The Ethernet service OAM function is maintained based on Ethernet service streams to provide automatic detection, fault location, and performance detection for the connectivity of Ethernet links. By using this function, you can maintain Ethernet Layer 2.
Overview
With the growth of Ethernet services, more Ethernet services are applied in MAN and WAN networks. Thus, there is an increased focus on the maintainability of Ethernet services. Currently, the ways to maintain Ethernet Layer 2 are limited. Thus, the ways of operating, managing and maintaining the Ethernet services through a transport network are limited.
- Detect the connectivity of Ethernet services.
- Find faults of Ethernet services.
- Locate faults of Ethernet services.
- Detect performance events of Ethernet services.
The Ethernet service OAM complies with the following standards: IEEE 802.1ag
Basic Concepts
As a protocol based on the MAC layer, the Ethernet service OAM detects Ethernet links by transmitting OAM packets. The OAM packets are processed at only the MAC layer.
- Maintenance domain (MD): The MD is a network that requires the OAM.
- Maintenance association (MA): The MA can be considered as a service-related domain, which consists of many maintenance end points (MEPs).
- Maintenance end point (MEP): The MEP is the transmitting and terminating points of all OAM packets. It is relevant to services. The MEP has a unique MEP ID in the MA. In a network, the MA and MEP ID can uniquely determine an MEP.
- Maintenance intermediate point (MIP): The MIP is relevant to the MD but irrelevant to the MA. The MIP cannot transmit OAM packets. The MIP can respond to and forward LB and LT packets, but can only forward CC packets.
- MD for customers: Create an MEP on NE CE, and create an MIP on the UNI of NE PE. In this way, you can detect the entire link that carries customer services, the MD level is 3.
- MD for carriers: Create an MEP on the UNI of NE PE. In this way, you can neglect the link quality on the user access side but focus on the network of carriers, the MD level is 5.
NOTE: The MD level displays the level of OAM frames, used to identify different customers. The smaller the number is, the higher the level is.
Basic Principle
- Application scenario: The IEEE 802.1ag ETHOAM is used to realize network-wide end-to-end Ethernet OAM function.
- Basic principle: Ethernet OAM implements the automatic fault detection and fault location through continuity check (CC), in-service loopback (LB), link trace (LT), Ping Test and Performance Detection operations. In this way, OAM capability of Ethernet services is improved.
- Operation objects: The operation objects of Ethernet OAM are the maintenance points, which include maintenance end points (MEPs) and maintenance intermediate points (MIPs).
- Realization process: The Ethernet OAM is implemented when an MEP initiates fault detection. When the MP detects a fault, it reports alarms. These alarms are correctly synchronized to the Ethernet trails, which are related to the MPs.
Continuity Check
The continuity check (CC) is used to test the link state unidirectionally.
The source MEP constructs the continuity check message (CCM) packets and transmits them periodically. After receiving the CCM packets from the source MEP, the sink MEP directly enables the CC function that focuses on this source MEP.
If the sink MEP fails to receive the CCM packets from the source MEP within a check period, it reports the CC_LOS alarm automatically until the link is recovered. That is, the sink MEP does not clear the alarm until it receives the CCM packets from the sink MEP.
As shown in Figure 2, the CC function of the MEP1 is enabled. The MEP1 transmits the CCM packets externally. After receiving the first CCM packet, the MEP2, MEP3 and MEP4 in the same maintenance domain start the timer respectively to receive all CCM packets from the MEP1 periodically. Once the link is faulty, the sink MEP fails to receive the CCM packets within a check period. In this case, the sink MEP reports the EX_ETHOAM_CC_LOS alarm. The alarm is not cleared until the link is recovered.
When the CC function is enabled at MEP1, MEP2, MEP3 and MEP4 at the same time, each MEP is both the source end and sink end for the CC function. In this way, the bidirectional continuity test is realized.
NOTE: Only the MEP can enable the continuity check and be the receive respond end for the check.
Based on the regular check mechanism of the CC, the check is performed automatically after it is configured successfully. In this way, the link fault check and fault auto-discovery are realized. Additionally, the broadcast protocol packets generated with the CC are used to realize the point-to-multipoint detection and multipoint-to-multipoint detection, especially the detection of the Layer 2 switching network. In this way, the network detection of the entire maintenance domain is realized.
Loopback Test
The loopback (LB) can be used to test the state of any MP link from the source MEP to the maintenance domain.
Based on the bidirectional service, the loopback is a test performed manually at a time. The source MEP constructs the loopback message (LBM) packets and adds the destination MP (MIP or MEP) IDs to the packets. When the packets are transmitted, the timer is started.
