Provider Edge Access Ring Question
alemoo
Member Posts: 9 ■□□□□□□□□□
in CCNA & CCENT
Hello,
I am a junior network engineer learning the ropes at my company. We are extremely busy and therefore there is a high amount of pressure on me to learn on my own and not ask too many questions. I am hoping someone can help clear up some confusion as to the reason why things are done a certain way.
We are on a contract for a customer who has hundreds of routers across the country, providing triple play services. In one city for example, we will have two provider edge routers as part of an access ring of switches. So basically the logical topology is a horseshoe shape of switches, with one provider edge router at each end. Attached to each switch are one or more devices that aggregate fiber optic connections, which in turn, lead to customers. On the switches there are hundreds of VLANs for different customers.
All access rings have provider edges, and we have VPWS (VLL) connections from the PE's to a core router where the VLLs terminate. Sometimes we simply establish a VLL from one PE to another PE. They say we do this because sometimes in reality there are several other unrelated routers on the path between one PE and another PE in the same access ring. Also they say that using VPWS is necessary because of spanning tree.
I understand that it might be very difficult to understand what is going on without a diagram. I also can not provide too much detail because of privacy issues. Does anyone have any idea what I am talking about? I just want some feedback so that I can become a better network engineer and understand more about the reasons why we do these things.
I am a junior network engineer learning the ropes at my company. We are extremely busy and therefore there is a high amount of pressure on me to learn on my own and not ask too many questions. I am hoping someone can help clear up some confusion as to the reason why things are done a certain way.
We are on a contract for a customer who has hundreds of routers across the country, providing triple play services. In one city for example, we will have two provider edge routers as part of an access ring of switches. So basically the logical topology is a horseshoe shape of switches, with one provider edge router at each end. Attached to each switch are one or more devices that aggregate fiber optic connections, which in turn, lead to customers. On the switches there are hundreds of VLANs for different customers.
All access rings have provider edges, and we have VPWS (VLL) connections from the PE's to a core router where the VLLs terminate. Sometimes we simply establish a VLL from one PE to another PE. They say we do this because sometimes in reality there are several other unrelated routers on the path between one PE and another PE in the same access ring. Also they say that using VPWS is necessary because of spanning tree.
I understand that it might be very difficult to understand what is going on without a diagram. I also can not provide too much detail because of privacy issues. Does anyone have any idea what I am talking about? I just want some feedback so that I can become a better network engineer and understand more about the reasons why we do these things.
Comments
-
networker050184
Mod Posts: 11,962 Mod
I understand what you are talking about but hard to give much specific info without more information. Basically the point of doing stuff like this is to provide L2 connectivity over an L3 network.An expert is a man who has made all the mistakes which can be made. -
alemoo
Member Posts: 9 ■□□□□□□□□□
Thanks for the reply, networker050184.
I believe that is the general premise of the design, yes. They were mentioning how you can not close the access ring, because if you do close it with one (or many different) VLAN, you create an L2 loop, so instead you must create an L2 psuedowire (VPWS) between two provider edges of the ring, and then STP will block it. I'm just trying to find a way to understand what is happening here.