_spool ...)"
- In reply to: Michael Tokarev: "Re: "table has changed -- exiting" in maillog"
- Reply: Wietse Venema: "Re: "table has changed -- exiting" in maillog"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Michael Tokarev:
> Ralf Hildebrandt wrote:
> >
> []
> > It's not a problem
> > It's perfectly normal -- could it be that you rebuild aliases using a cronjob?
> > That would cause e.g. local to reload the table each time it changed.
>
> Just little thought -- will it be ok for postfix process to _not_ to exit
> if it encountered a table change _and_ it is not run chrooted?
> I.e., in the place where this "table changed" check is performed, is it
> ok to just reopen the map _if_ it is not chrooted? Such a change should
> be trivial, but I want to be shure if this is ok, what other people
> think about this?
What is the problem? If you're updating maps constantly, perhaps
it is better to use a database instead. Updating a map means writing
the whole thing from scratch, which is expensive when you do it
frequently.
Wiese
- Next message: Lars Hecking: "Re: Postfix snapshot 20001005 available"
- Previous message: Wietse Venema: "Re: alternate hashing scheme (was: Re: Feature ... more generic $mail_spool ...)"
- In reply to: Michael Tokarev: "Re: "table has changed -- exiting" in maillog"
- Reply: Wietse Venema: "Re: "table has changed -- exiting" in maillog"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]