Jan,

>From the looks of your two screen shots, It's not clear if this is just
the traffic between the monitoring system and the specific server or if
this is  all traffic being seen on the network to/from the monitored
server, so I'm not going to hazard or second guess the network guys
there.

Being on an infrastructure team here, we use WUG to monitor the
presence of all of our equipment plus all of the servers (and the
services they are running) on the network. What my peers initially
complained about as objectionable bandwidth utilization by the WUG
system turned out to be unverified and based on past experiences without
actually measuring anything and in reality, ended up being negligable in
the grand scheme of things when actually measured against all traffic on
the backbone portions of the network. What I would suggest, is to ask
specifically how much bandwidth do the network folks actually see being
used by your WUG system and what sort of traffic is using the majority
of it, and that will tell you what you need to re-think to satisfy their
complaints about bandwidth utilization.

I know in our case, when I first got WUG up and running, the SNMP Traps
and generic SYSLOG traffic was really bad- for our Novell servers, the
traffic for 'monitoring' a few servers was overwhelming the SYSLOG
server to the point of being useless in getting the information we were
after and was in fact chewing up huge amounts of bandwidth (25%+) that
just didn't need to be used sending information about litterally every
transaction that the server performed because of the configuration of
what should be trapped. Not exactly what we (both server team and
network teams) were looking to accomplish with network monitoring. We
finaly settled on other solutions for monitoring the log information on
the servers in an 'as needed' basis rather than SYSLOG'g it to a common
remote server. WUG is now our primary alert system to let us know if
something has gone down which was all we really wanted out of our
monitoring system at this time (that and collecting uptime data to prove
we were meeting our SLA's).

------------------------------------------------------------
Bryan Harrell, SPII 
Network Transport & Administration - Tallahassee
Fla. Dept of Revenue
(850)-921-0700  S/C 291-0700
------------------------------------------------------------


>>> [EMAIL PROTECTED] 12/15/2003 3:06:06 AM >>>
Hi all,

We are running WUG since a week or three now.
Last week we stopped the monitoring, because we had complaints from
our
network team (and their superiors) that we were using way to much
bandwith.
We poll every 10 minutes and are checking just a few (Exchange)
services and
an event in the eventlog.

What's the best to tune down the polling, without loosing to much
functionality?

Is there a difference in traffic load between ICMP and "Services
only"?

Thanks,

Jan

Please visit http://www.ipswitch.com/support/mailing-lists.html 
to be removed from this list.

An Archive of this list is available at:
http://www.mail-archive.com/whatsup_forum%40list.ipswitch.com/

Reply via email to