The networking diagram is as follows:
LS-1 (CE and vlan1002)----NE40E----MAN---T128 made by vender Z
LS-2 (CE, the device fails to be registered)
The above networking diagram shows that the NE40E replaces the T160G made by vender Z on the existing network.
Symptom of fault:
After the NE40E replaces the T160G made by vender Z on the existing network, all the services run normally except the VSI service.
Because all the services run normally except the VSI service after the NE40E replaces the T160G made by vender Z on the existing network, the fault cannot he caused by the link failure or the abnormality of other devices on the existing network.
1. Check the configuration of the NE40E.
vsi LvGuanXinXiXiTong static
l2 binding vsi LvGuanXinXiXiTong
Data is configured correctly.
2. Check the VSI status of the NE40E:
[NE40E]display vpls connectio
7 total connections,
connections: 7 up, 0 down, 7 ldp, 0 bgp
VSI Name: LvGuanXinXiXiTong Signaling: ldp
VsiID EncapType PeerAddr InLabel OutLabel VCState
100 vlan 188.8.131.52 142336 66 up
100 vlan 184.108.40.206 142337 120 up
100 vlan 220.127.116.11 142338 107 up
100 vlan 18.104.22.168 142342 125 up
The above information shows that the Layer 2 tunnel of VPLS is set up, but the end user cannot access the server and the interconnection of the two CEs fails.
3. Ping the server from LS-1 to check the traffic count of the VSI service on the NE40E.
<NE40E>disp vpls statistics vsi LvGuanXinXiXiTong
*VsiName : LvGuanXinXiXiTong
In Frames : 338
Out Frames : 541
In Bytes : 11341
Out Bytes : 41411
In Discard : 0
The preceding information shows that the packets can be received and sent normally.
4. When you ping the server from LS-1, capture the VPLS packets to and from the NE40E on other devices (GSR) in the MAN:
The packets flowing into the NE40E captured are as follows:
0018 821D 2010 0014 1CD2 FC06 8847 22C0
01FE 0019 E019 0D9E 0019 21D5 5FD6 0806
0001 0800 0604 0002 0019 21D5 5FD6 0303
0301 0019 E019 0D9E 0303 0302 0000 0000
The field 0806 shows that the VPLS packets from LS-1 do not carry the VLAN tag, and thus are only the common ARP packets. Because the negotiation mode of VSI between the two CEs is configured as VLAN, the NE40E adds the VLAN tag to the VSI packets. That is, after carrying the VLAN tag from the NE40E, the packets are sent out from the outbound interface of the NE40E.
The packets forwarded by the NE40E are as follows:
0019 E019 0D9E 0019 21D5 5FD6 8100 019b
0800 0604 0002 0019 21D5 5FD6 0303 0301
0019 E019 0D9E 0303 0302 0000 0000 0000
The NE40E replaces the field 0806 (indicating the ARP packet) in the ARP packet coming form the T128 of Z with the field 8100 (indicating the VLAN packet). Thus the ARP packet changes into the VLAN packet and the VSI service is interrupted. To fix the fault, configure the encapsulation modes of the VSI on the NE40E and the T128 as Ethernet or add a VLAN tag to the VLAN encapsulation mode on the T128 of Z.
Method 1: Run the following command to modify the VLAN tag on the T128:
(config terminal) # mpls l2transfer vpls transfer vlan-tag
Method 2: Run the following command to configure the encapsulation modes of the VSIs on the NE40E and the T128 as Ethernet:
Run the following command to modify the encapsulation mode of the NE40E:
[Quidway-vsi-LvGuanXinXiXiTong] encapsulation ethernet
To modify the encapsulation mode of the T128, remove the VLAN at the sub-interface of the T128.
Why is the VPLS Layer 2 tunnel Up when the NE40E resolves the packets incorrectly? Check the configuration of the T128. The packet receiving mode of the VPLS packet on the T128 is mixed. That is, any packet can be processed, but the received VLAN-encapsulated VPLS packets are forwarded without the VLAN tag. That is why the VPLS Layer 2 tunnel is Up when the NE40E resolves the packets incorrectly.
According to RFC4448, if the packets are encapsulated in tagged mode, packets transmitted on PW must have a VLAN tag. Therefore, the operation on the NE40E conforms to the RFC standard, but the T128 does not.
1. The data configuration of the NE40E is wrong.
2. Whether the VPLS can set up the Layer 2 tunnel and learn the MAC address of the peer device normally.
3. The VSI configurations of the two ends are inconsistent.
4. Packets are discarded on the network.
Relavant specifications in RFC4448:
If an Ethernet PW is operating in raw mode, service-delimiting tags
are NEVER sent over the PW. If a service-delimiting tag is present
when the frame is received from the attachment circuit by the PE, it
MUST be stripped (by the NSP) from the frame before the frame is sent to the PW.
If an Ethernet PW is operating in tagged mode, every frame sent on
the PW MUST have a service-delimiting VLAN tag. If the frame as
received by the PE from the attachment circuit does not have a
service-delimiting VLAN tag, the PE must prepend the frame with a
dummy VLAN tag before sending the frame on the PW. This is the
default operating mode. This is the only REQUIRED mode.