Release Notes

OpenFlow does not support the scaled-l3-routes forwarding-table mode (UFT mode 4) on the C9010. To configure the
maximum scalable size for the L2 MAC, L3 host, and L3 route tables, use the hardware forwarding-table mode
command.
● On the C9010, Data Center Bridging (DCB) is supported per-line card. If the traffic handled by a DCB map is transmitted on
ports on different line cards, you must manually configure the DCB map on the backplane ports of the C9010 line cards on
which the ports reside (dcb-map linecard all backplane all command).
If the DCB map you apply to the backplane ports of C9010 RPMs (linecard 10–11) configures two or more priority
groups, you must increase the size of the PFC shared and total buffers (dcb pfc-shared-buffer-size and dcb
pfc-total-buffer-size commands).
● The C1048P does not support more than one (redundant) port channel to the same network device on its front-end ports.
● The C9010 does not support the user datagram protocol (UDP) helper. The UDP helper allows you to forward IP/UDP
broadcast traffic by creating a broadcast address and rewriting the destination IP address of packets with the new
broadcast address.
● The C9010 supports the following Open Automation (OA) features: Bare Metal Provisioning (BMP), Open Management
Infrastructure (OMI), Puppet, Representational State Transfer (REST) application programming interface (API), Smart
Scripting, and Virtual Server Networking (VSN).
● PE loop detection — In the current code, when a loop is detected between PEX ports P1 and P2, the port numbers are
compared numerically and the port which has the least ifIndex stays UP and the other port stays DOWN, and is moved to an
error-disabled state. In most scenarios, bringing down the IfIndex with a large number is not correct. For example, MAC M1
is learnt on PE port P3 and there is a loop because PE port P1 and P2 are connected back-to-back. In this case, if P3 has
a greater ifIndex than P1/P2, the current code will bring down P3. So, irrespective of the ifIndex value, it is appropriate to
keep the P3 UP where the mac was originally learnt first, and bring down the port P1/P2 on detecting a loop.
● Controlling Bridge ports cannot be part of VLAN that is non-spanned across VLT domain.
● Layer 3 physical ports (CB ports) and Layer 3 Port-channel (with members as CB ports) cannot be part of non default VRF.
If they need to be placed in VRF then they can be tagged/untagged to spanned VLT VLAN and this VLAN can be placed
under the VRF.
● FEFD is not a HA-aware protocol. Due to this, the protocol states and corresponding data are not available on the standby
system. While inter-operating with a third-party switch, the FEFD might move into an unknown state on the new RPM after
failover. Due to this, the line protocol might go down.
Changes to Default Behavior and CLI Syntax
Following default behavior and CLI syntax changes occurred during the Dell EMC Networking OS release 9.14(2.10):
None.
Restrictions
● Prerequisite steps to upgrade the Dell EMC Networking OS from earlier version to 9.14.2.0 or later:
1. Uninstall the older version of the Open Automation (OA) package
2. Upgrade the Dell EMC Networking OS to 9.14.2.0 or later version
3. Install the following OA packages from the respective upgraded version:
a. SmartScripts
b. Puppet
c. Open management infrastructure (OMI)
d. SNMP MIB
Prerequisite steps to downgrade the Dell EMC Networking OS from 9.14.2.0 or later to the earlier version:
1. Uninstall the OA package of 9.14.2.0 or later version
2. Downgrade the Dell EMC Networking OS to an earlier version
3. Install the respective OA package from an earlier version
For more information about installing, uninstalling and upgrading the Dell EMC Networking OS and OA package, refer the
respective Dell EMC System Release Notes.
4