nv-l
[Top] [All Lists]

Re: CISCO and "no keepalive"

To: nv-l@lists.tivoli.com
Subject: Re: CISCO and "no keepalive"
From: Jeff Fitzwater <jfitz@Princeton.EDU>
Date: Wed, 01 Aug 2001 15:36:37 -0400
Leslie Clark wrote:

> Being ignorant on the subject of router configuration, I am curious about
> this 'no keepalive' configuration. Does this mean that you want the
> interface to go down when it is not in use? Or that you do not want a trap
> from the router when it does go down? What is it for?
>
> If I wanted Netview to ignore the fact that an interface is
> up/down/up/down,
> then I would just unmanage it.  If that is what is going on...
>
> Cordially,
>
> Leslie A. Clark
> IBM Global Services - Systems Mgmt & Networking
> Detroit
>
> Jeff Fitzwater <jfitz@Princeton.EDU>@tkg.com on 08/01/2001 01:42:04 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] CISCO and "no keepalive"
>
> Mercier Alain wrote:
>
> > > -----Message d'origine-----
> > > De:   Jeff Fitzwater [SMTP:jfitz@Princeton.EDU]
> > > Date: mercredi 1 août 2001 17:32
> > > À:    IBM NetView Discussion
> > > Objet:        Re: [NV-L] CISCO and "no keepalive"
> > >
> > > Mercier Alain wrote:
> > >
> > > > the router is configured to send trap to the NetView station, but how
> > > can I
> > > > determine the source of the events ?
> > > >
> > > > > -----Message d'origine-----
> > > > > De:   Jeff Fitzwater [SMTP:jfitz@Princeton.EDU]
> > > > > Date: mercredi 1 août 2001 15:33
> > > > > À:    IBM NetView Discussion
> > > > > Objet:        Re: [NV-L] CISCO and "no keepalive"
> > > > >
> > > > > Mercier Alain wrote:
> > > > >
> > > > > > Hi list,
> > > > > >
> > > > > > Config :
> > > > > > AIX 4.3.3
> > > > > > NV 6.0.2
> > > > > >
> > > > > > I've some CISCO routers with interfaces configured with the "no
> > > > > keepalive"
> > > > > > parameter.
> > > > > > In NetView, log is full of alert from these routers :
> > > > > > - <ROUTER_NAME> Interface <INTERFACE_NAME> is down
> > > > > > - <ROUTER_NAME> Node is Down
> > > > > > and 1 or 2 seconds after :
> > > > > > - <ROUTER_NAME> Interface <INTERFACE_NAME> is up
> > > > > >
> > > > > > How can I fix it in NetView ? with a filter ?
> > > > > >
> > > > > > Thanks,
> > > > > >
> > > > > > Alain
> > > > > >
> > > > >
> > >
> _________________________________________________________________________
> > > > > > NV-L List information and Archives: http://www.tkg.com/nv-l
> > > > >
> > > > > Is the source of the event Netview or the router?   If you have set
> > > the
> > > > > interface on the router to "no Keepalive" then you should never get
> an
> > > > > interface down from the router.   If the source is netview then the
> > > > > problem
> > > > > is either a ping timeout too short or routing problem from the
> point
> > > of
> > > > > view
> > > > > of your Netview station.  Try just pinging it from the command line
> > > and
> > > > > see
> > > > > it the ping fails.   Also check that your routing path is not
> > > flapping.
> > > > >
> > > > > First check the source of the trap.
> > > > >
> > > > > Jeff Fitzwater
> > > > > CIT Systems & Networking
> > > > > Princeton University
> > > > >
> > > > >
> > >
> _________________________________________________________________________
> > > > > NV-L List information and Archives: http://www.tkg.com/nv-l
> > > >
> > >
> _________________________________________________________________________
> > > > NV-L List information and Archives: http://www.tkg.com/nv-l
> > >
> > > In event window, if you are using lists, you should see an "A" after
> the
> > > node
> > > name if the event is from the node itself (A meaning Agent) or an "N"
> > > meaning
> > > source is one of Netview processes.  You can also view the event as a
> card
> > > and
> > > see the SOURCE as an item in the card.
> > >
> > > Hope this helps.
> > >
> > > Jeff F
> > >
> >         [Alain MERCIER]  The source of the events is NetView (Netmon).
> >         Ping from Netview station (command line) is OK.
> >         Where can I configue the ping timeout in NetView ?
> >
> > _________________________________________________________________________
> > NV-L List information and Archives: http://www.tkg.com/nv-l
>
> If it is a timeout it can be set via the OPTION menu from any map view.
>
> Under the OPTION menu select SNMP CONFIGURATION.
>
> NOTE if do not have an individual entry for that router and you only want
> to
> change its' timeout, then you must add one to the list.
>
> To add an entry, select the  "Global Default" section on the SNMP
> CONFIGURATION
> page  and you should see all the DEFAULT SNMP settings come up below.
>
> Now change the TARGET to the IP  of the router interface in question and
> also
> change the timeout value to something a little longer in seconds.
>
> As you can see you can change other parameters as you wish for only this
> device
> if needed.
>
> When you are done making changes , select ADD button on right side.
> Now select APPLY button at bottom of page.
> You should now see an entry for that router interface in top portion of
> page.
> You can now enter OK or CANCEL at bottom to exit.
>
> One last thing to do since you made an SNMP config change.  All the SNMP
> values
> for all devices are CACHED and therefor must be refreshed to aquire the new
> settings.  The following will accomplish this..
>
> I assume you are on a UNIX box as Netview admin...
>
> Enter the following at the UNIX command prompt ...
>
> xnmsnmpconf -clearCache
>
> Now the new settings will take effect.
>
> Remember I am only taking a good guess at your problem.  There are many
> other
> factors that could cause the problem, but the root is that Netmon does not
> get
> a ping response in time so he thinks the interface is down.
>
> Jeff F.
>
> _________________________________________________________________________
> NV-L List information and Archives: http://www.tkg.com/nv-l
>
> _________________________________________________________________________
> NV-L List information and Archives: http://www.tkg.com/nv-l

Hi Leslie;

As far as CISCO keepalive goes, it is an interface command that is on by
default.  It sends a KIND OF ping to the interface to check the link state.
If no link, the keepalive fails and the interface goes down and the router
sends a trap.   Now if we want to ignor the interface link and bring up or
keep the interface up even if there is nothing connected to it,  we set the
KEEPALIVE to off or NO KEEPALIVE on that interface.   So if you ping that
interface IP via another interface on the router it responds.  Also the router
will advertize the subnet since it is up.

Glad to help.

Jeff F.


<Prev in Thread] Current Thread [Next in Thread>

Archive operated by Skills 1st Ltd

See also: The NetView Web