RichardLetts ✭✭✭✭✭

the inventory collection scripts are stored in C:\Program Files (x86)\SolarWinds\Orion\NCM\Inventory if the devices support JUNIPER-MIB:jnxContents* Then it should be just a matter of updating the 14-EntityMib... files to pull the required values. (sadly the inventory templates are not well documented)

and I unflag this as implemented, because as implemented it's less than useful :-(

this would be taken care of if the interface speed is a type-2 slowly changing dimension (see the work i cite below)

Yes, it works okay on 12.2 -- though the trigger got moved to NodesData as Nodes is now a view.

the SMTP server settings needs to honor the TTL returned from DNS queries. We added three new SMTP gateways to our cluster and retired three old ones, and the Alert manager was still using the old [cached] IP addresses a week later [TTL=5 minutes]. Fortunately our retirement process includes making sure no-one is using…

You will have to push up the log level to DEBUG on the alert manager to figure out why. what is your reset condition? if you did not specify one then the generated one is probably wrong. also, you can way simplify your SQL to something like: WHERE Nodes.NodeID IN ( SELECT traps.nodeid FROM traps with (nolock) INNER JOIN…

given Juniper doesn't really have running vs startup, we could subvert the startup config for the rescue config: so the backup command for the startup [actually rescue] config would be: file show /config/rescue.conf.gz so the backup command for the running config would be: show configuration I think that might be e

Wow, I use this weekly. The spreadsheet import was one of the nicest way to populate custom properties in NPM

Where did the SDK forum end up? Thanks

This idea was created the same month that the Avengers Movie was released Here are some great quotes from the Movie... Tony Stark: That man is playing Galaga! Thought we wouldn't notice, but we did. Jarvis: Sir, the Mark VII is not ready for deployment. Tony Stark/Iron Man: Well then skip the spinning rims, we're on the…

Change 2: Integrate with Aruba airwave AWAMP-MIB:configAlert traps Create or update thin WAP entity with properties awampApName L100.OFFICE.AP01 APName awampAPFirmware 6.3.1.5 APFirmware awampAPModel AP 105 MachineType awampAPMFG Aruba Vendor awampEventAPICON ICON awampEventGroupMonURL ... awampEventGroupMngURL ...…

in many cases I can write queries against existing database tables to populate dependencies, sometimes adding thousands of dependencies. I'd like a UI that allowed that, and also for the dependency evaluation engine to be able to cope with tens of thousands of entries...

BUMP -- yet another request about handling generating alerts from traps for a new network device being added has entered our Jira queue. Two months ago I noted this was becoming ridiculous. Now I an going to call this concerning.

I see this says 'implemented', but I can't find where to edit these. I can't see anything in the release notes or online manual: can someone guide me to where I can customize the MachineType values? thanks

I don't use account management on individual accounts -- I do all of this through [AD] groups. see also: https://thwack.solarwinds.com/ideas/1246

In a PCI/HIPPA environment it's useful to have splunk cleanse the logfiles of any credit card, PII, or other sensitive data before loading them into the server, and then restrict the types of search. If you're handling PPT (position of public trust) data (e.g. student loans) then you can restrict the data to those with…

use custom SQL on the [node] alert: trigger condition: WHERE nodeid IN (SELECT nodeid FROM traps WHERE TrapType = 'PowerNet-MIB:apc.0.3' AND DateTime > DateAdd(day, -1, Sysdatetime()) AND Acknowledged=0 ) reset condition: WHERE nodeid NOT IN (SELECT nodeid FROM traps WHERE TrapType = 'PowerNet-MIB:apc.0.3' AND DateTime >…

BUMP: I have several hundred of these we could try this feature on...

the same is also true for Juniper routers: certain models of interface card, CPU card or other physical entity might be EOL/EOS before the whole chassis is.

at one place I worked, we had Cisco Call Manager. Unfortunately a memory leak on it meant that periodically the primary system would stop responding and the only way forward would be to restart the process, or reboot the box (depending if we noticed in time). one day when we rebooted the box it claimed it had no disk…

Dealing with my ISP: Me: There is a problem with your head-end router, you need to get an engineer to troubleshoot it Them: no the problem is with your cable modem and router, we can see it fine on our network Me: That's interesting because I powered it off and disconnected it from the wall before we started this…

NCM already discovers the 'ports' in its inventory collection of interfaces on a node -- why not just have the ability to define rules that automatically import those into UDT and NPM as monitored ports (for UDT) and monitored interfaces (for NPM)

Hi Kellie, I'd really like to give you a tour of our home-grown CMDB, because that is one area where NCM/NPM is very weak. There are some UI features there that help when you have tens of thousands of devices that work for us, and might also work for other customers. We also use Aruba Airwave -- I think there is a lot of…

As an aside, you can get traps into the current alert manager; create a custom SQL alert on the NODE and put something like the following in the custom SQL clause: WHERE nodeid IN (SELECT nodeid FROM traps WHERE traps.traptype like 'CERENT-454-MIB:%' AND Datediff(day, traps.datetime, Sysdatetime()) < 1 AND…

http://thwack.solarwinds.com/ideas/2637 data warehouse is a more complete idea that also solves this problem.

My additional polling engines are crashing with out of memory exceptions due to the 32-bit limits on businesslayer processes. Apart from that limitation I'm not seeing what resource on the servers they are actually running short of.

Can you provide a standards-reference for NQA? I'm trying to find out if this is the same as Juniper's RPM (which is what I would like from IPSLA)

Our unsampled netflow PDU-rate would currently be over 100,000 per second (6 million/minute) (and we're not at a time of peak network traffic). We sample the data at 2000:1 to bring it into a more reasonable range; auto-scaling would be a good feature.

Since I wrote this feature request Solarwinds have acquired Wheb Help Desk, which includes SAML authentication... I'd like to see that component ported over to the core product.

Are there any API operations to a) add a subnet b) set its properties (e.g. neighbor for discovery) and having found a free IP address, are there any API verbs that allow you to set it its properties (for example that it is now in use, so the next person trying to find a free IP address gets a different one)? Thanks