mcbridea

NPM latency measures the ping response time from the NPM server to the managed device. This includes the entire path between these two. Users on a switch are connected directly so their response time will be faster than the NPM latency. Latency over a network is measured in milliseconds; latency to a directly connected…

Not at this point. How are you picturing this working? On a per user basis or change only for this session or...?

Skype has done a good job of keeping its traffic from being identified or controlled by port. They use 443 and 80 for inbound , ports necessary for many other services and a random port per installed user for outbound. I don't think you can narrow down unless you can add those ports and a Skype server IP combination to…

Voip cannot do this now. I have not received other requests for this feature so I don't believe we would add it soon.

Yes, just be sure to choose a new database in the config wizard and give it a unique name. Depending on how many elements you have you might bog down the SQL server. Check the SQL server health first. http://www.solarwinds.com/documentation/Orion/docs/OrionPerformance.pdf

You can deploy the NTA module and get CBQoS stats from any CBQoS interface without enabling NetFlow export. NetFlow can be applied carefully to help you understand the nature of your traffic issues without impacting the traffic. Check out this paper on NetFlow Deployment. Hope this helps - Andy

Hi AJ, Just use the Search NetFlow Endpoint resource. Andy

In the customer portal, under "Regester Your Software", you will find a link to Upgrade Instructions. Download and documents, extract to a local folder and run the Start_Here file. You will find detailed instructions for the upgrade. There is also a video explaining how to properly open and use the linked instruction set…

Have you tried setting them to unmanaged?

Hmmm. If the file transfers are associated with an unwanted app like Bit Torrent you could use NTA to detect Bit Torrent and use an http APM monitor to detect the phrase "Bit Torrent" from the NTA view.

Thanks for the tip - I'll give it a try

In Custom Property Editor, click on Nodes and add ${NodeID} to each Node_ID field. Requires copy/paste and patience as there is no click/drag or paging. Hope this helps!

Hi rucky, Network wide utilization is the average over all circuits being managed. It should never be >100%. In fact, it should never be near 100% at all. There may be some high speed interfaces on your network with NPM using 32 bit counters. This can cause ludicrous results when the counters roll. High speed interfaces…

De-duplication on a NetFlow collector is a tricky thing so we show you all the flow data you collect. Because NetFlow allows you to apply ingress and egress on all interfaces there is good opportunity to over-collect. Are you collecting flow information for accounting or for traffic analysis? Also look for a paper I'm…

Hi Scorpio, Does the Traffic Analysis Event show an error? This can give you the best place to look. Common causes are stuck process or a lost connection to the database. Support can walk you through the fix. Andy

aLTeReGo, Can you clarify this portion of you post "...When I drill down to some of the interfaces this client shouldn't be passing traffic to NTA ..." Andy

Thanks Sean! BTW, do you know what the default time are?

I would ask SolarWinds support to tell you *exactly* what is missing from the switch so that you can work with Cisco on fixing it. It might be worthwhile to discover/add a lab switch that works with UDT and one of the 3750's and do a packet capture from the Orion server and compare the results. This will give yopu concrete…

I found an old thread where a user saw the same thing on 3750s running 12.2. That thread was not resolved but if you have the same IOS that might be the issue. http://thwack.solarwinds.com/message/176516#176516

Hi scriag84, Sounds like the IP address scheme is a mess. If you choose to collect at each spoke it will be easier to segregate that data by accessing reports from the spoke interface views, but you will be adding X amount of data to your WAN to export NetFlow to the hub collector. It might be worth it to try this on one…

Do you mean that you would only want to see CRC errors in the NPM web? If so, here is a CRC error UnDP. From the information you have provided you are dropping inbound packets due to buffer failures. Buffer misses are normal and just mean that a packet had to be handeled by a secondary buffer, but failures = drops. You can…

This should work on the 877s. The 837 will require NetFlow v9 config using ip flow ingress and ip flow egress commands at the (config-if) level. Make sure to configure netflow at the (config) level with Router(config)# ip flow-export version 9 Router(config)# ip flow-export destination NTA server ip address 2055ip…

You could setup a TCP port IP SLA operation without a responder. This requires high level processing. This also will give you true latency from any place on the network you have a Cisco IP SLA capable device. http://www.solarwinds.com/products/orion/ip_sla_monitoring/

At this point that is too many servers to scan in one job. I would break it up into 20 servers max at a time.

You can accomplish that with 2 alerts. One for node A down and one for node A down and exchange down too. For the page you'll need a text gateway like Notepage.

Please open a ticket on this one.

Hi Jonathan, I have not seen this before but you should open a ticket and see if we can find a solution that does not involve you redoing all that work. Andy

Have you been able to follow the examples in the Admin Guide?

Hi John, Ingress/Egress views are on the roadmap for a version due out next year. Andy

Yes, you could remove those. I recommend a couple of things here, * Make sure the changes are consistent with Brandons post on this feature. * Test before you roll out to all alerts