No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>Search

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

CX11x, CX31x, CX710 (Earlier Than V6.03), and CX91x Series Switch Modules V100R001C10 Configuration Guide 12

The documents describe the configuration of various services supported by the CX11x&CX31x&CX91x series switch modules The description covers configuration examples and function configurations.
Rate and give feedback:
Huawei uses machine translation combined with human proofreading to translate this document to different languages in order to help you better understand the content of this document. Note: Even the most advanced machine translation cannot match the quality of professional translators. Huawei shall not bear any responsibility for translation accuracy and it is recommended that you refer to the English document (a link for which has been provided).
Principle Description

Principle Description

This topic describes the basic concepts and principles of the Smart Channel technology.

Basic Concepts

This topic describes the basic concepts of Smart Channels.

A Smart Channel contains two LAGs. After physical ports are added to the LAGs, the link statues dynamically change through LACP. Only the active LAG forward data, and other LAGs are blocked. In this way, network loops and broadcast storms are prevented.

Smart Channel

A Smart Channel contains one or multiple ports on switch module panels. The E9000 connects to data center TOR switches through these ports. Unlike Eth-Trunk, a Smart Channel can contain ports of different rates. In Channel mode, Smart Channel 1 is automatically created after the system is started. All panel ports are added to Smart Channel 1. You can also create or delete Smart Channels using the CLI. When a Smart Channel is created, two link aggregation groups (LAGs), Trunk A and Trunk B, are automatically created. You can manually add physical ports in the Smart Channel to the LAGs, or the ports are added to the LAGs based on the Link Aggregation Control Protocol (LACP) status.

LAG

Trunk A and Trunk B are LAGs automatically created when a Smart Channel is created. You can enter the port view and set parameters such as the port type, VLAN, and QoS. You can add panel ports to a Smart Channel as required. You can use commands to manually add the ports to Trunk A or Trunk B, or enable the ports to be added automatically based on the port LACP status and port priority rules. Trunk A and Trunk B work in active/standby mode by default. VLANs of the active LAG forward data, while those of the standby LAG are blocked and can receive and send only LACP BPDU packets. When the active LAG fails (that is, Eth-Trunk status is down), a failover occurs. The standby LAG becomes active and starts forwarding data.

Member Port

Smart Channel ports that have not been assigned to Trunk A or Trunk B are member ports. All member ports are blocked and can only receive or send LACP BPDU packets. If you specify a LAG (Trunk A or Trunk B) when adding a port to a Smart Channel, the port becomes a trunk port. Otherwise, the port becomes a member port. When the member port goes from down to up, it may be added to Trunk A or Trunk B based on port priority rules (LACP status, port rate, and port number) and become a trunk port.

Trunk Port

Ports can be added to Trunk A or Trunk B in two ways: manual assignment or automatic assignment based on port LACP status, port priorities, and LAG statuses. A trunk port is in the same forwarding status as the LAG where it resides.

Basic Rules

This topic describes the basic principles of Smart Channels.

Channel Mode

In the user view of the CLI, run the following command to set the working mode:

working mode channel | switch

  • Channel mode: Ports on the switch module can be added to Smart Channels.
  • Switch mode: The switch module works as a common switch and does not support Smart Channels. STP is usually used to prevent loops.
Link Aggregation Mode
  • Manual link aggregation mode: You need to manually create LAGs, and manually add ports to LAGs. LACP is not involved. All active links forward data and traffic is distributed using the specified hash algorithm. If an active link is faulty, the remaining active links evenly share the traffic.
  • Static LACP link aggregation mode: LACP defined by IEEE 802.3ad is a protocol for implementing dynamic link aggregation and disaggregation. LACP communicates with peer devices using the link aggregation control protocol data units (LACPDUs). After ports are added to a LAG, the ports send LACPDUs to inform the peer devices of their system priorities, MAC addresses, port priorities, port numbers, and operation keys (used to determine whether the peer devices are in the same LAG and whether the port bandwidths are the same). After receiving the information, the peer devices compare the information with their own port information, determine the ports that can be aggregated and supported by the local ports, and form active links. IEEE802.3ad defines the following two priorities:
    • System LACP priority: A smaller value indicates a higher priority. The end with a higher priority determines the active ports. The end with a lower priority accepts the determined active links.
    • Port LACP priority: indicates the priority of ports in the same LAG. A larger value indicates a higher priority. A port with a higher priority will be selected as the active port.
    NOTE:

    If switch modules use Smart Channels to connect to Cisco switches with port channels configured, set the port channel mode to active or on. Smart Channels cannot connect to passive port channels. For details about how to set the port channel mode, see related Cisco documents.

