cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
749
Views
0
Helpful
3
Replies

Starting eval of LMS 4.2 and running into issues with inventory

sdavids5670
Level 2
Level 2

After defining default credentials (for SSH, SNMP, etc) I created a standard discovery job and specified a seed device.  The job immediately found the seed device and then proceeded to discover all of the downstream devices (10 total including the seed).  The summary of the discovery job stated that 10 devices were added to DCR.  When I look at the inventory, however, I do not see any of the devices.  When I look at jobs associated with the devices found during discovery I find several references to failures related to accessing the devices.  However, when I do an export of the devices out of DCR all of the credentials check out.   Does anybody know whether or not the eval copy of full-featured?  Am I running into a limitation of the eval?  Or is this just a bug?  Or is the discovery/inventory process not as seamless as I am assuming it is?               

3 Replies 3

jstam
Level 1
Level 1

Hello, just make use of custom discovery instead of standard discovery and you get rid of the issues.

Good luck!

Regards,

John-Paul

Turns out the problem was related to the decision making that LMS does with SNMP.  In the "SNMP Settings" section of the "Standard Discovery Settings" page I checked both fallback boxes ("SNMPv2c to SNMPv1 Fallback" and "SNMPv3 to SNMPv2 Fallback" and I defined configuration for V1, V2 and V3 (I'm in the process of switching over to V3 so most of my devices are not configured for V3).  My expectation (using what I think is common sense) was that the software would try v3 and fallback to v2c and if v2c didn't work then fallback to v1.  Makes sense to me.  However, here's what really happened.  When the initial discovery added devices to DCR it applied the Default Credential Set Policy that I created before running the discovery to each of the devices it found (this default policy includes configuration for all 3 version).  It seems to me that one thing (the fallback settings under "Standard Discovery Settings") doesn't have anything to do with the other (ie, behavior during inventory) because the inventory process tried v3 and v3 failed so it just decided to quit at that even though it had options for v2c and v1.  I went into device management for one of the devices I couldn't inventory, turned off v3 and re-ran inventory and it inventoried that device.  Can anybody tell me how to change the behavior so that it tried all available options during inventory (preferring v3 over v2c and v1)?  Is that possible?

I tried looking for a v3 to v2 fallback however that never went well nor did it ever work.

I ran into the same issue as well, the best solution I found was to  leave my v2 community strings and the v3 configs on the few devices I  had configured and just had LMS manage the devices via snmpv2. 

Now that all my devices are inventoried via v2, when I make the switch to fully integrate v3 I will simply just switch over and ditch v2.

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community:

Innovations in Cisco Full Stack Observability - A new webinar from Cisco