Hello,

I kindly ask you to clarify whether you changed the names of computers or not. Please, find the answers below:

1) You can combine the nodes by editing the base manually. According to the information you provided, the inventory data is identical. Therefore, you can simply delete the duplicates from the Assets tree and from the Information folder and leave the nodes with actual names only.

2) To find the duplicates, please build Tabular report with either "IP Address" or "MAC Address" field. After that, sort the report by target field. If you see two nodes with the same MAC address value, please, delete one of them. Note, that it is necessary to add "Node Name" field to the report to be able to distinguish nodes.

3) Please, be informed that the duplicates will be created only if you scan via Wizard. Network Inventory Advisor cannot determine that "FinanceFred" PC is a "FredPC" which is already present in the Assets tree. We would recommend you to use "Scan Network" option to avoid creating duplicates.

And I just wanted to add a comment here. Kevin, the main reason why this happens is the way Network Inventory Advisor scans your nodes: by IP address or by DNS name. Both can be changed. Unfortunately, we don't have a mechanism to detect such changes in the core, yet. Since this is the way we identify nodes. While in most networks IP addresses can change frequently, the node name stays the same. In your case it seems to be different, as both, IP addresses and DNS names, are changed, that is what leads to duplication. We initiated an internal discussion on this, however, as you know mac-address can also be easily altered, and then we come to another issue: determining the set of parameters that can identify a node exactly, which might also be quite client-specific. Sounds like another huge feature though... need to think it over and plan.

Dear Jeff, I am Adriano from PAT. We are having the same issues as you know, and they are not easy to manage.

I was reading the forum and found this discussion.

I would like to help (I hope) by giving an idea on "how" to detect the NODE UNIQUE ID.

Basically I was thinking about that: how about store on the machine LOCAL registry -or- somewhere else maybe also through WMI interface if possible a GUID that YOU define at first scan, ... and use it as reference for that node(?):

MSDN Topic how to write and read registry through WMI

This can be useful, coupled with your current logic to identify nodes in the case this GUID is not found, in order to have the node uniquely identified. Also, note, that this could be managed "internally" to NI, and not necessarily published to the end-user. Basically you just need to know a NODE HANDLE, which obviously can't be neither the NODE NAME or IP Address, or MAC Address... (for instance, Teamed NICs build dynamic MAC address, and many times this can change... I know I already had this kind of issues a lot of time ago...)

Hope it helps.

Adriano, let me talk to my devs on that. We had so many internal meetings devoted to this feature in 2009 and 2010 now, and there so many do's and dont's of the idea and possible implementations... we need to make very careful decisions here.

Adriano, a technical comment from our developers:

Generally, there definitely IS sense in identifying nodes by unique ID. However, this solution will only work for Windows nodes. There's no registry in Mac OS (, Linux?). However, we are taking this into consideration and will get back to you with the better implementation of nodes analysis after the next update we are going to release soon.

Dear Jeff, do you have any news about the solution of this node duplication issues? This is becoming a REALLY blocking issue. On 1000 nodes we get 3 identical workstations on the tree...

THANK YOU!

We are looking into your issue at the moment. Thanks for providing the details. We'll be in touch via email.

HI Jeff,

I recently puchased your software last week and I ran the network scan today and found there is a duplicate computers, I emailed you as well. From this forum, your last comment was you havent fixed the issue. Can you tell me, if this has a fix or completely set aside?

Seems like the issue was resolved, right?

this is not resolved, the workaround you provided is just a temp fix but when we re-scan it always find duplicates and then I have to fix it manually and takes so much time

Norman, we are currently working on this issue. The global fix will be available soon. I'll keep you posted.