Known Issues

Linux Kernel Upstream Release Notes v6.5

The following is a list of general limitations and known issues of the various components of this MLNX-NATIVE-ESX release.

Internal Ref.



Description: The maximum number of established active RDMA connections (QPs) is currently 5000.

Workaround: N/A

Keywords: QPs, RDMA

Discovered in Version:


Description: Setting ETS value to 0 may cause WQE timeout.

Workaround: Set ETS value of 1 instead of 0.

Keywords: ETS, QOS

Discovered in Version:


Description: Although the max_vfs module parameter range is "0-128", due to firmware limitations, the following are the supported VFs per single port devices:

  • ConnectX-4 / ConnectX-5: up to 127

Workaround: N/A

Keywords: SR-IOV, VFs per port

Discovered in Version:


Description: ECN tunable parameter initialAlphaValue for the Reaction Point protocol cannot be modified.

Workaround: N/A

Keywords: nmlx5 ecn nmlxcli

Discovered in Version:


Description: ECN statistic counters accumulatorsPeriod and

ecnMarkedRocePackets display wrong values and cannot be cleared.

Workaround: N/A

Keywords: nmlx5 ecn nmlxcli

Discovered in Version:


Description: The hardware can offload only up to 256B of headers.

Workaround: N/A

Keywords: Hardware offload

Discovered in Version:


Description: There is no traffic between PV and SR-IOV VF connected to different ports on the same HCA. This issue is applicable to ESXi 6.5 & ESXi 6.5 UP1. The issue is solved in ESXi 6.5 UP2.

Workaround: N/A

Keywords: PV, SR-IOV VF, HCA


Description: Setting the "Allow Guest MTU Change" option in vSphere Client is currently not functional. Although guest MTU changes in SR-IOV are allowed, they do not affect the port's MTU and the guest's MTU remains the same as the PF MTU.

Workaround: N/A

Keywords: MTU, SR-IOV


Description: Geneve options length support is limited to 56B. Received packets with options length bigger than 56B are dropped.


Keywords: Geneve


Description: Running with ConnectX-4/ConnectX-4 Lx older firmware versions, might result in the following internal firmware errors:

  • Device health compromised

  • synd 0x1: firmware internal error

  • extSync 0x94ee

Workaround: Upgrade your firmware to the latest version 12.17.2020/14.17.2020

Keywords: Firmware


Description: The 'esxcli mellanox uplink link info -u <vmnic_name>'

command reports the 'Auto negotiation' capability always as 'true'.

Workaround: N/A

Keywords: 'Auto negotiation' capability


Description: ESXi v4.16.10.3 cannot updated from v4.16.8.8 (GA) or from the Inbox driver using the "esxcli software vib update" command.

Workaround: To update it, run the "esxcli software vib install" command.

Keywords: Driver update


Description: SMP MADs (ibnetdiscover, sminfo, iblinkinfo, smpdump, ibqueryerr, ibdiagnet and smpquery) are not supported on the VFs.

Workaround: N/A

Keywords: SMP MADs


Description: Wake-on-LAN does not notify when invalid parameters are provided.

Workaround: N/A

Keywords: WoL


Description: Nested ESXi might not function properly.

Workaround: N/A

Keywords: Nested ESXi


Description: Device RSS fails to hash traffic to sufficient RX rings with Broadcast traffic.

Workaround: N/A

Keywords: RSS, RX rings


Description: In stress condition ‘Watchdog’ may appear, leading to uplink going up and down.

Workaround: N/A

Keywords: uplink, watchdog

© Copyright 2023, NVIDIA. Last updated on Aug 31, 2023.