This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

IPAM Feature: Discover subnets by obtaining routes from a seed router.

All of our plants with multiple VLANS/subnets export their subnets via OSPF to BGP in our WAN.  Our main BGP router has all these routes.  I'd like to be able to discover subnets by a query to a seed router.  These should appear in a list of discovered subnets the same way that DHCP discovered subnets are listed and need to be manually added.

This should ignore subnets that already exist (with the same CIDR length) in IPAM already.  There should also be a way to ignore subnets we don't want to manage like NPM discovery has.

DHCP doesn't know about many of our VLANs at plants.

  • Thanks for contacting us. I've created a feature request to track this problem (111972).

     

    Is there anyone else that have the similar infrastructure and would like to apply the same scenario?

  • Normally the router is authoritative for subnets, unless you are doing route summarization somewhere to reduce route count.  We will take a class /24 for a given site, and bust it into several VLANS/Subnets.  After messing with IPAM, it seems that making each location a supernet, and adding subnets in it works best.  This is especially true since supernets and subnets don't sort in IP order.

  • We are setup very similarly, although we do not use BGP.  We have large plants with class B's and many other smaller sites with 18 and 20 bit subnets that all tie back to the core summarized.

    I would love it if we could determine which layer 3 devices that IPAM could use for auto population via the route table.  Seems it would be a fairly simple calculation.  Then it could verify the subnet is not a duplicate (already does this in import) and assign it to its smallest supernet as a parent.

    Human error is our biggest problem....  This would eliminate that....

    Thanks for listening guys!!