matrix o' monitors

[Paul Lantinga <prl@xxxxxx>]

> > Definitely.  Check this page for a start and look for "mon"
> > or "ldirectord"
> > - http://www.linuxvirtualserver.org/HighAvailability.html. 
> whoops, I forgot keepalived! - 
> http://keepalived.sourceforge.net/.  sorry

After seeing Peter's posts on monitoring, I'm wondering if anyone has put together a blow by blow matrix/comparison chart of all the available monitoring methods (or at least the most well known or used ones) for lvs?

I'm not trying to start an arguing match over which monitor is bigger, better, or shinier.  I'm thinking that, depending on your particular setup of lvs, one particular type of monitor may be more suitable than another.

An example of such could list the features/options of:  shell scripts, cron, mon, ldierctord, keepalived, lvs' daemon, heartbeat, fping, big brother, et al

and a listing of most of the common lvs setup scenarios: 

LVS related:
lvs-dr
lvs-nat
lvs-tun
lvs-ha  -using internal lvs daemon (v9.10)
        -using external ha
1 lvs box
2 lvs boxen
>2 lvs boxen
lvs box local node
lvs box with other non-lvs related production services

Real Server related:
<5 real servers, 5-10 real servers, >10 real servers
Equal weight real servers
UnEqual weight real servers

Routing related:
ipsec
tunnels (gre, etc)
ipchains
iptables
multiple ingress/egress networks
limited bandwidth
unlimited bandwidth

Contact methods:
snmp
paging
email
winpopup
write tty (e.g. wall, heh heh)
wav file
klaxon alarms
missile launches to lvs authors,   or better yet,
auto $ xfer to lvs authors after n*100 days uptime.

ok, well you get the idea.

Paul.