After receiving the LBM packets, the sink MP constructs the loopback return (LBR) packets and transmits them back to the source MEP. In this case, the loopback succeeds. If the source MEP timer expires and fails to receive the LBR from the sink MP, the loopback fails.
As shown in Figure 3, the MEP1 transmits the LBM packets to the sink MEP4. After receiving the packets, the MIP2 and MIP3 in the same maintenance domain detect that the sink MPIDs contained in the packets are different from those of them. In this case, the MIP2 and MIP3 transparently transmit the packets. After receiving the packets, the sink MEP4 transmits the LBR packets back to the source MEP1. At this moment, the loopback is complete.
NOTE: Only the MEP can initiate the loopback test, while both the MEP and MIP can be the receive end in the test.
The LB can be used to test the state of any node link from the source to the maintenance domain. As the sink MP can be the MIP, the loopback can be used to locate the faults. Compared with the continuity test, the loopback test is not a continuous test. Thus, you need start the test manually each time.
Link Trace Test
Compared with the LB test, the link trace (LT) test provides enhanced fault locating capability. That is, the faulty network segment can be located the first time in the LT test.
The source MEP constructs the link trace message (LTM) packets and adds the ID of the sink MEP to the packets for transmission. At the same time, the timer is started.
All MIPs that belong to this maintenance domain in the link continuously transmit the packets to the sink MEP. At the same time, a link trace replay (LTR) packet is transmitted back to the source MEP.
After the sink MEP receives the LTM packets, the packet transmission is complete. Then the sink MEP transmits LTR packets back to the sink. In this case, the link trace test is successful. If the source MEP timer expires and fails to receive the LTR from the sink MEP, the loopback fails.
Additionally, the parameter hop is added to the packet that is transmitted back. The hop is used to indicate the MP ID of the returned maintenance point in the link during the link trace test. If the first MP passed through the LTM is on the same board with the source MP, the hop starts from 0 and accumulates in the sequence. Otherwise, the hop starts from 1.
The function of the link trace test is similar to the function of the loopback test. The difference lies in the response to the LBM frames. In the loopback test, only the sink MP responds to the LBM frames. In the link trace test, all MPs in the link respond to the LTM frames. According to these response messages, all MIPs that are involved from the source MEP to the destination MEP can be recognized, as shown in Figure 4.
- The source MEP1 transmits the LTM packets to the destination MEP4.
- After receiving the LTM packets, the MIP2 transmits the LTR packets to the source MEP1 and forwards the LTM packets at the same time.
- After receiving the LTM packets, the MIP3 transmits the LTR packets to the source MEP1 and forwards the LTM packets at the same time.
- After receiving the LTM packets, the destination MEP4 concludes the LTM packets and transmits the LTR packets to the source MEP1.
NOTE: Only the MEP can initiate the link trace test and be the termination of the test.
Compared with the LB test, the LT test enhances the capability of fault locating.
- According to the returned MIP, the route that involves the protocol packets can be determined.
- Once a link segment where the fault occurs is determined, the link state can be checked according to the returned LTP packets. That is, links that transmit the packets back are normal. In this way, the faulty network segment is located.
Ping Test
The Ping test provides a method to test the in-service packet loss rate and hold-off time that result from bit errors. Besides the continuity test, the Ping test further manages the performance of the Ethernet link at the MAC layer.
The Ping test includes:
- MPID-Ping: In the case that both the Ethernet service processing boards of Huawei equipment at two ends support the IEEE 802.1ag OAM protocol, the ping test is initiated from the maintenance point on the local Ethernet service processing board to that on the opposite board.
- IP-Ping: In the case that both the equipment at two ends supports the ARP protocol and ICMP protocol, only the Huawei equipment can initiate the ping test. Moreover, the Huawei equipment does not respond to the ping test of the opposite end.
The transmit end in Ping acquires the IP address of the maintenance point first, and then constructs the ARP packets and ICMP packets for transmission. The maintenance point that receives the ARP packets or ICMP packets analyzes the packets and then transmits the response packets back to the transmit end. After receiving the response packets, the OAM_Ping transmit end reports the test results of the Ping according to the information contained in the response packets. The results include the packet loss rate and delay information.
The Ping test provides a method to test the in-service packet loss rate and hold-off time that result from bit errors. Compared with the continuity test, the Ping test provides more detailed and accurate OAM information. In this case, the OAM not only ensures the service continuity, but also detects the packet loss and service delay to a greater extent.
