Dell PowerSwitch S4820T Configuration Guide for the S4820T System 9.100.0 - Page 895

• If the link between the VLT peer switches is established, changing the VLT system MAC address or the VLT unit-id causes the link between the VLT peer switches to become disabled. However, removing the VLT system MAC address or the VLT unit-id may disable the VLT ports if you happen to configure the unit ID or system MAC address on only one VLT peer at any time. • If the link between VLT peer switches is established, any change to the VLT system MAC address or unit-id fails if the changes made create a mismatch by causing the VLT unit-ID to be the same on both peers and/or the VLT system MAC address does not match on both peers. • If you replace a VLT peer node, preconfigure the switch with the VLT system MAC address, unit-id, and other VLT parameters before connecting it to the existing VLT peer switch using the VLTi connection. • If the size of the MTU for VLTi members is less than 1496 bytes, MAC addresses may not be synced. Dell Networking recommends retaining the default MTU allocation (1554 bytes) for VLTi members. • VLT backup link • In the backup link between peer switches, heartbeat messages are exchanged between the two chassis for health checks. The default time interval between heartbeat messages over the backup link is 1 second. You can configure this interval. The range is from 1 to 5 seconds. DSCP marking on heartbeat messages is CS6. • In order that the chassis backup link does not share the same physical path as the interconnect trunk, Dell Networking recommends using the management ports on the chassis and traverse an out-of-band management network. The backup link can use user ports, but not the same ports the interconnect trunk uses. • The chassis backup link does not carry control plane information or data traffic. Its use is restricted to health checks only. • Virtual link trunks (VLTs) between access devices and VLT peer switches • To connect servers and access switches with VLT peer switches, you use a VLT port channel, as shown in Overview. Up to 48 portchannels are supported; up to 16 member links are supported in each port channel between the VLT domain and an access device. • The discovery protocol running between VLT peers automatically generates the ID number of the port channel that connects an access device and a VLT switch. The discovery protocol uses LACP properties to identify connectivity to a common client device and automatically generates a VLT number for port channels on VLT peers that connects to the device. The discovery protocol requires that an attached device always runs LACP over the port-channel interface. • VLT provides a loop-free topology for port channels with endpoints on different chassis in the VLT domain. • VLT uses shortest path routing so that traffic destined to hosts via directly attached links on a chassis does not traverse the chassis-interconnect link. • VLT allows multiple active parallel paths from access switches to VLT chassis. • VLT supports port-channel links with LACP between access switches and VLT peer switches. Dell Networking recommends using static port channels on VLTi. • If VLTi connectivity with a peer is lost but the VLT backup connectivity indicates that the peer is still alive, the VLT ports on the Secondary peer are orphaned and are shut down. • In one possible topology, a switch uses the BMP feature to receive its IP address, configuration files, and boot image from a DHCP server that connects to the switch through the VLT domain. In the port-channel used by the switch to connect to the VLT domain, configure the port interfaces on each VLT peer as hybrid ports before adding them to the port channel (see Connecting a VLT Domain to an Attached Access Device (Switch or Server)). To configure a port in Hybrid mode so that it can carry untagged, single-tagged, and double-tagged traffic, use the portmode hybrid command in Interface Configuration mode as described in Configuring Native VLANs. • For example, if the DHCP server is on the ToR and VLTi (ICL) is down (due to either an unavailable peer or a link failure), whether you configured the VLT LAG as static or LACP, when a single VLT peer is rebooted in BMP mode, it cannot reach the DHCP server, resulting in BMP failure. • Software features supported on VLT port-channels • In a VLT domain, the following software features are supported on VLT port-channels: 802.1p, ingress and egress ACLs, BGP, DHCP relay, IS-IS, OSPF, active-active PIM-SM, PIM-SSM, VRRP, Layer 3 VLANs, LLDP, flow control, port monitoring, jumbo frames, IGMP snooping, sFlow, ingress and egress ACLs, and Layer 2 control protocols RSTP and PVST only. NOTE: Peer VLAN spanning tree plus (PVST+) passthrough is supported in a VLT domain. PVST+ BPDUs does not result in an interface shutdown. PVST+ BPDUs for a nondefault VLAN is flooded out as any other L2 multicast packet. On a default VLAN, RTSP is part of the PVST+ topology in that specific VLAN (default VLAN). • In a VLT domain, ingress and egress QoS policies are supported on physical VLT ports, which can be members of VLT port channels in the domain. • Ingress and egress QoS policies applied on VLT ports must be the same on both VLT peers. • Apply the same ingress and egress QoS policies on VLTi (ICL) member ports to handle failed links. • For detailed information about how to use VRRP in a VLT domain, see the following VLT and VRRP interoperability section. • For information about configuring IGMP Snooping in a VLT domain, see VLT and IGMP Snooping. • All system management protocols are supported on VLT ports, including SNMP, RMON, AAA, ACL, DNS, FTP, SSH, Syslog, NTP, RADIUS, SCP, TACACS+, Telnet, and LLDP. Virtual Link Trunking (VLT) 895