GregD

Yeah, that was going to be my next step this AM :-) I've opened case 107932 for this issue.

 Yep, a year. We'll eventually keep 3 years - it's being driven by upper management and the new corporate overlords. They say they need 3 "cycles" worth of data to accurately see trends. So now it's up to us to hold onto all the data - and ever so much fun for network operations :) We poll every 5 minutes, and do stats…

I'm seeing this same problem with Engineer's Edition 10.2 and NPM 9.5 SP2. Any idea on what I can do to correct it?

 If by maintenance interval, you mean "unmanage", then yes, you can unmanage and re-manage via the web console. The new custom poller is definitely easier, but you still have to know OIDs or MIBs. I'm not crazy about that fact, but that's the nature of SNMP. This sounds great! I know V9 is in beta now, any idea (ballpark)…

 I know, I'm digging this post up from the deep freeze :-) We are about to expand our data retention policies from 7 days of detailed statistics to 1 year of deatiled statistics (for about 1500 routers). We're also going to start running a lot more reports, using a lot more of the data - like, "What are my most utilized…

 You mentioned that, and I went back into the View manager and noticed there were two options for web reports. I chose the other one and now it works. Hooray for n00bs! :)

The only issue with that is that I'm running SQL 2000 and going straignt to 2008 because of time and bugetary concerns. If I was able to make a stop at SQL 2005 along the way I would upgrade to 2005, upgrade to 9.5, then move to 2008. Right now my plan of attack is to shut down the entire Orion system first thing in the…

 But that stuff never goes down, right? RIGHT?? :-P Seriously, I've given thought to implementing something like this myself, even if it's a dial-up modem sending text messages through the provider. Hopefully someone will come along with some answers soon.

 I know, I was having a bit of fun. I got the distinction in your titles :-)

X2 - I'd like to see the config. Are you basing the "flap" on getting X number of "Device/Interface up" traps in a specified time period?

Thanks for that :-) It answers half my question, I suppose. I guess I would do this procedure to move the DB, then install the new 9.5 pollers on the new VMs, and then run Poller Load Balancing to move the nodes to the new pollers. I guess the only question that remains is how to get the references to the old pollers out…

That's a decent idea, you can use the report scheduler to pull any web page and email it to you... 

I had seen that, and had gone into the "poller load balancing" option in that program, but the only columns that were available to me were Engine, Caption, and IP_Address. I couldn't find a way to add my custom property. What I ended up doing was this: 1.) Shut down all the polling engines. 2.) In Query Analyzer, run this…

 Hmmm... dueling developers :) I'll play it safe and wait. They aren't in production now anyway, so if I take an extra day to get them up and running it shouldn't be a big deal. Thanks for the quick replies!!

 We have it set up to send to a standard email box in our corporate email system, which then has a rule to forward it to the provider's email-to-sms gateway for the phone. That way you can change the address in a single location, and you have the added benefit of an "audit log" of sorts.

No, if I add that filter, I get nothing in the "nodes with problems" list. I would imagine I need to find a way to make a custom module (if that's the correct terminology) to display in the view.

Excellent, thanks! I'll apply that Tuesday morning, then. 

 Thanks. I contacted customer support, who gave me another download entry for the Orion additional web server *smacks forehead* so I've opened a support ticket and referenced this thread. :)

 Thanks for the reply. I found that posting, but I still can't get the disk space to show up. I can get everything but the disk, it seems. If I dig out the OID for the NetWare disk space, it gathers the information properly.

 Additionally, I've noticed that when the web reports stop running I lose the response time/packet loss and CPU load graphs on the node page. Curiouser and curiouser...

 Come on, I *know* someone has to know *something* :-)

You could write a custom SQL report based on the Accounts table in the NetPerfMon database. I'm covered up at the moment, but if you need help just post back and I'll throw together a quick SQL query that you should be able to turn into a report. 

I'm also getting messages like the following: "Error "No Such Name " occurred during polling VoIP Call Path AL Decatur Bankcard GW01->AL Brock GW01 (simulation node aldcbcgw01 [172.20.131.10]) " Anyone? Anyone? Beuller? :-)

 We are all MPLS to all locations. Admittedly, we're just starting our VOIP implementation, but I don't see moving to another environment anytime soon :)