rharland2012

That's a fantastic idea - thanks. I'm going to do so today.

I'll hijack the reports thing, thanks. 1. Change the interface. Use drop-downs or something for condition definitions. Make less typing possible. It just takes. too. long. Better yet, build some functionality into NPM so we can build reports easily from the web console! 2. The report output, with graphs added via kludge or…

bubowski‌ - first, go to your NTA settings page and make sure the following is enabled:

Absolutely I would. I can't imagine I'm the only one.

I've tested with both scenarios to eliminate that as a cause of the problem. Error persists with both states. I also added a separate test node with each management method as well and receive the same error. Strange stuff!

Great post and very informative - thank you! Have you taken a look at a trial of the formerly-Confio DPA? I'd be interested to hear if this product might address some of the challenges you faced. I have no experience of the product but am interested to hear from people who use it.

I think of it like this - I want to see two types of devices. One, the switches my users connect to, so I can easily pull the MACs from polling. Two, the L3 gateways for those switches so I can easily resolve MACs to IPs. Sometimes, the L3 gateway is a switch itself, etc. You can monitor a device without choosing the…

It's not 'automate' oneself out a job - it would be 'outsource' oneself out of a job because perhaps a future-date services provider specializes in real-time traffic flow adjustment in an automated fashion...someone still has to automate the automation.

It usually means that the comment in question is being verified/checked by community managers on the Solarwinds side before allowing publication. This may be because your thread is discussing a significant bug or other reason.

Yes, I have. They are identical. I've opened a case as well...hopefully I can get this resolved. It's very strange.

Additionally, you need to establish an SNMP community, etc. so that your NPM server will be authorized to perform SNMP GETs from the phone.

As long as your NPM server can reach the phone, you should be all set.

Set it to On Demand and check your results. It should perform DNS lookups for IP hosts as needed and will display names where resolvable. Now, NTA won't show you a simple hostname like 'www.google.com', since the Netflow table refers to the resolution of the specific IP to which the client is connecting. You'll likely see…

Can I assume that you're running SAM standalone with no NPM present or integrated?

Currently working through it with support - rerunning config wizards and a complete rebuild of the website have as of yet yielded no results. I'll keep all posted with any updates. I wouldn't call it a showstopper for us. However, I can see in a larger environment (tons of agents) this problem would be more than…

Assuming that you can work with the administrators of said firewalls and get appropriate permissions on the various vCenter entities, then yes.

I believe he is referring to that external server. You would define a group that consists of the external server in question, and then use (or define) groups that encapsulate each of your sites. You can then use the Top XX IP Groups Conversations widget to get a view similar to this (adjust your timeframe to suit):

It refers to the switch. What you're trying to do is set sampling/polling rates for each interface.

If you wanted to be cautious, you could always limit the BW that UDP port 2055 would accept on the off-chance that some malicious fool tried to hammer your pipe/server with bandwidth on that port.....but as jswan ably pointed out, there's precious little for bad people to do with an open UDP port if your perimeter box is…

Nicole, thanks much - that seems to be my takeaway thus far as well. I appreciate the feedback.

Radioteacher​ - care to elaborate?

Thanks - got my support case updated as well. I appreciate the followup.

Thanks, Captain. I'll get on that now.

I hear you - informally polling my peers on thwack seemed a valid way of injecting some time and reflection into the situation before upgrading. I'd love to upgrade, but I wouldn't relish playing a frantic game of 'ROLLBACK' right about now. Nodewise our shop is smaller than most, but we have many modules and I'd like to…

I doubt that - since his results from the CLI on the poller don't agree with the results in the web console, which is pulling data from the poller. However, there could be some arcane network policy forcing the packets from the SW processes through some different path. Highly unlikely, though.

Glad to hear about Dave...

Oh, man.....that doesn't sound fun at all. Thanks for the information. Were you going from 2008 to 2012?

No problem - glad it worked for you.

Got this fixed - full repair of Core Services from Control Panel/Add/Remove Programs did it.

Kudos are well-deserved!