We are in the process of deploying Cisco 2901's running IOS 15.0 to several locations across the United States. Each of these routers is connected to its own AT&T T1 circuit via VWIC2-1MFT-T1/E1 cards. Our head office has 4 trunked T1 circuits @ 6.0Mb to a single 2901 via two 2-port VWIC2 cards The majority of the last mile copper/linecards for the out of state T1's are delivered by LECS such as Qwest, Windstream, etc. All routers have essentially the same configuration aside from the ip's unique to each site All sites will be on the same MPLS network and ASN
The idea was that after each router is physically connected to the T1 linecard at each location and the T1's are turned up by AT&T, each router would then establish BGP connectivity and join all existing locations on our MPLS network. This plan seems to have worked initially for the first 2-3 sites, however, every remaining location has instead been dead in the water after turn up. Power cycling the routers does not fix the problem. We now find ourselves in a position where we have to get personnel on site with a console cable and remote desktop to allow us access to each Cisco to correct the problem. All we then have to do is turn BGP off and back on which establishes the BGP connectivity with AT&T and our MPLS network as the router should have done without requiring our intervention. The details of why we have to use console cable access vs existing ethernet/lan infrastructure are another story which isn't relevant. It also appears that if a router loses power we again have to coordinate someone on site to get us console cable access to disable/enable BGP which fixes connectivity. We haven't been able to do enough testing to be 100% certain about this power aspect but it seems to be another symptom of the same problem.
I'm unable to paste the exact configs tonight, but the following will give you an idea.
router bgp (ASN) no synchronization bgp log-neighbor-changes network (IP) mask (subnet) neighbor (IP) remote-as (AS) no auto-summary
the command sequence we run afterwards which establishes BGP & MPLS connectivity bringing the router online:
We immediately then see all of our existing MPLS locations listed and the router is good to go.
All we are doing to fix the problem is simply taking bgp off/on and then just typing out what was already in the config of the router for bgp as 'show run' would demonstrate. This simple task becomes a major headache when it has to be coordinated a thousand miles away each time. I'm considering putting together an EEM script which will handle this task for us at system startup, but that seems like an awkward patch rather than a fix. We've spoken with AT&T a couple times and forwarded them our router configs, but at this time they have suggested getting in touch with Cisco which we'll be doing shortly if need be. The concensus seems to be that the routers should be automatically negotiating the BGP connectivity with AT&T's side but for an unknown reason are not. I am by no means an expert on Ciscos or the topics we're dealing with, I'm only relaying what I know so far of the problem hoping others in the community with more experience may notice an obvious solution right off the bat. I can follow up with any additional details of the configs in the morning if requested.
router bgp (ASN) no synchronization bgp log-neighbor-changes network (local IP) mask 255.255.252.0 neighbor (IP) remote-as (AS) neighbor (IP) update-source GigabitEthernet0/0 no auto-summary ip forward-protocol nd !
after some further experimenting we determined that removing "neighbor (IP) update-source GigabitEthernet0/0" altogether seems to solve the problem and we'll investigate any impact this may have on the lan side as time permits.
Question We run asr9001 with XR 6.1.3, and we have a very long delay to
login w/ SSH 1 or 2 to the device compare to IOS device. After
investigation, the there is 1s delay between the client KEXDH_INIT and
the server (XR) KEXDH_REPLY. After debug ssh serv...
Introduction The purpose of this document is to demonstrate the Open
Shortest Path First (OSPF) behavior when the V-bit (Virtual-link bit) is
present in a non-backbone area. The V-bit is signaled in Type-1 LSA only
if the router is the endpoint of one or ...
Hi, I am seeing quite a few issues with patch install and wanted to
share my experience and workaround to this. Login to admin via CLI, then
access root with the “shell” command Issue “df –h” and you’ll probably
see the following directory full or nearly ...