Resolved Issues

Resolved Issues in 19.1R3

• On VMX instances running in SR-IOV (single-root I/O virtualization) mode with vlan-offload enabled, the input-vlan-map option might not work correctly. PR1460544

• When an interface on vMX receives a malformed packet, it might cause a PF reset. This issue causes the interface to stop forwarding traffic. PR1483224

• When TCP-MSS is configured on dynamic-profile or enabled for plain IP traffic, only the initial SYN packet might be marked correct with configured TCP-MSS rest SYNC packets are not marked correctly. PR1475381

Resolved Issues in 19.1R2

This section lists the resolved issues in the release 19.1R2.

• On VMX instances deployed in subscriber scenario, RIOT core might be observed on VMX instances. The respective FPC might reboot after generating RIOT core. All the traffic or service related to that FPC are impacted. PR1393660

• On VMX platform, when the value of the "remote-gateway" is changed (from example, it is changed from a.a.a.a to b.b.b.b) on the device acting as the initiator of IPSec VPN tunnel, the device might still use value a.a.a.a to negotiate the IKE-phase-1 with its peer, which will lead to the failure of the negotiation, thus the establishment of IPSec-VPN tunnel will fail. PR1421977

• On vMX instances, continuous disk error logs requesting for switchover due to disk failure on ada1 are seen on vCP console. PR1424771

Resolved Issues in 19.1R1

This section lists the issues fixed in the release 19.1R1.

• On vMX platform, when a physical function (PF) resets, the underlying virtual function (VF) must also be reset to continue data path functionality. There are no options available for registering to receive PF reset notification. Consequently whenever something happens on the PF causing a reset, the data path functionality halts till SR-IOV is restarted (that is FPC restart in order to recover). When PF reset takes place, it sends a reset message to all associated VFs and attempts to stop and restart again. In case of i40e drivers, VFs are not able to reset properly. PR1382787

• On vMX system, when you configure large number of interfaces, the vFPC CPU utilization might go very high periodically because of interface statistics collection running repeatedly. PR1385853

• On vMX platform, kernel core files are generated when the kernel state (ifstates) exceptions occur. PR1398320

• While using the IXGBE driver with SR-IOV, during the interface flap at the peer device, the vMX occasionally stops receiving traffic. You must restart the FPC to recover the functionality on the affected interface. PR1401672