RME Conflicting Device Types

Answered Question
Dec 11th, 2007

I found 12 devices being in conflicting state, before resolving them I thought I will confirm what the GUI states and found the following:

- I did an export of these 12 devices from DCR to get the sysObjectID stored in DCR.

- I did an snmpwalk to get the sysObjectID from the device

=> for all devices (except of 2 which had wrong DNS entries and 1 being unreachable) sysObjectID was correct in DCR

The 'Conflicting Device Type' report listed the correct sysObjectID 'found by RME' and wrong entries for 'sysObjectID in DCR' (for the 9 devices)

Will RME only once pulls this info from DCR, generate the report and never reconfirm this if there are changes in DCR ?

I have this problem too.
0 votes
Correct Answer by Joe Clarke about 8 years 10 months ago

As it's documented (and working as designed), it would be an enhancement request. I agree with you that this is non-intuitive, though.

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 4.5 (2 ratings)
Joe Clarke Tue, 12/11/2007 - 12:25

When RME determines that a device is conflicting, it will move it into a conflicting state until a user resolves the problem. If the data (sysObjectID and MDF ID) are correct in DCR, delete, then resubmit the devices into RME.

Martin Ermel Tue, 12/11/2007 - 12:55

Thanks, I will do that.

But a question to the understanding of the overall process:

I assume the report was correct at the time the automange feature of RME tried to add the devices to RME. But according to the philosophy of DCR (that every change in DCR will be pushed to the apps) I thought that changeing the info in DCR (the obvious change here is sysObjectID ) should be pushed to RME as well and start a process to verify the RME device 'values' regardless of the device state. In this example the devices could be dropped from this list automatically and could be added to the normal devices - if I do not miss anything in the update process.

Is this a wrong understanding of the DCR concept?

(it would be more practicable in big environments)

Joe Clarke Tue, 12/11/2007 - 13:00

That would be nice, but the way the algorithm works is that once DCR is updated, the conflicting device must be manually resubmitted for RME to manage it. This is noted in the online help, though it may not be very clear:

RME detects an conflict device using this algorithm:

1. RME gets sysObjectID from Device and Credential Repository.

If the sysObjectID is null, RME updates the sysObjectID collected by Inventory collection, else go to step 2.

2. RME compares the sysObjectID in the Device and Credential Repository with the sysObjectID collected by Inventory application for a given device.

If they match, the device is moved to either Normal/Pre-deployed/Pending state. Otherwise, RME moves the device state to Conflict and allows you to update the Device and Credential Repository and resubmit the device for management or delete the device.

Martin Ermel Tue, 12/11/2007 - 13:25

Yes, I read that. But I was stumbling over the fact that the 'Conflicting device types' is a static list and presents outdated data. In my case in fact DCR sysObjectID is correct (now) but RME will never remove these devices from the list and even presents wrong data - this could be hard to argue and for the most users hard to understand. 'Conflicting Device Type' does not presents its data like a job history where you can say 'at the time when these data was collected and compared this was truth'. Instead it seems to be current data (like 'Normal Devices').

But from what you said, I am not sure if my thoughts are a feature request or (from what I thought) a bug (because of the basic DCR concept)

Correct Answer
Joe Clarke Tue, 12/11/2007 - 14:30

As it's documented (and working as designed), it would be an enhancement request. I agree with you that this is non-intuitive, though.

Martin Ermel Wed, 12/12/2007 - 00:00

Thanks! I will inform the AM of the customer that he file an ER for this.


This Discussion