Azure Virtual Machines
An Azure service that is used to provision Windows and Linux virtual machines.
8,294 questions
Thanks for replying.
It's customized Linux based on ubuntu rootfs.
kernel 5.15.53
linux vm dpdk test error "hn_vf_add(): RNDIS reports VF but device not found, retrying"
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(134.4, 10%, 23%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EJ%3C/text%3E%3C/svg%3E)
Jimmy
0
Reputation points
I built a customized linux vm, when I run dpdk test with dpdk-testpmd, it shows error as
"hn_vf_attach(): Couldn't find port for VF
hn_vf_add(): RNDIS reports VF but device not found, retrying"
the full log as below.
app/dpdk-testpmd -l 1-3 --vdev="$BUS_INFO,mac=$MANA_MAC" -- --forward-mode=txonly --auto-start --txd=128 --rxd=128 --stats 2
EAL: Detected CPU lcores: 8
EAL: Detected NUMA nodes: 1
EAL: Detected static linkage of DPDK
EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
EAL: Selected IOVA mode 'PA'
EAL: VFIO support initialized
EAL: Probe PCI driver: mlx5_pci (15b3:1016) device: deed:00:02.0 (socket 0)
mlx5_common: DevX read access NIC register=0X9055 failed errno=22 status=0 syndrome=0
mlx5_net: No available register for sampler.
mlx5_common: DevX create q counter set failed errno=22 status=0 syndrome=0
mlx5_common: Key "mac" is unknown for the provided classes.
EAL: Requested device deed:00:02.0 cannot be used
EAL: Bus (pci) probe failed.
hn_vf_attach(): Couldn't find port for VF
hn_vf_add(): RNDIS reports VF but device not found, retrying
TELEMETRY: No legacy callbacks, legacy socket not created
Set txonly packet forwarding mode
Auto-start selected
testpmd: create a new mbuf pool <mb_pool_0>: n=163456, size=2176, socket=0
testpmd: preferred mempool ops selected: ring_mp_mc
Warning! port-topology=paired and odd forward ports number, the last port will pair with itself.
Configuring Port 0 (socket 0)
hn_vf_attach(): Couldn't find port for VF
hn_vf_add(): RNDIS reports VF but device not found, retrying
Port 0: 00:0D:3A:34:69:DA
Checking link statuses...
Done
No commandline core given, start packet forwarding
txonly packet forwarding - ports=1 - cores=1 - streams=1 - NUMA support enabled, MP allocation mode: native
Logical Core 2 (socket 0) forwards packets on 1 streams:
RX P=0/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
txonly packet forwarding packets/burst=32
packet len=64 - nb packet segments=1
nb forwarding cores=1 - nb forwarding ports=1
port 0: RX queue number: 1 Tx queue number: 1
Rx offloads=0x0 Tx offloads=0x0
RX queue: 0
RX desc=128 - RX free threshold=0
RX threshold registers: pthresh=0 hthresh=0 wthresh=0
RX Offloads=0x0
TX queue: 0
TX desc=128 - TX free threshold=0
TX threshold registers: pthresh=0 hthresh=0 wthresh=0
TX offloads=0x0 - TX RS bit threshold=0
Port statistics ====================================
######################## NIC statistics for port 0 ########################
RX-packets: 0 RX-missed: 0 RX-bytes: 0
RX-errors: 0
RX-nombuf: 0
TX-packets: 0 TX-errors: 0 TX-bytes: 0
Throughput (since last show)
Rx-pps: 0 Rx-bps: 0
Tx-pps: 0 Tx-bps: 0
############################################################################
hn_vf_attach(): Couldn't find port for VF
hn_vf_add(): RNDIS reports VF but device not found, retrying
hn_vf_attach(): Couldn't find port for VF
