Community
- Command Central
- MVP Program
- Monthly Mission
- Blogs
- Groups
- Events
- Media Vault
Products
- Observability
- Network Management
- Application Management
- IT Security
- IT Service Management
- System Management
- Database Management
Content Exchange
- SolarWinds Platform
- Server & Application Monitor
- Database Performance Analyzer
- Server Configuration Monitor
- Network Performance Monitor
- Network Configuration Manager
- SQL Sentry
- Web Help Desk
Free Tools & Trials

Error when loading the UnDP node assign group by option

Hi All,

I'm getting an error message when using the custom properties on the group by in the UnDP node assignment.

ProvideFault failed,check fault information.

I have checked and verified all the services are up on all the polling engines, checked if any errors on logs(NPM.BusinessLayer)

, Verified the custom properties are not deleted or empty. Not all the CPs are getting the error message

Find more posts tagged with

Accepted answers

All comments

Fred.Loucks

From your posting, it sounds like you're getting random CP errors on various devices, but not all devices. Odd question. Are all your nodes build with the exact same hardware/model types?

Here are some examples of problems I've had throughout the years where minor differences between devices shot me in the foot regarding OIDs and Custom Property information. This information might help you in troubleshooting.

1) Cisco Catalyst Platform - 3750, 3750-G, & 3750-X switches are not the same and use different OIDs to pull information.

2) Cisco Catalyst Platform - 3850 switches come in different flavors. The WS-C3850-24T has 24 ports that support 1Gb each. The WS-C3850-24XU has 24 UPOE ports that support 100Mbps/1/2.5/5/10 Gbps connections based on hardware & inserts. Both devices use different OIDs based on insert and SFPs.

3) Cisco ASR Platform - 9000 series routers run on either a 32 bit OS or a 64 bit OS. 32 bit OS views hardware differently than 64 bit OS. Using the same OIDs on a 9006 running 32 b OS vs a 9006 running 64 b OS will have various CP failures. The underlying OS reads the hardware differently and controls data output differently.

4) Cisco ISR Platform - 4431 routers with different insert modules can be problematic. If routers are not uniformly built, there can be issues. If one routerhas a switch port model installed and another router doesn't, do routers should have/use additional OIDs.

5) Cisco Nexus Platform - Use Insight. Insight seems to poll without massive issues.

I hope this might help you a little in isolating what the issue is.

ralphpost

Bumping this as I had the same issue today.

Right clicking a UNDP - select Assign - when selecting the Group By: dropdown - select a CP value.

It does not happen on all CPs - only some.

All services running etc.

At a loss to pinpoint the cause/resolution.

ralphpost

And solution found.

Issue was with the data in various CP columns/data.

For example - we had 7 out of 57 CPs that would produce this error when selecting in the "Group By" in Universal Device Poller.

Investigating those CPs in the DB tables found various issues - some may have caused but a couple of definite causes:-

Duplicate CP values (for example - in CP Device_Function (set list of values, required CP) we had 2 entries for "Storage System")
Same CP as above - reserved word used (Unknown) - technically it should not have been possible to have this but there you go.
CP values (obviously copy/pasted in) that had a leading blank (space) - i.e. ' 1234' - these are obvious if you go into Manage CPs, select Node and the CP in question, select View/Edit values and then Group By and select the CP being edited - the leading blank values will appear at the top of the list and above the [Unknown] Group
By far the biggest issue and definitely causing the bulk of the problems - CPs entries that were blank - i.e. were not NULL in the DB tables but appeared as blank or a space. This one had us stumped for a while until we looked at the tables in the DB and noticed that many values for a CP against the nodes showed as NULL but some were just blank,

To fix the blank entries (not Null) various methods:

For those CPs in the table with only a few blank (not NULL) entries - just edit CPs and enter a value - test UnDP and now works with selecting that CP in Group By,
For those CPs in the table with lots of blank (not NULL) entries (one table/CP had over 200) - export the CP, delete the CP, re-add the CP then import the CP values from your export (ensuring you edit the import excel to only import the entries that had a (valid) value for the CP (ensuring the blank entries are not imported). Once this was done, selecting the CP in UnDP via Group By worked.

To find the offending entries use SQL query like:

Select * FROM [NetPerfMon].[dbo].[Nodes]
where Your_CP_Name is not null

This will return all entries where your CP has a value (is not NULL) and you can easily see which are blank.