Does Prime, any variant, do bandwidth monitoring?

[nellson]

I have been looking at Prime Infrastructure, trying to eliminate a mix of Entuity's Eye of the Storm and BMC's Network Automation. The LMS bit of CPI seems to cover what BMC does, with out any gaps. But my manager was interested in the reports on bandwidth utilization of key links. Entuity's Eye (Or NetScout, SolarWinds NPM, and such) would let you pick an interface or two and show the IN/OUT stats. From what I can tell, CPI is primarily aimed at TOP-N based statistics in a more general sense. Does anyone use a variant of Prime to watch traffic stats, and if so, what modules sould I be looking at?

Nick

[Marvin Rhoads]

It's under performance reports.

See here for details:

http://www.cisco.com/en/US/customer/docs/net_mgmt/ciscoworks_lan_management_solution/4.2/user/guide/reports/Perf.html#wp1359365

[Captain80]

You are making reference to the LAN Management flavour of the Cisco Prime Suite. Cisco Prime Infrastructure does not have this report from what I can see, it's all focused on the TopN as Nellson mentioned.

[Karl Schuman]

Hi Nick,

Prime Infrastructure includes Assurance Manager which can act as a NetFlow collector to gather and report on interface utilization and other pertinent statistics. It also can act as the GUI for multiple NAM modules and do event corelation.

Karl

[Marvin Rhoads]

Another reason to stick with Prime LMS for most non-wireless management for now. Assurance requires a separate license, an end device that supports Netflow, setting up exporters, etc.

Why in the world would PI not have a simple report you could generate based on the standard SNMP ifInOctets / if OutOctets MIB variables like every other network management system built in the past 20 years?!

[gary.gatten]

An old one, but I am just digging into PI 3.1.7 and am disappointed in a number of things.  It seems like whomever was responsible for the development and features is not familiar with what info / views a network admin would use daily, weekly, monthly, etc.  Right now I am trying to see what host is using all my bandwidth and what apps / protocols they are using and who they are talking to.  This is a very basic function you generally need several times a day.  I can get this for free with "ntop" and netflow, why when I pay for PI can't I see this?  Maybe it's hiding somewhere - I have only looked for 3 hours.  I will look a bit more and open a case with TAC.

[paulpi-cn6300]

To add to this, I'd like to see prime notify on high input utilization.  I get high output utilization emails, but none for high input utilization.  Prime LMS 3.1.  

[delsenyj]

Ok, I am glad to see that I am not the only one in this boat! (but saddened at the same time)

I'm using PI 3.3, and have been pulling my hair out trying to find some way to have it produce some fairly simple bandwidth usage reports/details (the same as everyone else here).  It took me a while to just find exactly where I can pull some netflow reports from, which looks like they are just raw data reports over a given timeline.  Great, if I want to import that into something else and drill down a little further, but I was hoping that PI was going to provide that capability itself. 

To add to the frustration, with all the playing around with PI that I've been doing, I started to notice that a lot of the traffic volumes that PI does report on the dashlets, are not accurate.  I've been comparing the PI volumes to those taken directly from our routers CLI, webui, and PRTG (just to confirm my findings with an alternate reporting system).  All numbers from the router CLI, webui, and PRTG, match exactly.  The PI numbers, do not.  So now I'm finding myself questioning and checking everything.  The last thing I need is to provide senior management with some numbers or reports, that are actually inaccurate!

I'll continue using PI for some of the other features, but hopefully Cisco will update it down the road, to include some of the more commonly used, basic bandwidth monitoring & reporting features and tools, because the price point for what PI has the potential of doing is great.  For now, I guess that I will just go back to using some of the other systems (ntop, prtg, plixer, solarwinds, etc.) for my usage monitoring & reporting.

[david.stanley@amerisbank.com]

Has anyone found out if any of the newer releases offer this kind of bandwidth utilization monitoring, and alerting when certain thresholds are breached? I am in a similar boat with wanting a one-stop shop for bandwidth utilization and alerting. We currently use Plixer Scrutinzer and are in the process of trying to eliminate the need for multiple tools that can perform the same function. Any help is greatly appreciated.

[gary.gatten]

PRIME definitely does bandwidth monitoring and alerting. Setting it up is far from intuitive IMHO, but it's there and seems to work. It's lacking some details re conversation level info from netflow - or at least I haven't found it yet. But, simple stuff like "this interface is < 90% for 5 minutes" - yes - PRIME does that.
IIRC, Scrutenizer does other stuff as well - more focused on security isn't it? Send it a bunch of data using netflow and it acts as an IDS?

[david.stanley@amerisbank.com]

Correct, Plixer is much more than a bandwidth monitor, however that's primarily what we use it for here. So we're looking to eliminate the need to have multiple software services doing the same thing. And really our need for bandwidth monitoring is pretty simple. We really just want to setup alerts at a certain percentage of utilization and then be able to go back and look at the netflow information around that time...nothing fancy.

[gary.gatten]

Cool - here's what I found. Here's a typical use case: My circuit is maxed out - who is doing what right now - who's using all my bandwidth and what application(s) are they using?
This info is difficult at best and perhaps impossible. If it is possible, hopefully someone will enlighten me. PRIME will show interface util, QoS info, and "application" usage info. But, let's say you have CIFS using all your bandwidth; drilling down to see what specific hosts are the top users of CIFS is where I run into issues. I am far from a PRIME expert, but I have not been able to figure out how to do this. I have worked with other tools such as nTop and Netscout nGenius which easily present this info.
One option is to enable the top-talkers and other statistics in your routers / switches doing netflow. Then, even if PRIME doesn't present the info, you can hop on the router and look at the stats there.
PRIME does a lot of cool stuff, but if you are wanting "basic" bandwidth and conversation level traffic stats, IMHO it's not the best choice for just that.

[david.stanley@amerisbank.com]

Thanks for the quick response Gary. Seems like Prime might not be the best option for netflow analytics in that case. Any idea if they are going to be improving this feature set in upcoming releases?

[gary.gatten]

No idea. Please don't take my word for it. As mentioned I am far from an expert. The interface is not very intuitive, so maybe this data is available somewhere and I just can't find it. Maybe open a TAC case and see what they have to say?
Either way, if your needs are basic I think PRIME is too much: It's a big app with relatively high resource requirements. Maybe take a look at the features and use cases and see if it's worth it? No sense paying for 50 features when you will only use 2.

[david.stanley@amerisbank.com]

Okay, thanks Gary. Will do. We are currently using Prime for a lot of other functionality as well, so I'd just like to take advantage of its netflow and bandwidth utilization reporting as well. Thanks again.