[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [hobbit] Bandwidth monitoring

On Wed, 2008-09-10 at 08:04 -0500, Hubbard, Greg L wrote:
> How about devmon?  

Devmon doesn't support snmp V3, which is a requirement for me.  MRTG
does.

> It requires much less care and feeding than MRTG if you are concerned about the entire device 

mrtg supports templates just like devmon does.  Once I have the
templates built, it works fine.  The only big issue is that you have to
read the rrd files twice, and that requires really fast hardware when
you have several thousand points to monitor.  we had to upgrade this
year when we had over 5500 points on snmp v3.  Currently I'm monitoring
6502 snmp v3 targets and 1921 snmp v1/2 targets, and feeding it all into
Hobbit.  With the new hardware, our load average: 0.49, 0.70, 0.79

> -- if you only care about certain interfaces than MRTG is probably a
> better fit.  However, devmon only supports a few (though common!)
> device types, so you might also be better off with MRTG if you are
> using "unusual" equipment.
> 
> GLH 
> 
> -----Original Message-----
> From: McDonald, Dan [mailto:Dan.McDonald (at) austinenergy.com] 
> Sent: Wednesday, September 10, 2008 7:48 AM
> To: hobbit (at) hswn.dk
> Subject: Re: [hobbit] Bandwidth monitoring
> 
> On Wed, 2008-09-10 at 11:04 +0200, pkc_mls wrote:
> > Stef Coene a écrit :
> > > On Wednesday 10 September 2008, DNS wrote:
> > >   
> > >> Hi All,
> > >>
> > >> Do anyone know (have server-side scripts) how to centraly monitor 
> > >> Bandwidth usage of switches/servers/router etc.? If so how do I 
> > >> configure this step-by-step in Hobbit?
> > >>     
> > > I use mrtg.  It can query via snmp routers and servers and store the 
> > > data in a rrd file (just like hobbit).  I never tried to integrade 
> > > the information in hobbit, but I'm sure it can be done.
> > >
> > >   
> > the integration is described in here :
> > http://www.hswn.dk/hobbit/help/hobbit-mrtg.html
> > 
> > this will simply add the graphs in the trend columns (by default) for 
> > the client.
> 
> 
> Or, you can run bbmrtg.pl and have it generate alerts.
> 
> --

-- 
Daniel J McDonald, CCIE #2495, CISSP #78281, CNX
Austin Energy
http://www.austinenergy.com