04-27-2012 07:40 AM - edited 03-07-2019 06:22 AM
I have the scenario as pictured in the attachment.
Company A migration with Company B
I built a layer 2 trunk between the Company A and B core switches. Both switches can talk over the newly built Layer2 trunk. Both switches are same VTP domain and running same STP. All the vlans in Company A advertise to ISP via eigrp and they want to keep eigrp running in future too, similarly company B advertises its routes using BGP /MPLS.
After the migration I am not able to reach the new switch from NOC/Remote management [ packets dropping at Company B core ].What routing changes do I need to make on both core switches to make sure the new switch connected to Company A can be reached from Remote management. Attached the Configs files for both companies.
Challenge is: I cannot ask ISP to advertise company B’s subnets come through Company A Core router. A typical path to reach company B subnets in Company A is: to come from ISP-> Company B core-> laye2 trunk-> company A.
Example. To reach SW3[ 192.168.30.0/24] from Remote management, it has to go through MPLS cloud- ISPA-> Company B core-Via L2 trunk to Company A-> SW3.
Any help advise is appreciated.
Thanks
Blach.
04-27-2012 11:40 AM
Hi,
After a very lean understanding of your network I am able to figure out that you are missing a reverse route on your Company B core switch.
Just try to add a static route for your NOC and this should solve your problem.
Regards,
Pawan Sharma
04-27-2012 11:58 AM
I am not sure what you mean reverse route? Can you be more specific?I can reach anyof the Company B switches from NOC that are connected to Company B core router, The swich connected to Company A Core is no different than anyother switches that are connected to Company B core.
04-27-2012 12:04 PM
Hi,
By reverse route I mean, default gateway configuration the switch.
Regards,
Pawan Sharma
04-27-2012 12:10 PM
I do have default on each core switches to ISP; Sorry I did not include in the Configuration file.
04-27-2012 01:08 PM
Hi,
I did not understood where does stands the SVI ( interface vlan) for the new vlan (192.168.30.0/24) on Company A or B ?
Can you post from the Company B :
show bgp all summ
Dan
04-27-2012 09:23 PM
Thanks for the reply. The SVI for company B's subnets is defined on company B core.The Idea is to use the span company B's subnets to company A.
Example:192.168.30.1[which is management vlan for switches in company B] is defined on core switch B. When a new switch is added to core switch A with management IP 192.168.30.4[example]. I cannot reach it from remote management.
On company core A I added the below static route for remote management/Noc subnet.[166.44.50.0].
Ip route 166.44.50.0 255.255.255.0 192.168.30.1
However when I tracert from Noc to new switch IP 192.168.30.4 , it drops at core B[ 192.168.30.1]
I am not sure what static route I have to add on core B to reach from 192.168.30.1 to 192.168.30.4 as I do not have any SVI defined .
04-28-2012 12:33 AM
Hi ,
If the new switch on company A has the default gateway 192.168.30.1 ( which is the SVI on Company B ) there is no need for a static route.
On company B :
Can you ping the new switch from the Company B ?
Is the SVI configured in the VRF ?
Your interface toward the ISP in the VRF ?
Is your BGP adjacency with the BGP estableshed per address-family ipv4 vrf also on ipv4 unicast ? show bgp all summ
Dan
04-28-2012 12:25 PM
If the new switch on company A has the default gateway 192.168.30.1 ( which is the SVI on Company B ) there is no need for a static route.
->I agree, However I already have an default gatway defined to ISP for any routes that doesnt getadvertised via eigrp.I dont think it makes any sense to add one more default route to SVI on company B.
Apart from I explained above in my messages ,I bascially want to manage the Company core A switch by assigning the management IP from the management vlan[192.168.30.0.24/subnet].and currently it drops after Company B when traced from Noc. Iknow there is route missing and I am not what should the route from company B to reach company A switch
On company B :
Can you ping the new switch from the Company B ?
-> I can ping the new switch from company B and viceversa, but not from NOC
Is the SVI configured in the VRF ?
-> Yes it is configured in VRF
Your interface toward the ISP in the VRF ?
-> Yes
Is your BGP adjacency with the BGP estableshed per address-family ipv4 vrf also on ipv4 unicast ? show bgp all summ
->Yes
For address family: IPv4 Unicast
BGP router identifier 192.168.10.3, local AS number YYYYY
BGP table version is 5, main routing table version 5
Neighbor V AS MsgRcvd MsgSent TblVer InQ OutQ Up/Down State/PfxRcd
192.168.10.1 4 XXXXX 0 0 0 0 0 never Idle
192.168.10.2 4 XXXXX 0 0 0 0 0 never Idle
For address family: VPNv4 Unicast
BGP router identifier 192.168.10.3, local AS number YYYYY
BGP table version is 1280, main routing table version 1280
109 network entries using 16259 bytes of memory
190 path entries using 10880 bytes of memory
18/9 BGP path/bestpath attribute entries using 2520 bytes of memory
16 BGP AS-PATH entries using 168 bytes of memory
1 BGP community entries using 24 bytes of memory
1 BGP extended community entries using 24 bytes of memory
0 BGP route-map cache entries using 0 bytes of memory
0 BGP filter-list cache entries using 0 bytes of memory
BGP using 24850 total bytes of memory
BGP activity 789/324 prefixes, 1930/1170 paths, scan interval 15 secs
Neighbor V AS MsgRcvd MsgSent TblVer InQ OutQ Up/Down State/PfxRcd
192.168.10.1 4 XXXXX 447234 445672 1280 0 0 46w6d 78
192.168.10.2 4 XXXXX 443791 445863 1280 0 0 53w0d 78
04-27-2012 02:18 PM
By route back I think he is implying that you have a way there, but have you checked the routing tables to verify a way back? Just another place to start.
04-28-2012 12:47 PM
yeah... thats exactly what I was trying to say...
Regards,
Pawan Sharma
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide