Allow me to steal Francois' thunder on this one since it's
a weekend.
In my shop I use Ping polling for devices which have only
one interface and SNMP for routers and such which have more than one. Ping
polling is simpler because the SNMP poll is a query for the number of interfaces
present and a query for the ADMIN and OPER status of the interface.
This gives a more fine grained status than the Ping
can.
The result of Ping polling is either UP (green) or DOWN
(red). The result of SNMP polling may be UP, DOWN or ADMIN DOWN
(pink). Not a great difference but sometimes useful.
The big difference is in a secure environment where all but
the primary interface on a router is blocked from responding to Ping.
(I.e. Firewalls in our shop.) For Ping polling of multi-interface devices
you also must have every interface defined in the DNS or Hosts file resolving
back to the primary name. Those interfaces are also blocked from SNMP as
well.
As far as timing, either means of polling uses the same
configuration setting.
Francois did not identify what system he is using as the
NetView host. Windows systems default to a 20 minute polling interval and
Unix/Linux systems default to five minutes. This reflects the
capabilities of the underlying operating system. That said, Windows
on a powerful system can easily handle a five minute cycle if the universe of
supported devices isn't too large. 1200 nodes would not be considered a
heavy load. Read the hardware and software prerequisites for both
versions of NetView to get a good feel for the capabilities but keep in mind the
documents were written some time ago and todays medium size systems would have
been brutes back then.
Bill
Evans
Francois,
Just elaborate on your response, does it matter if the Poll is SNMP or
ICMP? Considering we have plenty of resources avaialble on the Server.
Regards,
Usman Taokeer
On 10/17/06, Francois Le
Hir <flehir AT ca.ibm DOT com>
wrote:
It
all depend on the resources of your server, but netview is able to
poll thousands of nodes every 5mn and even at a shorter interval. The way
it does it is that it doesn't wait for an answer before sending pings to
other objects. My server poll over 6000 interfaces every 2mn so if your
server has the resources, I would not worry about netmon not being
multi-threaded.
Salutations, / Regards,
Francois Le
Hir Network Projects & Consulting Services IBM Global Technology
Services
"Usman
Taokeer" <usman.taokeer@gm
ail.com>
To Sent
by: "Tivoli
NetView
Discussions" nv-l-bounces@list
<nv-l AT lists.ca.ibm DOT com> s.ca.ibm.com
cc
Subject
10/17/2006
11:05 [NV-L] Fine
Tunning
Polling AM Intervals!
Please
respond
to
Tivoli
NetView
Discussions
<nv-l AT lists.ca DOT ib
m.com>
Hi
list,
NV 7.1.4 FP04 Windows 2003 Std SP1
We have around 1200
nodes, consisting of Routers, Servers and ATMs, or to elaborate 400
locations, and each location has one (1) Router, One Server and
One ATM respectively.
We want to configure/tune netview in a such a way
that if any one of the node for e.g the "Server" goes down it(netview)
should be able to alert within five(5) minutes. Now as per my
understanding configuring netview to poll 1200 nodes every 5 minutes is not
the answer, there is a high possibility that it could ever poll all 1200
nodes in one go. I also read that "netmon" is not multi-threaded so it
makes it even kinda of worst.
So what would you suggest we configure or
tune netview so that it should be able to alert the operations staff within
5 minutes if any (Server,Router,ATM) goes down. What have been your
experiences or how did u handle these kind of situations in your
experiences? or just a head start for me where shall i look
into.
Thanks in advance!
Regards,
Usman
Taokeer Si3._______________________________________________ NV-L mailing
list NV-L AT lists.ca.ibm DOT com Unsubscribe:NV-L-leave AT lists.ca.ibm DOT com http://lists.ca.ibm.com/mailman/listinfo/nv-l
(Browser access limited to internal IBM'ers
only)
_______________________________________________ NV-L
mailing list NV-L AT lists.ca.ibm DOT com Unsubscribe:NV-L-leave AT lists.ca.ibm DOT com http://lists.ca.ibm.com/mailman/listinfo/nv-l
(Browser access limited to internal IBM'ers only)
-- No virus found in this incoming message. Checked by AVG
Free Edition. Version: 7.1.408 / Virus Database: 268.13.9/490 - Release Date:
10/20/2006
--
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.408 / Virus Database: 268.13.9/490 - Release Date: 10/20/2006
_______________________________________________
NV-L mailing list
NV-L AT lists.ca.ibm DOT com
Unsubscribe:NV-L-leave AT lists.ca.ibm DOT com
http://lists.ca.ibm.com/mailman/listinfo/nv-l (Browser access limited to
internal IBM'ers only)
|