hn_vf_add(): RNDIS reports VF but device not found, retrying
**
dmesg log as:
[ 6.151804] IPv6: ADDRCONF(NETDEV_CHANGE): enp2s0: link becomes ready
[ 8.609137] mlx5_core 75d3:00:02.0 enP30163s1: Disabling LRO, not supported in legacy RQ
[ 8.610859] mlx5_core 75d3:00:02.0 enP30163s1: Disabling LRO, not supported in legacy RQ
[ 8.612864] mlx5_core 75d3:00:02.0 enP30163s1: Disabling LRO, not supported in legacy RQ
[ 8.613831] mlx5_core 75d3:00:02.0 enP30163s1: Disabling LRO, not supported in legacy RQ
[ 8.821302] loop0: detected capacity change from 0 to 8
[ 8.842848] mlx5_core 80bc:00:02.0 enP32956s2: Disabling LRO, not supported in legacy RQ
[ 8.844700] mlx5_core 80bc:00:02.0 enP32956s2: Disabling LRO, not supported in legacy RQ
[ 8.847495] mlx5_core 80bc:00:02.0 enP32956s2: Disabling LRO, not supported in legacy RQ
[ 8.849066] mlx5_core 80bc:00:02.0 enP32956s2: Disabling LRO, not supported in legacy RQ
[ 9.035447] kauditd_printk_skb: 3 callbacks suppressed
[ 9.035451] audit: type=1400 audit(1738022329.631:14): apparmor="STATUS" operation="profile_replace" profile="unconfined" name="/usr/lib/snapd/snap-confine" pid=1035 comm="apparmor_parser"
[ 9.064196] audit: type=1400 audit(1738022329.659:15): apparmor="STATUS" operation="profile_replace" profile="unconfined" name="/usr/lib/snapd/snap-confine//mount-namespace-capture-helper" pid=1035 comm="apparmor_parser"
[ 9.067151] mlx5_core deed:00:02.0 enP57069s3: Disabling LRO, not supported in legacy RQ
[ 9.068993] mlx5_core deed:00:02.0 enP57069s3: Disabling LRO, not supported in legacy RQ
[ 9.071117] mlx5_core deed:00:02.0 enP57069s3: Disabling LRO, not supported in legacy RQ
[ 9.072348] mlx5_core deed:00:02.0 enP57069s3: Disabling LRO, not supported in legacy RQ
[ 10.017708] fbcon: Taking over console
[ 10.017853] Console: switching to colour frame buffer device 128x48
[ 51.026576] hv_balloon: Max. dynamic memory size: 32768 MB
[ 534.610255] RPC: Registered named UNIX socket transport module.
[ 534.610260] RPC: Registered udp transport module.
[ 534.610261] RPC: Registered tcp transport module.
[ 534.610261] RPC: Registered tcp NFSv4.1 backchannel transport module.
[ 535.082260] RPC: Registered rdma transport module.
[ 535.082264] RPC: Registered rdma backchannel transport module.
[ 1115.762368] hv_netvsc 000d3a34-69da-000d-3a34-69da000d3a34 enp2s2: Data path switched from VF: enP57069s3
[ 1115.825184] hv_vmbus: registering driver uio_hv_generic
[ 1115.825710] hv_netvsc 000d3a34-69da-000d-3a34-69da000d3a34 enp2s2: VF unregistering: enP57069s3
[ 1115.825719] mlx5_core deed:00:02.0 enP57069s3: Disabling LRO, not supported in legacy RQ
[ 1164.343188] mlx5_core deed:00:02.0 enP57069s3: Link up
{count} votes
-
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(25.6, 0%, 12%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ESW%3C/text%3E%3C/svg%3E)
Silvia Wibowo 5,126 Reputation points Microsoft Employee2025-01-28T01:25:11.9666667+00:00 Hi @Jimmy , what linux and kernel version is your OS?
-
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(64, 38%, 16%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EMS%3C/text%3E%3C/svg%3E)
Markapuram Sudheer Reddy 425 Reputation points Microsoft Vendor2025-01-28T18:00:10.5666667+00:00 Hi Jimmy ,
Thank You for reaching out to Microsoft Q&A forum.
