Can anyone give me some clues as to the best method to build a nagios 
plug-in to check for the existance (and correct running) of a slon 
daemon on a server? (We're using slony 1.0.5)

Some options include:

1) Looking at netstat to see if slon is listening

2) looking at the process list, looking for slon, and the cluster name

3) Checking the slon output log

4) Is it possible to "talk" to the slon daemon to see if it is OK?

5) Querying the master cluster database - does the slony schema record 
which nodes are subscribed and actually connected (for the purposes of 
replication)? ie Find out the state of all slaves nodes by talking to 
the master/origin only.

I think that the slon_watchdog2 script queries the database to see 
whether the node is working properly - is this the best approach?

We want to use nagios to monitor our (currently) two slon processes (on 
different servers) and to let us know if a slon process dies.

A plugin has to report an exit code of 0,1,2 or 3 (OK, Warning, 
Critical, Unknown) + a one line text message (eg "OK", "Slon daemon not 
found", "Slon node x not replicating" etc)

Ideally a slon nagios plugin wants to do the following:

a) Test to see if the slon daemon is running
b) If running, test to see if it is replicating data correctly
c) Return the appropriate result code

Can anyone suggest a reliable method of testing for correct running of 
the slon process that would meet the requirement above? Ot give me some 
pointers on how to get started. BTW, I'm no perl programmer, but can 
learn :)

Thanks

John Sidney-Woollett