Thanks to everyone for your suggestions,
unfortunately most of which I had already
either checked or tried.
I filed a case with Tivoli support yesterday
and, at their suggestion, set up a seedfile with
only one router that was not being discovered,
turning discovery OFF, and doing the
'automapgen'. Not surprisingly, netmon found
the one router and kept it. Then (on my own), I
restored the original seedfile (with about 20
routers in it), left discovery OFF, and re-did
the 'automapgen'. Success again, all of the routers
stayed around. Then after I was sure they weren't
going away, I turned discovery back ON, did not do
an 'automapgen', and after a while netmon discovered
all of the other nodes in the network, and kept the
original routers too! So I now have what I wanted.
This a an acceptable work around because, after a few
more rounds of cleanup (I'm trying to get all of my
networks into the location.conf file) which will
involve some 'automapgens', I don't foresee doing
'automapgen' very often. If we do, we'll just have to
remember to turn discovery OFF and then back ON.
BTW, I did do nvTurboDatabase with no apparent
change.
I am pursuing getting NetView Version 6.0.1 and will see if
it exhibits the same behavior. Will let you know via NV-L.
So...all said, all is fine here now!
Thanks again,
Les
>From: lclark@us.ibm.com
>To: "Les Dickert" <lesdickert@hotmail.com>
>Subject: RE: [NV-L] NetView 6.0 Discovery Problem
>Date: Tue, 19 Sep 2000 20:48:52 -0400
>
>
>
>Les, these discussions should all go to the listserver, so everyone
>benefits, and because it is none of our jobs to provide personal support
>for customer. In theory, you customers are supporting each other and
>some of us from Tivoli and IBM hang out and make sure you don't
>steer each other wrong.
>
>That said, I suggest you look for a couple of things. One is your seedfile.
>Make sure you only have one entry in the seedfile for a particular
>device, regardless of how many interfaces it has. Second, make sure
>you don't have any duplicate IP addresses. That can cause this sort
>of behavior. For this, I use the rnetstat command with the -I (capital I)
>flag. Compare the output with your name resolution. Try both the
>host command and the nslookup command to make sure you have
>not 'assigned' an address to a router incorrectly just by naming it
>the same. Third, is it possible that you are dealing with HSRP addresses?
>That is, addresses that move between routers? If so, be sure you
>do NOT resolve those to the same name as the routers. And make
>sure the snmp sysname is unique on those routers. Other
>than HSRP interfaces, you could (and I do) name all addresses on
>a device the same.
>
>All of this could behave differently in V6 than in V5 because there were
>massive code changes in these areas, changes that were needed
>because of frequent bad behavior in V5 on all of these subjects.
>Check into some of these things, and then tell us how you made out,
>on the forum.
>
>No, you should not need Turbo for less than (in my experience) 70
>interfaces. But I have never seen it hurt.
>
>Cordially,
>
>Leslie A. Clark
>IBM Global Services - Systems Mgmt & Networking
>(248) 552-4968 Voicemail, Fax, Pager
>
>
>"Les Dickert" <lesdickert@hotmail.com> on 09/19/2000 09:16:13 AM
>
>To: Leslie Clark/Southfield/IBM@IBMUS
>cc:
>Subject: RE: [NV-L] NetView 6.0 Discovery Problem
>
>
>
>Leslie,
>
>Let me know if I am supposed to direct this
>type of reply back to the NV-L or to you
>directly. Thanks.
>
>Thanks also for the App G reference, I found it.
>
>I checked but failed to mention on my original posting
>that the reverse and forward DNS entries for the addresses
>of the routers in the seedfile appear to be correct and
>consistent. I'll check again.
>
>I also checked the number of interfaces on each of the routers.
>The ones that are not discovered (actually I think they are
>deleted or something) have about 15 each, the one that I mentioned
>that stays around has 6. I know you mentioned a limit of about
>20, but could 15 be causing a problem.
>
>Off to read up on Turbo....
