grody: munin is pretty -: grody uses RRDtool brycec: Oh sure munin is nice. So is nagios/opsview. And cacti. And I use them all. But I didn't think that munin (or the others) really meets their desire for host ping monitoring (while keeping it simple) grody: :)
i had all kinds of fun with cacti mercutio: i like cacti pyvpx: smokeping is exactly that
but I find ICMP ping'ing hosts to be of little value mercutio: you could do tcp pinging with hping with smokeing
i use curl with smokeping
i didn't really read what was said before, but it does sound like smokeping is what he wants
pyvpx: also icmp isn't of little value, its' just not the be all and end all.. pyvpx: ICMP is of great value
that I don't contest!
I just don't see value in solely pinging hosts w/ ICMP
for the effort involved, what does it get you? mercutio: it tends to show transit congestion quite easily
shows when people are getting ddos'ed and poor network availability
i use smokeping with small ping size
and curl to some known servers
curl uses lots of bandwidth
but yeah ping can stay the same and curl change jpalmer: brycec: I used to use opsview at my last place. not too long after I left, opsview announced the free edition was going away
brycec: did they change their minds, or are you paying or it, or using an old version? brycec: jpalmer: Last I knew, there was still a free version. you had to jump through a hoop or two (email signup form?) to get it, but it was still there, and... limited. Much more limited than Opsview had been. For awhile, I kept running the older Opsview (pre-rename), but eventually stopped bothering due to the amount of maintenance it needed (it seemed as if a slave was always becoming disconnected)
outweighing my need for it. Those places I did need something like it, I moved to monit (and mmonit)
I will say that Opsview made a real fine GUI for Nagios, and the master+slave nodes feature was a really great one.
(Oh not to mention that it was surprisingly "heavy" resource requirement-wise.)
Hm, I think I heard that BigBrother is still around... ***: gizmoguy has joined #arpnetworks grody: some little wannabe has been slowly port scanning my irc host
i dont know whether to laugh or cry at the sheer stupidity
there is like, one port open on this IP - and his IP got blocked at the 4000th probe
roughly mercutio: there is xymon kellytk: mercutio: Xymon (http://xymon.sourceforge.net/) looks interesting, thanks for the heads up. It seems generalized like collectd. Ideally I would like to find or either custom build a tool which monitors large numbers of hosts from a more rudimentary network perspective
Application layer is beyond the scope of interest as I have it mercutio: why not just use one of the existing monitoring systems
like pingdom etc. kellytk: It's a service I need to have in house brycec: Oh man I had totally forgotten about xymon mercutio: xymon is nice and efficient kellytk: I would be interested to know how it compares to collectd in that regard ***: mnathani_ has joined #arpnetworks
mnathani_ has quit IRC ()