On 29/01/08 17:35 -0500, Alan and Pete wrote:
> I'm not all that familiar with syslog-ng, but couldn't you just have
> it write 
> to a (rotating) file which is picked up, processed & databasified by a
> cron 
> job?  Seems to be a pretty common way of doing such things...

Latency. I can't have more than (perhaps) 30 seconds latent.

On 29/01/08 17:54 -0500, Chris Knadle wrote:
> The way I envision you doing essentially the same thing is having the
> script read the log file via 'tail -f' (or something like it).

Its not a single file, we have to rotate it daily (not hourly like I
said before) so we can archive or delete logs by date. I may be able to
add that rotation into the script, but it seems like an easy place to
have a script failure. It may come to that.

On 29/01/08 17:54 -0500, Chris Knadle wrote:
>  Linux/Documentation/devices.txt says that the device is a syslog
>  local
> socket.  I'm not sure why you'd want to use sockets in this situation;
> maybe
> you can explain what you were considering.

If I can have my script open and listen to a socket and have syslog-ng
deliver straight into it. I think it would work, but I'm not sure if
syslog-ng would need to be restarted if the socket went away. I guess I
just have to test it.
-porkchop
_______________________________________________
Mid-Hudson Valley Linux Users Group                  http://mhvlug.org          
   
http://mhvlug.org/cgi-bin/mailman/listinfo/mhvlug                           
Upcoming Meetings (6pm - 8pm)                         MHVLS Auditorium          
        
  Feb 6 - DBUS
  Mar 5 - Setting up a platform-independent home/small office network using 
Linux

Reply via email to