>
>Les
>
>
>
> >From: lclark@US.IBM.COM
> >Reply-To: IBM NetView Discussion <nv-l@tkg.com>
> >To: IBM NetView Discussion <nv-l@tkg.com>
> >Subject: RE: [NV-L] NetView 6.0 Discovery Problem
> >Date: Mon, 18 Sep 2000 21:45:06 -0400
> >
> >
> >
> >The turbo business is covered in Appendix G of the Inst & Config
> >manual for Netview for Unix.
> >
> >Guillermo may be on to something. Poor or incomplete name
> >resolution can cause discovery problems.
> >
> >Cordially,
> >
> >Leslie A. Clark
> >IBM Global Services - Systems Mgmt & Networking
> >Detroit
> >
> >
> >Salinas Sangines Guillermo <gsalinas@scanda.com.mx>@tkg.com on 09/18/2000
> >06:04:26 PM
> >
> >Please respond to IBM NetView Discussion <nv-l@tkg.com>
> >
> >Sent by: owner-nv-l@tkg.com
> >
> >
> >To: "'IBM NetView Discussion'" <nv-l@tkg.com>
> >cc:
> >Subject: RE: [NV-L] NetView 6.0 Discovery Problem
> >
> >
> >
> >We had the same error, and we correct it after various tests.
> >Just disable the DNS resolve method at the AIX server, and use /etc/hosts
> >file.
> >Be shure to DISABLE DNS.
> >
> >Regards
> >
> > > -----Mensaje original-----
> > > De: lesdickert@nsimail.com [SMTP:lesdickert@nsimail.com]
> > > Enviado el: Monday, September 18, 2000 4:13 PM
> > > Para: nv-l@tkg.com
> > > Asunto: [NV-L] NetView 6.0 Discovery Problem
> > >
> > > Running NetView 6.0 under AIX 4.3:
> > >
> > > I have 3 main Cisco routers that are not being
> > > discovered, which is new since converting to
> > > NetView 6.0 from NetView 5.1.1. Some things that
> > > I know are:
> > >
> > > 1. The routers are not in the object database
> > > (according to ovobjprint) after discovery
> > > settles down.
> > >
> > > 2. For a brief few seconds, one of the routers
> > > (and maybe the others) is in the object database,
> > > but not after that.
> > >
> > > 3. The router addresses are in the seedfile.
> > >
> > > 4. I can snmpget and snmpwalk the routers using the
> > > addresses from the seedfile, verifying (I think)
> > > that the entries in ovsnmp.conf are correct.
> > >
> > > 5. I can ping the routers.
> > >
> > > 6. I have been doing nv6000_smit automapgen and ovtopofix -a
> > > many times to diagnose this problem.
> > >
> > > 7. I turned on netmon -M 12 tracing, and it looks like the
> > > routers are returning SNMP data according to entries in
> > > netmon.trace.
> > >
> > > 8. Another nearby router that is similar in function gets
> > > discovered fine.
> > >
> > > I saw a comment (by Leslie) on a previous posting about using the
> > > Turbo form of the databases if routers have more than 20 interface,
> > > however I can't find any reference to 'Turbo' databases in the NetView
> > > for UNIX Installation and Configuration Version 6.0 manual.
> > >
> > > Thanks for any help/suggestions, other than going back to Version
>5.1.1!
> > >
> > > Les Dickert
> > > Network Solutions
> > >
> > >
> > >
> > > -------------------------------------------------
> > > Get personalized e-mail and a web address or your
> > > own free e-mail at http://www.networksolutions.com.
> > >
> > >
> > >
> > >
> > >
> >_________________________________________________________________________
> > > NV-L List information and Archives: http://www.tkg.com/nv-l
> >_________________________________________________________________________
> >NV-L List information and Archives: http://www.tkg.com/nv-l
> >
> >
> >
> >_________________________________________________________________________
> >NV-L List information and Archives: http://www.tkg.com/nv-l
>
>_________________________________________________________________________
>Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.
>
>Share information about yourself, create your own public profile at
>http://profiles.msn.com.
>
>
>
>
_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.
Share information about yourself, create your own public profile at
|