Re: [exim] Logging to a database

Top Page
Delete this message
Reply to this message
Author: Wakko Warner
Date:  
To: David Powers
CC: exim-users
Subject: Re: [exim] Logging to a database
Please CC.

> I see advantages and disadvantages to using the existing exim logging
> mechanism. On the one hand you are sure to get everything that is currently
> logged to the flat file in your database, but on the other you are going to
> get everything. ;) At the very least you are going to have to pick through
> the log output for the items you are really interested in. I think the
> advantage of the database logging via acls is that it allows you to save
> exactly what you want and easily change the format. No patching, no external
> programs. If the acl runs, the item gets logged with exactly the information
> you choose. This, of course, is only useful for mail logging. Panic
> conditions, crashes, spool run times, etc... those are still in the basic
> logs and I don't know a good way to capture most of them via the acls.


For what I wanted, everything in sql was fine. queue runners, smtp
connections, etc. I was wanting this at work so that viewing logs would be
easy for the people who don't know how to navigate the command prompt.

Since I've only been asked a few times in the 5 years the server is running
to find something, it was a moot point at best. Which is why I never
bothered to make the patch.

> > Ok, I've seen 3 messages about this.
> >
> > 1) peter bowyer: use of readsocket
> >
> > 2) hochstrasser benedikt: use of syslog
> >
> > 3) david powers: using acls to update sql.
> >
> > For 1 and 3, I was thinking more of having the logging facility that's
> > already in exim to do the logging. I had thought of doing this, but I have
> > not felt strongly enough about it to attempt it.
> >
> > For 2, I prefer using the logfiles exim writes. syslog (the one I use
> > anyway) does not log the year.
>
> --
>
> ## List details at http://www.exim.org/mailman/listinfo/exim-users Exim details at http://www.exim.org/ ##

--
Lab tests show that use of micro$oft causes cancer in lab animals