FormerMember ✭✭✭

updated update... now it works. Can't explain it. I tried it again and it worked. Did I make a mistake the first time? Perahps. Maybe the browser didn't update the objects. How do I know? Even if I did make a mistake... don't think I did. But that would have been obvious in v9... instead of being hidden behind a browser…

Is this fixed in SP1? Our Orion takes a good 5 minutes to stop the service. In Task Manager, I watch memory usage slowly tick down from about 218M avg to around 170+/-M and finally, poof, it's gone. This is a longstanding quirk of with the way the NetPerfMonService handles shutdown. We didn't touch that in SP1. Tim, I'm…

Edit: duplicate post.. odd..

My chart gaps (which were few and far between, like days and only for 5-10min on a handful of systems) became much more prevalent with v9, ICMP working well, but any SNMP stats gap for 1-2hrs on most devices. I haven't applied SP1 yet (that will most likely be this evening). For the most part though v9 has helped in a few…

We just captured the INSERTs via SQL Profiler and examined which fields were required and then wrote our own SOAP service to add nodes. With it being highly experimental and not supported (though we've never had issues) I don't want to give advice to do this, but there isn't actually a lot to insert into the DB if you're…

That link is not working.

Thanks, Scott. I'll change it to Megabytes.

this is a core requirement... it should be badged as part of making NPM more scalable... i often get involved in threads where we would be adding performance measures to make NPM better... The problem with not having this, is that every udnp is like another element which can really kill your orion installation. I worry…

What about other folks? Did SP2 makes All Nodes faster or not?

How do you get the switch to send these snmp traps? Is it one of the default ones, or do you need to enable it? snmp-server enable traps mac-notification change mac-address-table notification change int range fast 0/1 - 48 snmp trap mac-notification change added

The reason we didn't support this in the UnDP out of the gate is that it greatly increases the complexity of the tool. We expect to support the table-join scenario in the future, but it needs to be designed so that it doesn't make the UnDP too complicated for the much more common single-table case.

Using Basic Alert solved my issue. Thank you all for your support.

Interesting tidbit. It seems that transfers via Telnet/TFTP, such as a VPN Concentrator 3015, the top of the config look like this: ########################################################## # # # When saving this file from a browser, you must save it # # as a text (.txt) file. Most browsers default to saving # # as an…

It's not a supported configuration, so it's not something we test. Consequently, I don't have a lot of wisdom about why it's not working. In the past, customers who have tried this have run into latency issues, among others. The reason we don't support it is because we have never been able to get it to work consistently. 

That's not the plan. It would be Orion DB at Site A. Orion DB at Site B. New consolidated console would see all of the real-time data from both DBs. 

On mine, which exhibits the same behavior (one of the Fabric Interconnects, along with the blades show N/A) The NPM_UCSFabrics table is missing the IPAddress value for switch-A and shows NULL for the NodeID.

I'm working now. Turned out to be my Kaspersky anti-virus software. I added *.solarwinds.com as a trusted site and pages load just fine.

One last question on this. Now that it's been running for a while, occasionally when it backups up the config on the Open BSD machine, it captures the command it issues to get the config "cat /etc/pf.conf". Is there anyway to prevent that? Thanks 

So when you try to unmanage, what happens? 

This is where we're going. We're introducing more each release and across modules, but it won't be done in Orion 9.

As I said earlier, NTA and VoIP are not licensed by interface. They're licensed according to the size of the network because the value of NetFlow is a function of the network size. And we use the size of your Orion implementation as a proxy (admittedly, an imperfect one) for network size. So you're not paying for more…

New problem... the site comes up but errors when clicking on any node. Files attached. Orion Website Error An error has occurred with the Orion website. Additional Information System.IO.DirectoryNotFoundException: Could not find a part of the path…

Hi Bill, Sorry to hear you're still facing the problem. What version of Windows are you using ? Ours is 2003 Server, looking at the KB article looks like procedurs differ from one OS to another. Regards

It's pretty straightforward and most users do it. You can learn how to from here. www.solarwinds.com/.../wwhelp.htm

Dependencies are a very high priority item. I can't promise a date, but it's hard for me to imagine what might come up that would push dependencies off.

 We're adding a feature in Orion to allow you to set some interfaces to show as "unplugged" instead of "down". Then a node is plugged into a switch, you would see it as up, but when the PC or laptop is removed from the network, the status will show as unplugged, not as down. This solution should allow you to collect…

I found, that switches can transfer multiple mac addresses in one trap. So we have to parse OIDValue. I built a small database application that extracts the MAC addresses from Orion database and store it in a table. There is also a web interface (ASP.Net) to access this data (search for MAC, track MAC, list unknown MACs…

pacetti, this is not the information I am looking for. I want to find, on a node-by-node basis, the poller being used to check CPU & Memory on these ESX servers, so I can see which snmp agent Orion is expecting to find on these boxes. Edit: Brandon, given your latest information, we were able to resolve the issue. Thanks…

Yes I understand that NCM is a module, THAT is part of the problem as not being intuitive. It such a waste of time jumping around from interface to interface and logging into this server and that server to get simple tasks done. We admins have enough on our plate to deal with and SWI DOES not make that any easier. PRTG…

NG, It should work fine.