Network device with multiple IP addresses

Ronald Lucas
Tera Contributor

Using Istanbul...

When scanning a single network device with multiple IP addresses from a single discovery job, my understanding is the identification phase will recognize these IP addresses are on the same device and will only discover the device using one of its IP addresses.   The others will be ignored during that scan.   And the Management IP of that device will set to the IP used during this discovery.   Is that correct?

The next time the discovery job runs, will discovery continue to use the same Management IP address from before?   Or is it possible it will use any of the IP addresses on that device and update the Management IP?  

Thanks.

4 REPLIES 4

adilrathore
ServiceNow Employee
ServiceNow Employee

Absolutely correct. When we discover a computer with multiple IPs in the same range we stop discovery on the IPs that we find are already associated to the same class CI.



During the identification phase, Discovery at some point realizes that the three IP addresses are actually the same device and simply pick one to finish the Discovery (whatever IP Discovery happens to find first).



ServiceNow KB: Troubleshooting the Identification Phase in Discovery (KB0535238)


Dave Ainsworth
ServiceNow Employee
ServiceNow Employee

Hi Ron,


 


There is no guarantee that the IP address will be set to the management address. It will simply be one of the IP addresses. One of the reasons for this is that sometimes there is often no way for discovery to know what the management address is. There is some information about how the IP address is chosen here:



Control IP address selection



As Adil mentions, discovery will stop at the identification stage so the other IPs will not be totally ignored because identification still needs to happen and so it will reach the identification stage on the discovery, mark the IP as 'Identified, ignored extra IP' and then not continue with exploration.



The next time, exactly the same behaviour will occur, i.e. it will scan all IP addresses it has been told to do in the schedule. The first IP that it successfully identifies on (could be different IP to the previous scan) will go to the exploration phase and the others will stop after identification.



Regards,




Dave


Hi Dave.



So how about when a single network device reachable on 2 different IP ranges has 2 discovery jobs running at the same time?   For example, imagine servers are on both IP ranges below, and the single network device is reachable via 10.0.0.1 and 10.1.0.1.



10.0.0.0/24


10.1.0.0/24



If 2 separate discovery jobs are running at the same time, one job for each IP range above, will only 1 IP be selected between both jobs to scan the network device?   Or since each job is running independently, will the network device be scanned twice?



Thanks,


Ron


The explanations given above are true for a single discovery schedule. So the network device will be scanned twice since the two discoveries are independent of each other.