HA Nagios system

[Steve Shipway]

>We have a requirement that our monitoring system has to be 
>99.995% or greater availability.  has anyone run Nagios in 
>this kind of environment? If so what techniques have been used?
>
>I was trying to think of using a Linux cluster, that shared a 
>SAN disk array.  Am I going down the right lines ?

We have a pair of Linux servers, running Linux-HA as a heartbeat.   A
virtual IP address is passed between them, and the Apache web server listens
on the wildcard interface for https.  This takes care of the web interface.

The Nagios service is started/stopped by linux-ha when it passes control
back and forth.  Configuration of nagios is held on the 'master' server, and
periodically copied over to the standby.  This takes care of config
synchronisation, and not running two copies in parallel (which we do for
some stuff, but Nagios generates a lot of network queries and alerts, so
better to have only one copy running at once).

The database currently is not copied automatically, only manually.  However,
we're looking into having it on SAN (or shared SCSI, or something) so that
linux-ha can dismount/remount it between servers as part of the nagios
startup.  This allows the logs and data to be taken care of.

We also run BigBrother and MRTG/routers.cgi on the same server set.

Steve

---
Steve Shipway: ITSS, University of Auckland
Email: s.shipway at auckland.ac.nz  Web: http://www.steveshipway.org/  
** We can only discover new oceans when we have the **
** courage to lose sight of the shore.              **
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Steve Shipway.vcf
Type: text/x-vcard
Size: 154 bytes
Desc: not available
URL: <https://www.monitoring-lists.org/archive/users/attachments/20040526/59377a43/attachment.vcf>