Mainstream open source monitoring is pretty much about munin, cacti, nagios, zabbix. You can make any of these run on openbsd, AFAIK.
Even though they serve different purposes, my favourite (if no custom, tailored solution is crafted) between these is cacti. However, its pretty disappointing the lack of support for alternative (see psql) backends :( </rant> On 8/10/10, Eugene Yunak <[email protected]> wrote: > On 10 August 2010 02:28, Jiri B. <[email protected]> wrote: >> Hello, >> >> I'm thinking to choose a monitoring tool which would run on OpenBSD >> of course. >> >> I have been working with Tivoli and Netview for couple of years so my >> idea is: >> >> * clients >> >> - heartbeats of course >> - simple interface to give a client some input as alert >> - text configuration on client node (can be pushed from central repo) >> - light >> >> * infrastructure nodes >> >> - proxy feature for far networks or dmz >> - filtering rules (thresholds, time filters ...) >> - text configuration >> - light >> >> * main server(s) >> >> - good filtering >> - surveillance console for monitoring center >> - be able to change status of an alert (acknowledge, closed, solved...) >> - be able to have some categories of clients based on roles >> >> I'm watching zabbix... not sure... >> >> If I wouldn't want event console I would probably check snmp -> sec -> >> snmptt. >> >> jirib >> >> > > Definitely nagios/cacti pair or zabbix. Having used nagios for a year > or so, i would never want to get back to Tivoli. It also gives you > lots of flexibility in how you setup your monitoring, and can neatly > work with snmp as well. > > Eugene > > -- > The best the little guy can do is what > the little guy does right
