Joe,
Thanks for the info. In response to your comment about users suggesting
use of the 'S' flag, I can only speak from my own experience. I previously
had a PMR re: HSRP devices being deleted under NetView 5.1. They said it
would be fixed in 5.1.1, and for us it was. At that time, Tech Support
told me that the discovery/deletion was due to devices supporting IP
addresses that are not reported in the MIB (ip.ipAddrTable). This
corresponds to the description of the 'S' flag in the oid_to_type file.
(As an aside, use of the 'S' flag also cured HSRP discovery/map problems
for us previously on NNM.)
If Tivoli Tech Support says the flag should NOT be used for HSRP
interfaces, I'd like to know under what circumstances it should be used?
The approach we're taking is to monitor the actual interfaces and to watch
for HSRP traps to know when a failover of the standby address has occurred.
I'd like to know more about your plans for monitoring the standby address,
and any advantages you see.
I am also interested to hear what you find out about representation of the
virtual addresses on the map.
Karin
----------
> From: Prokott, Joe <Joe.Prokott@WESTGROUP.COM>
> To: NV-L@UCSBVM.UCSB.EDU
> Subject: Re: Fw: Cisco Cat5500-RSM
> Date: Wednesday, June 23, 1999 5:23 PM
>
> I have a PMR opened for the same issue. I too am querying a node with
the
> same OID you state below. Below is what I received so far from official
NV
> support:
>
> "You do not want to have the S flag set in the oid_to_type file for
> a HSRP device. This can create problems with NetView handling HSRP (per
> Mike P.).
> DNS name resolution is not required for HSRP support. Without name
> resolution,
> the selection name will default to the ip address instead of hostname."
>
> It is strange that many users have stated to put the "S" flag in the
> oid_to_type file, yet NV support says not to for HSRP discovery.
>
> I have not had any luck in getting the HSRP address to show yet either.
> I'll post something as I get more info.
>
> Joe Prokott - West Group
> Network Architect
> 610 Opperman Drive
> St. Paul, MN 55123
> Phone: 651-687-4536
> Fax: 651-687-6946
> E-mail: joe.prokott@westgroup.com
>
>
>
> -----Original Message-----
> From: Gil Irizarry [mailto:IRIZARRG@LABS.WYETH.COM]
> Sent: Wednesday, June 23, 1999 4:05 PM
> To: NV-L@UCSBVM.UCSB.EDU
> Subject: Re: Fw: Cisco Cat5500-RSM
>
>
> Karin:
> Thank you for the response. I removed the external DNS lookup and the
> problem went away.
> What should the oid_to_type entry look like? And do I have to modify any
> other members?
> In Netview the OID associated with the RSM is 1.3.6.1.4.1.9.1.168. Is
this
> correct for HSRP?
> Thanks in advance
> Gil Irizarry
>
> >>> Karin Binder <karin.binder@nwa.com> 06/23 12:09 PM >>>
> Hello Gil,
>
> I saw similar symptoms on the symbols for our Cisco routers running HSRP
> (Hot Standby Routing Protocol) when running NetView 5.1 on AIX 4.3.2.
I'm
> not sure what version you're running. Our problems were fixed in NetView
> 5.1.1.
>
> Other thoughts:
>
> 1) oid_to_type flags: Do you have the 'S' flag set for devices using
HSRP?
> 2) DNS: are the DNS forward/reverse definitions using the HSRP address?
> This could confuse discovery/config checks.
>
>
> ----------
> > From: Gil Irizarry <IRIZARRG@LABS.WYETH.COM>
> > To: NV-L@UCSBVM.UCSB.EDU
> > Subject: Cisco Cat5500-RSM
> > Date: Wednesday, June 23, 1999 10:47 AM
> >
> > Hi All:
> > I have 2 Cisco Cat 5500 with RSM cards & variable net mask
> 255.255.252.0. NETMON discovers the IP addresses, ok. The Event display
> window shows Added and 10 minutes later it shows Deleted for the same
> interfaces. This 2 switches are up all the time and function as backup to
> eachother(hot stand by). We are also using vlans.This problem was NOT
there
> on NV 4.2.1.
> > Any ideas?
> > Gil Irizarry
> >
> > Gil Irizarry
> > Wyeth-Ayerst Pharmaceuticals
> > 280 King of Prussia RD.
> > Radnor PA.
> > 19087
> > Internet:irizarrg@labs.wyeth.com
|