tdanner ✭✭✭✭✭

PowerShell is a bit stingy with error output by default, and I can't figure out what is going wrong from what it has printed here. Try using this function to get some more details: http://blogs.msdn.com/b/powershell/archive/2006/12/07/resolve-error.aspx Then I bet we can figure it out.

If you run Get-SwisObject $swis -Uri "swis://CALMON4AP01.fglcorporate.net/Orion/Orion.Nodes/NodeID=12360/CustomProperties" Does it return a list of the custom properties with all blank values? If you set some custom property values for that node through the UI and run that Get-SwisObject command again, does it return the…

For some verbs, we have documentation in the Orion SDK.pdf file installed with the SDK. Others you can figure out from looking at the sample programs. But there are a bunch more that don't have written descriptions. You can find out the names and types of the arguments with a query like this: SELECT EntityName, VerbName,…

The NodeIDs are actually generated by SQL Server.

Bulk update (and delete) through the REST interface is something I would like to see added, but I can't commit to a timeframe. It's 340423 in our bug tracking system if you want to ask product managers about it. Concerning the time difference in running a query through different channels - that's not something I would…

Post data for an "Invoke" call should always be a json array. StartDiscovery expects one parameter, so it should be an array of one item.

To remove a node from NCM without removing it from monitoring in general, use the Cirrus.Nodes.RemoveNode verb. It takes a single NCM NodeID (a guid) as the only parameter. The full Orion schema, including verbs but not much in the way of description, can be browsed at SolarWinds Information Service v3.0 Schema…

Does the same request work if you do it from the command line instead of the ticket system?

The simplest way to do this is just let discovery add everything (wait for it to complete) and then delete unwanted volumes and interfaces.

What is the query box? Is this a Custom Query resource? If so, then the problem is that this resource uses SWISv3.

Orion.Nodes.Unmanage requires the "node management" user right. And that HTTP response code should have been 403 Forbidden, not 400 Bad Request. I'll open a bug for that.

I tried your script, changing just the EngineID and IPAddress values, and this worked for me. How are you observing the ping request/response? Wireshark?

The AlertObjectID is unique for each combination of alert definition / object. It doesn't sound like you are using the AlertObjectID here.

I recommend using the Discovery API to handle this task: Discovery · solarwinds/OrionSDK Wiki · GitHub Sample: OrionSDK/DiscoverSnmpV3Node.ps1 at master · solarwinds/OrionSDK · GitHub (this sample says it is for an SNMPv3 node, but the same code will work for WMI if you just provide a WMI credential instead of an SNMPv3…

The first parameter to "update" should be a SWIS Uri. You got the Uri for the node in the call to swis.create above this. To get a uri that points to the custom properties for the node, just append "/CustomProperties" to it.

The web report writer modifies the query somewhat. Normally this does not affect the results, but it is getting confused by the correlated subquery in the where clause and producing a broken query. I came up with this as a workaround. It works in both SWQL Studio and the web report writer. SELECT N.Caption AS NodeCaption,…

What was the original issue that caused you to change IIS from port 80 to port 17777?

Could just be a scaling issue. I can't tell from the graph screenshot how the poller is configured, but I can see in the Cacti graph that it's showing "per 5 minutes". I don't think you can set UnDP to do "per 5 minutes", but you could do "per minute". Then you would expect the Orion chart to show a value 1/5 as large as…

There's no good alternative within Advanced Alerts. Depending on what you are trying to monitor, it might be possible to build this in Basic Alerts. We know that the existing "has changed" options are pretty limited, but we aren't scheduled to do anything about that in the next release. Maybe Denny could say more.

I always get a little more bald when I try to use curl from the cmd.exe command line with a complex URL. I recommend you use the POST form of /Query so you can avoid dealing with cmd.exe's insane quoting rules. Like this: curl -k -X POST -H "Content-Type: application/json" -d @postdata.txt…

Account permissions sound like a good solution. You could create an "EMEA-IP-Reservation-Service" account and restrict it to only see EMEA groups/subnets.

The CreateSubnet verb is very basic, but you can use a Create operation for IPAM.Subnet to specify more properties, like the ParentId. See https://github.com/solarwinds/OrionSDK/wiki/IPAM-API#crud-operations-for-subnets

Thanks. Are you getting that error from SWISv2 or SWISv3?

No progress.

SWQL doesn't have a general-purpose CAST or CONVERT operation. There is a "ToString" function that will force any data to be a string. What kind of conversion do you need to do?

Try including the "-Trusted" parameter on your call to Connect-Swis. This will cause it to use your current Windows identity instead of prompting you for credentials.

In your script you have these lines: CredentialName = 'Service Account' #Getting the credentialID credentialID = "SELECT ID from Orion.CredentialName where Name = " + str (CredentialName) credentialidquery = swis.query(credentialID) This is sending a query to swis that looks like this: SELECT ID from Orion.CredentialName…

According to https://github.com/solarwinds/OrionSDK/wiki/Credential-Management#create-snmp-v3-credentials, CreateSNMPCredentials is "Supported since: Orion Platform 2017.3 (NPM 12.2)". What version are you using?

Madhavan is out, so I'll try to pick up where he was. What I see is that the Information Service is getting "System.ServiceModel.FaultException`1[SolarWinds.Orion.Core.Common.CoreFaultContract]" and apparently not logging the details. That tells me that it called the Core Business Layer service and got some exception back.…

That's even trickier! Can your firewall map the ports for you? I've done that with an ssh tunnel before, but not with a firewall.