automag928 ✭✭✭✭✭

peter.ksenzsigh‌ - I'm glad I could help, and appreciate you guys jumping on this. My thwack points bank is wide open for deposit !!! lol

Ok - trying to help you here - Did you look ConfigMgmt DB itself, using SQL mgmt studio or even the database manager built into SW, look through the tables, one is labeled NCM_NCMJobs. Query that table and look for your NCM Job name, scroll over to the enabled column and does it say 1 or 0? Try re-running the NCM…

fahadahmedsyed‌ - The robustness of your platform determines the robustness needed for your sql server. But one correction to what silverbacksays, the NTA db server is not based on SQL, but a flat file proprietary db that connects to the solarwinds sql server. So you would have your primary polling engine with NPM and NTA…

aLTeReGo - so that worked on the server that I had removed and re-added to solarwinds. I went in, listed resources, hw health showed back up - checked on it and sure enough hw health polling is working. On the 1st dl380 however, the hw health never left the list resources page. I uninstalled HP SIM agent, 9.40, reinstalled…

Just curious - is the p6000 similar to the p2000, and if so it may be worth a look over at Storage Resource Manager. I know monitoring the P2000s was a little weird trying to use NPM & SAM, and I think the thought on our side was to investigate SRM or utilized the snmp traps from the P2000.

So you have two separate OAs you are monitoring, both are using the OA FW of 3.71 yet one is correct and the other is pegged? That seems odd.

stevenrlong‌ - as far as your nexus - take a look at these two threads - UDT not working on Nexus FEX units and UDT not seeing port details for Nexus 5k Switches I've done a ton of research and opened cisco tac cases and sw support cases for our nexus problems and think i have the problem with our nexus identified. And if…

HolyGuacamole - for "the undp threshold breach will not change the node status" I think that is what the OP is trying to be accomplished, although I could be wrong. Screenshots from @bonag would help to make sure we are understanding the ask.

Do you guys have FEX units connected to your Nexus?

Wow!! lol it was literally right in front of me and I missed it! Thanks!!

I think I got it...after seeing how easy you did with the command script..I started playing with excluding snmp etc. Finally came up with : show run | exclude (snmp-server|username|enable) Hopefully that will cover all my bases! Thanks again for the help!!

We had issues with our ucs chassis, we needed the blades to have CIMC IPs assigned, once they were assigned, orion was able to correlate them correctly.

oops - yeah misread the question! But with that said, why not create a custom property like you mentioned and then in your alert, adjust for that custom property? I mean you would have to adjust the alert anyway for an "unplugged" or the like condition so why not just make that custom property as "Alarm on device for…

Nice job on the detailed write up!! Glad I could help, I stumbled across that thread myself and its been perfect!

Hey joepoutre‌ - I've got all traps from all devices sent to our main poller. Same with syslogging...

Can you give me an example or screenshot of how to "using variables pull specific portions of the variables" ? I'm trying to do this same thing, I've got a log message coming in and am trying to strip out all the info I do not need and then fwd the message on.

If you are looking at the customize page (pressing edit in the box where the nodes with problems is) there are options to "Filter Nodes (SQL)" and setup a query to display specific devices that might be where one filter is setup

Hi peter.ksenzsigh - I just emailed you. For the life of me, I couldn't figure out how to message you privately on here...

there is a table in the configmgmt db, I think it is NCM jobs or similar. It shows all of the jobs within NCM, is the job showing enabled in there?

I'm not sure if this helps but I've got a crap ton of interfaces as well. We created a policy report that goes out and looks at interfaces that have regex strings of "ip address" in them, and then looks for "no ip redirects". Same sort of thing you are looking for but you could do it with ip helper. So you could do the…

Ok so I found a not so nice way to delete these from another thread and just tried it and it works. I figured I would post it here. Go to the Config Mgmt DB, and delete everything in dbo.TransferQueue. Like I said, rather not do it directly in the db, rather have it work via gui but this cleared our 522 statuses that were…

Yep SAM is installed - 6.1.0 - I guess this thread needs to move SAM forum?

tjreeddoc - Click on general tab and grab the name of the alert. Then head over to your sql server, and open up the dbo.AlertDefinitions table in the solarwinds db. I typically do a right click on the table, and "Select Top 1000 Rows" since I have less than 1000 created alerts. Once that query is run, I usually add a where…

 Why don't you just remove the pollers for these devices? Then they won't be alerted on. The problem is I will have one device that is say 6 bgp neighbors, 2 might be in an active state for 9 weeks or more while the other 4 neighbors are currently established. If I remove the pollers for those devices I won't get any bgp…

Yeah you should be ok.

Are you familiar with the concatenate process in excel? And just to note, I've blasted out hundreds of nodes at a time updating several different custom properties assigned to nodes, and node names etc - it works great for bulk stuff!! You could do a spreadsheet to update the nodes table like this: Column F is where you…

Was there any solution to this? We had a similar issue on a Nexus 7018, where the sup's failed over and in the process the line cards reloaded. Now all polled interfaces are in an "unknown" state.

Was there a resolution to this? I am seeing the same issue, where IPs are showing directly connected on devices that are no where related.

Yeah seriously - solar winds doesn't offer it directly to my knowledge and I know one of the companies they recommend is loop1, who I have had personal experience with.

So after doing some searching around, I figured out how to do a sql query within the alert definitions table using the trigger query I configured. The response I get from that is successful and displays the nodes alerting on what I would expect. So next I ended up looking at the actual snmp trap itself. I downloaded the…