mcbridea

Progressive uses a different x axis data period. (3 days vs 1 day)

I have been away from SolarWinds products for about 2 years but we just bought NPM so I'm back. I have had 4 issues that required support calls so far. One of the issues was handled very quickly by an Austin support associate. Twice I waited on hold until I had to hang up and abandon the call. I'm on hold now for the forth…

A couple of questions. Are the routers inside or outside the MPLS network (PE or CE routers). Do you have ingress and egress NetFlow data exporters set up on all 3 routers?

Great stuff! Thanks for the contribution.

There are probably some settings we can assist you with but the best way to proceed on this one would be to open a ticket.

You could do so for 30 day eval licences of each and that will give you 30 days to get the existing NPM up to snuff. Be aware that merging the data between the two seperate servers will not happen once you go to production with IPAM and IP SLAM.

If you open Advanced Alert Manager you will probably find an active response time alert that you can edit. Here is a paper on using Advanced Alert Manager

Hi Jon, Can you open a ticket on this one? Andy

If it has an SNMP agent we can monitor it using Universal Device Pollers.

As of now were investigating all the different flavors of CCM and determining support. Some versions just don't have very much SNMP support and so are not manageable. I don't know if Unix CCM 6.1 will be supported at this time.

Hi Tony, The example on the demo is intended to demonstrate the type of integration that is possible, based on the abilities of you ticket system. This can normally be accomplished by adding a custom HTML resource to the view you want to it to appear on. Your ticket system admin can probably show you what HTML pages the…

Have you opened a ticket for this? It may be a bug.

Look at the report time frame. If you need a report for a static time frame you need to specify it. Otherwise it will update.

I would check to see that all the services are running first then run configuration wizard.

What version of ESX are you using?

Hi Jeff, It's one of the mechanisms SP1 uses to keep the server from overloading. Andy

cs3 = 96 af31 = 104 ef = 184

Would a report do the trick? There is a reporrt on Content exchange that shows which Cisco devices rebooted, when and why. You could place the report on you Summary web page.

I am assuming you are using NPM. In advanced alerts you can set a timer for how nolg to wait before triggering. For status polling the default poll cycle is 120 sec. You can set the trigger delay to avoid alerting for shrt outages. Another feature that may be useful is grouping. Here is an example of how to do this with a…

I'd say give support a call on that one. They'll be able to find the issue. Andy

Are you seeing the thin AP as down in NPM? What type of AP is this?

I think it is a timing thing so will happen when the timing is right. You could add a reset condition for the primary switch or firewall equal to up. Then it would reset and not trigger the dependent devices.

I don't believe that you can alter that rounding. I usually don't report availability at any more than four nines. The issue is that reporting to six or seven nines creates an expectation that the monitoring will detect very short outages on a per device basis.

Take a look at the NetFlow settings. If you have it set to persistent, then it will resolve addresses in the background. If it is set to on demand, then it will resolve when you access a resource.

Thanks for the feedback. I can see how a global threshold would be a problem there. I'll submit that as a feature enhancement. Andy

VMware has changed SNMP in v4. See the below. http://www.vmware.com/pdf/vsp_4_snmp_config.pdf

Have you tried the NetFlow real time tool?

I'm collecting Feature Requests and will note this one. Thanks - Andy

Hi Joey, I can help out with some guidelines and experience. 1. How much bandwidth does Netflow use to collect data from remote routers/switches? Netflow admin guide does not give much detail on how much bandwidth and min/max bandwidth recommandation. I believe Cisco says that v9 and v5 exporters will use about 1.5% to 3%…

You should add the RAM requirements but 6G of CPU is probably a bit much. I'd try 3G. You should use SQL on a separate non-virtual server. If you use SQL on a VM it will be unbearably slow. I have heard there is better performance with Server 2008 but have not seen this myself. Andy