I had to move about 350 satellite IRDs from one set of 4 DMAs to another and I have about 74 that are giving the error messages of "Could not retrieve community library credential for port: 161..." The error is regenerated after a restart of the IRDs. Do I simply need to delete these 74 and import them again? What is the best resolution and should I have included an additional step to the procedure that is shown below.
DataMiner Version - Feature release 10.3.11
Below are the steps that I had used during the moves.
- Exported the IRDs to csv.
- Deleted the original IRDs in groups of about 45 each.
- Removed the Element IDs from all IRDs on the exported csv then separated the IRDs on the csv into 8 separate csv files, 2 files for each new DMA location.
- Updated the 8 csv import files with the appropriate DataMiner ID and View numbers.
- Imported each csv into the cluster and verified that all imports were present.
- Reapplied the Service Templates to pull the IRDs back into their appropriate services.
- Spot checked about 5 services on each DMA ensuring that the IRDs were shown in the service. Found one that was missing so I reapplied the Service Templates again to get any IRDs that might have been missed the first time.
Hi Joe,
Are these elements using credentials that are stored in the credentials library? Could it be that these elements are linked to a set of credentials that don't exist on the new DMAs?
For more information see: https://docs.dataminer.services/user-guide/Advanced_Functionality/Security/Basic_security_configuration/Managing_predefined_sets_of_credentials_for_SNMP_authentication.html
Hi Joe, since the error message mentions ‘community library credential’, I still think it could be related with that. Could you please check in the CSV file if the following columns don’t contain expected values: ‘SNMPReadCommunity’, ‘SNMPWriteCommunity’, ‘CredentialGUID’, …?
Hi Tom. Well you solved the mystery. When I exported these and moved them to the new csv files I ran “text to columns” and must have left the tab checked along with the comma. These elements happened to be in more than one view and of course DataMiner uses a tab to separate multiple views in the View IDs column. Leaving the tab checked caused the data to shift to the right for all columns to the right. I just tried to edit/save one of the elements and the error cleared. I’ll do that on the remaining elements instead re-importing them. I appreciate the explanation.
Actually I just exported the element that I tried the edit/save on and it still has some data in the incorrect columns. I’ll fix my original csv import and will delete these and re-import them just to be sure there aren’t any other issues. Again I appreciate your help.
Hi Joe, glad to hear that the problem has been found!
Hi Tom. Thanks for your response. I don’t believe there would be any credentials saved. These aren’t SNMPv3 or HTTPS. Just simple SNMPv2. No login needed. As long as the read/write community strings are correct we should be able to poll them. They are currently pulling data even with the error present. I’ll check out the article that you had provided and see if there is something else I missed. If not I’ll try deleting and importing these again. I appreciate it.