The error your are facing when you execute the command can be due to a compatibility issues between your kernel version, DPDK setup and network drivers.
- The mlx5 driver may have compatibility issues with your current kernel version (5.15.53)
- RNDIS driver detects the Virtual Function on a network interface card (NIC) and it is unable to effectively manage or use that VF. If RNDIS driver cannot properly utilize a VF due to configuration issues or driver incompatibility, it may lead to errors.
- In the error logs, it is observed that, Mellanox NIC is using legacy RQ, which does not support LRO.
- Ensure that the necessary RDMA packages are configured whenever you are using Mellanox NICs with DPDK.
Ensure that your NIC is bound to the appropriate DPDK driver i.e mlx5.
You can try testing with a newer kernel version if the current kernel version facing any bugs affecting functionality of drivers.
Make sure that RNDIS driver has been installed properly and it should be compatible with your kernel version.
Please follow below documentation for more information;
https://learn.microsoft.com/en-us/azure/virtual-network/setup-dpdk?tabs=redhat
https://doc.dpdk.org/guides/linux_gsg/sys_reqs.html#running-dpdk-applications
https://learn.microsoft.com/en-us/azure/virtual-network/setup-dpdk-mana
If the information is helpful, please click on "Upvote"
If you have any queries, please let us know, we are glad to help you.
-
Jimmy 0 Reputation points
2025-01-28T23:11:18.12+00:00 Thank you for replying.
- which kernel version has azure dpdk support, we would like to stick to 5.15.xx if possible. (as same version with our other products). appreciate if you can share the git repo link and branch/tag.
- how to get dependency lib like rdma? do I need to build from https://github.com/linux-rdma/rdma-core or I can install as "apt-get install -y libibverbs-dev rdma-core"
- please comment if we need other libs.
thanks
-
Markapuram Sudheer Reddy 425 Reputation points Microsoft Vendor
2025-01-29T16:51:28.5733333+00:00 Hi Jimmy ,
Thanks for the response.
Please use below github url for azure dpdk support information.
https://github.com/OpenDataPlane/odp-dpdk/blob/master/DEPENDENCIES
You can install rdma dependency lib either via package manager apt or from source until your kernel version supports rdma features and those libraries which were installed should be compatible with a kernel version.
Ensure 'vfio-pci' module is loaded and it should be bind it to your NIC.
sudo ./dpdk-devbind.py --status
sudo ./dpdk-devbind.py --bind=vfio-pci <your PCI address of your NIC>
If you have any queries, please let us know.
Sign in to comment
2 answers
Sort by: Most helpful
-
-
Jimmy 0 Reputation points
2025-01-29T23:22:39.2566667+00:00 Thank you for replying.
you suggested "sudo ./dpdk-devbind.py --bind=vfio-pci <your PCI address of your NIC>", is the PCI address of the VF?
Based on document below doc a), DPDK applications must run over the master PMD (NetVSC PMD) that is exposed in Azure. If the application runs directly over the VF PMD, it doesn't receive all packets that are destined to the VM, since some packets show up over the synthetic interface.
I followed the instructs in doc b), I use uio_hv_generic instead of you suggested vfio-pci, please help clarify this.
I used the following instructions.
modprobe uio_hv_generic
echo $NET_UUID > /sys/bus/vmbus/drivers/uio_hv_generic/new_id
echo $DEV_UUID > /sys/bus/vmbus/drivers/hv_netvsc/unbind
echo $DEV_UUID > /sys/bus/vmbus/drivers/uio_hv_generic/bind
dpdk-testpmd -l 1-3 --vdev="$BUS_INFO,mac=$MANA_MAC" -- --forward-mode=txonly --auto-start --txd=128 --rxd=128 --stats 2
a)https://learn.microsoft.com/en-us/azure/virtual-network/setup-dpdk?tabs=redhat
b)https://learn.microsoft.com/en-us/azure/virtual-network/setup-dpdk-mana