Thanks John.
Stephanie
--- On Mon, 9/13/10, John M Gatrell <John.Gatrell@uk.ibm.com> wrote:
From: John M Gatrell <John.Gatrell@uk.ibm.com> Subject: Re: [NV-L] Slow Response from nmdemandpoll To: "Tivoli NetView Discussions" <nv-l@lists.ca.ibm.com> Date: Monday, September 13, 2010, 6:18 AM
Try the following. Choose just one IP address (the most reliable or the nearest) to do snmp polls of a router (A loopback address if Cisco). Inform NetView of the HSRP address with the percent character. So in the seed file I would have the following %10.222.1.1 #Router1 and Router2 hsrp address 10.222.1.10 #Router1 Discover router =10.222.1.10 #Router1 Lock router poll address $10.222.1.10 #Router1 Prefer snmp to ping Alternatively, make sure router1 and router2 have same SNMP community string and use the '-V'
option that is described in 'man netmon'. This will involve editing the netmon.lrf file and running 'updatelfr netmon' While I think of it, might be worth making sure the HSRP addresses respond with the same community string, both on the routers and in NetView's config. reset_ci is to be run if you change the hostname or the IP address, or hardware of the NetView box itself. So reset_ci is probably not relevant to yourself. It should not be necessary to run demand polls regularly. I only run a demand poll if I suspect NetView does not know the latest config change of a router. John Gatrell We're
getting delayed responses when running demand poll against devices. What's the best way to troubleshoot this issue? We looked in netmon.trace and saw false positive DUP_ADDR errors on HSRP addresses. In what case would we use reset_ci? Would we use it to resolve the slowness with demand poll? Stephanie
Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU
-----Inline Attachment Follows-----
|