    After a port is added to a LAG and goes from down to up, LACP negotiation starts. Figure 11-88 shows the negotiation process.
    Figure 11-88 LACP negotiation

    If the LACP negotiation succeeds, the port starts forwarding LAG data with its status changed from inactive to active.

  • Dynamic LACP link aggregation mode: Ports are added to LAGs of a Smart Channel based on the LACP status of peer devices. After a port is added to a LAG and goes from down to up, LACP negotiation starts. The negotiation process is the same as that of the static LACP link aggregation mode, but the actions performed after negotiation fails are different. If LACP negotiation fails in dynamic LACP mode, the port inherits the VLAN attributes of the LAG, enters the Indep state, and independently forwards L2 data.
Adding Member Ports to LAGs
If you add a port to a Smart Channel without specifying a LAG, the Smart Channel adds the port to a LAG based on the port LACP status, port rate, and port number. If both LAGs in a Smart Channel do not contain ports, both LAGs are down. The Smart Channel adds qualified member ports to the standby LAG. When the standby LAG is up, an active/standby switchover occurs. For newly created LAGs without ports, Trunk A is standby and Trunk B is active by default. For an existing LAG, its active/standby status is the status of the last port that exits the LAG. Only member ports in the up state can be added to LAGs. The following are priority rules for adding ports to LAGs.
  1. High-bandwidth port first: ports with a higher bandwidth are added first.
  2. LACP port first: Ports with LACP enabled are added first.
  3. Small-number port first: Ports with a smaller port number are added first. The system compares the slot numbers, subcard numbers, and port numbers in sequence. For example, port 10GE 2/17/1 has a higher priority than port 10GE 3/17/2.

The preceding rules are used in sequence to determine the ports to be added first. For example, if the Smart Channel member ports in the up state include an LACP port and a common port, the LACP port is added to a LAG first. If Trunk A or Trunk B already includes an LACP port, the ports to be added must have the same Actor_System and Actor_Key as the added port, and the LAG must be in static LACP mode to perform LACP negotiation. If LACP is disabled for the peer device of a LAG port, the Smart Channel sets the LAG working mode to the manual mode, and the LAG can contain only one port.

If both LAGs contain ports, if a new member port has a higher priority than a port in the standby LAG, the port in the standby LAG can be replaced, that is, the new member port is added to the standby LAG, and the standby LAG port becomes a member port.

Usually, ports added to the same Smart Channel are of the same type. The high-bandwidth port first policy is not commonly used. The small-number port first policy is used for facilitating network troubleshooting. Real applications usually use 1+1 redundancy. 1+N redundancy is rarely used. LAG port replacement is also rarely used. Two LAGs can meet the requirements of most scenarios.

NOTE:
In Channel mode, panel ports cannot be used as stack or FCoE ports.
Removing a Port from a LAG

A trunk port is removed from the trunk (LAG) if any of the following occurs: the trunk port status goes from up to down; the trunk port is deleted using the CLI; the trunk port is replaced. If a trunk port becomes inactive due to a low LACP priority, it will not be removed from the LAG.

Active/Standby Switchover
An active/standby switchover occurs in the following situations:
  • If both LAGs in a Smart Channel contain ports, and the active LAG fails, a switchover occurs. The standby LAG becomes active and forwards data. The active LAG becomes standby and is blocked. If a port of the original active LAG goes down, the port exits the LAG and changes from a trunk port to a member port. If all ports of the original active LAG become inactive, the ports remain in the LAG.
  • Besides the switchover triggered by a failure, a switchover also occurs if the standby LAG is higher priority than the active LAG. In the Smart Channel view of the CLI, you can set the switchover delay time using the switch delay command. You can also disable switchovers caused by high standby LAG priority so that switchovers occur only when the active LAG fails. Priority-triggered switchovers have priority rules similar to adding ports to LAGs. The system compares the LAG bandwidths, LACP enablement status, and smallest port numbers in sequence.
Fault Detection

Fault detection enables a faulty Smart Channel to inform the blade servers that reside in the same VLAN as the Smart Channel. The blade servers perform traffic switchover based on the NIC bonding mode to ensure network connectivity.

Translation
Download
Updated: 2019-08-09

Document ID: EDOC1000041694

Views: 58609

Downloads: 3621

Average rating:
This Document Applies to these Products
Related Version
Related Documents
Share
Previous Next