Here's some info on that topic from an older append. It will probably only
confuse things more. How frequently is this occurring? It would seem
reasonable that it should occur with netmon's regular configuration poll,
which by default occurs once per day and then again at startup of netmon.
Check the frequency of your configuratio polling in Options...SNMP...
==========================repost
===========================================
To: NV-L@UCSBVM.UCSB.EDU
cc: (bcc: Leslie Clark/Southfield/IBM)
Subject: Re: AW: polling problem
We ran into the same problem with NetView 4.1.2. What we discovered is the
fact that NetView does not treat nodes the same between up and down polling
cycles. We used a sniffer, and we observed that if a non-snmp node is
known to be Up by NetView, it would send an ICMP C Get Address Mask. If the
same node was known to be Down, it would send a ICMP Echo only.
Some devices on the network would not respond to the ICMP C Get Address
Mask , so NetView flagged it as being down. On the next polling cycle, the
ICMP Echo was sent and the device was back up.
The problem was solved by using an (as far as I know) undocumented daemon
option for Netmon. We added a "-x" to the /usr/OV/lrf/netmon.lrf file on
the NetView server. This option told the Netmon daemon not to perform the
ICMP C Get Address Mask request to non-snmp devices when attempting to get
configuration information. We had to work this one out with the support
center.
Don't know if this is the exact problem you are seeing, but I suspect it is
close.
Mearl Danner <jmdanner@SAMFORD.EDU> on 01/13/99 04:16:12 PM
Please respond to Discussion of IBM NetView and POLYCENTER Manager on
NetView <NV-L@UCSBVM.UCSB.EDU>
To: NV-L@UCSBVM.UCSB.EDU
cc: (bcc: Dean Grant/AIS/US/Ahold)
Subject: Re: AW: polling problem
We use NV6000 4.1.2 on an RS6000 running AIX 4.2.1.
We have been having similar problems with our network also.
It apparently was due to some problems with the bridging
tables in our 3Com 2500 layer 3 switches. Since we upgraded
the software in the 2500s to the latest release (8.3.1.3),
I haven't had a node down event I couldn't reconcile to an
obvious hardware problem.
You didn't specify what hardware you are pinging through,
but this is one possibility. It also seems that the pinging
that Netview does operates under different rules than the
command line ping. I can never see any problems using the
command line ping or the pulldown from the GUI. The device
always came back up on the next ping cycle.
On Wed, 13 Jan 1999 09:47:42 -0600 Stephane Robertson
<Stephane.Robertson@HEC.CA> wrote:
> > I don4t know where this problem belongs to, so I post it here and hope
that
> > it is known by someone:
> > We4re using NK 5.1 on AIX4.2.1. From time to time I see Interface down
and
> > node down events from active stations. When looking on the network I
see an
> > ARP request to the station, which isn4t answered. When I ping these
> > stations from command line, Interface and node up events are shown.
> > I know, this seems to be a network problem, but there was never a
"manual"
> > ping which wasn4t answered. Only the pings from NetView fail.
> > Thanks in advance for any help!
> >
> > Mario
>
> Mario,
>
> Do you an ATM Network?
> Did you see the ping request on the server side?
>
> I had the same problem and I fixed it by putting very small timeout and
many
> retries.
> Since I changed to that setting I don't have any problem but I don't
think the
> problem
> is really fixed. Can you send me what kind of device you have to compare
with
> my network because I also suspect something on the network. May be we
have
> something
> in common that could point us to the solution.
>
> SR
>
> --
> Stephane Robertson
> Ecole des Hautes Etudes commerciales
> 3000 chemin de la cote-sainte-catherine
> Montreal, Quebec
> H3T 2A7
> E-Mail: Stephane.Robertson@hec.ca
> Phone : 514-340-6560
> Fax : 514-340-5637
>
>
---------------------------------------------------------------------------
--
> Windows 98 supports real multitasking - it can boot and crash
simultaneously.
>
---------------------------------------------------------------------------
--
-----------------------------------------
Mearl Danner
Network Engineer
Email: jmdanner@samford.edu
Samford University
=============================end re-post===============================
Cordially,
Leslie A. Clark
IBM Global Services - Systems Mgmt & Networking
Detroit
Geoffrey Cheng/Markham/IBM@IBMCA@tkg.com on 07/03/2001 04:38:46 PM
Please respond to IBM NetView Discussion <nv-l@tkg.com>
Sent by: owner-nv-l@tkg.com
To: nv-l@tkg.com
cc:
Subject: [NV-L] netmon sends mask request
My NetView is configured to monitor ip resources in several networks
isolated by firewall. Only
snmpget/set and trap are allowed going through the firewall. I configured
MLMs on the other side of
the firewall to perform status polling for configured interfaces (defined
in MLM domain
smartsets) for NetView. The status information is propagated correctly.
According to manual, if I move interfaces into domain smartset for the
running MLM, netmon should
withdraw status polling of those interfaces and deligate the job to MLM.
While the status polling job
did deligate to MLM in my case, however, by monitoring the firewall log, I
found out that netmon is still
doing the pings.(of cause all being dropped by firewall). And in netmon
trace, it seems that netmon
is doing "sending mask request to <interface_ip_address>" quite frequently.
The communiation between
NetView and MLM is fine all the time.
I disabled the discovery and configuration polling globally, but netmon is
still doing the pings...
Any idea? I am on NetView for Solaris 6.0.2.
Thanks.
Best Regards,
Geoffrey Cheng
Email: gcheng@ca.ibm.com
_________________________________________________________________________
NV-L List information and Archives: http://www.tkg.com/nv-l
|