i don´t find them, can you help me? thanks

Greetings,

A recently acquired motherboard has an on-board Intel i219-V NIC. Within a fresh Win7 Pro SP1 x64 install with latest Intel drivers (driver v12.15.23.1, PROSet v21.0.504.0), I am experiencing stability issues with this NIC when a specific combination of factors are met:

1. NIC is set to any Link Speed above 10Mbps Full Duplex (100M-Full, 1G-Full, & Auto Negotiate)

2. Steam (the application from Valve) is downloading application and/or game updates

Basically once Steam starts its updates and reaches about 1MB/s transfer rate, the link drops and the whole computer loses its network connection (LAN and WAN) - all LAN ARP entries are flushed, cannot ping local gateway router, Windows shows the little yellow triangle over the network icon (link but no L2/L3, self-assigned 169 address)

Once this issue occurs, it is resolvable in only two ways:

- Reboot the computer

- Change the Link Speed settings under the NIC's Device Properties page in Device Manager. It doesn't matter what I set it to, just the action of re-setting this parameter is enough to re-negotiate the link again, it seems.

The only workaround that I've found so far is to manually set the NIC Link speed to 10Mbps Full - the NIC is stable (albeit much slower) in this configuration.

Other Troubleshooting & Notes:

- No other application or traffic type seems to result in this behavior. For example, speed tests, iperf tests, torrent, etc all max out my WAN (not LAN) with ease, without the NIC dropping the link

- Replaced relevant CAT6 patch cables (all test good as well)

- Tested different port on LAN switch (NIC is connected directly to an enterprise HP ProCurve switch)

- Tested with switchport speed/duplex settings always matching NIC Link Speed settings (100M-Full, 10M-Full, etc)

Edit: I tested with a USB3.0 Ethernet adapter, and Steam has no issues maxing out the ISP without the link dropping. All other equipment is the same and configured the same.

I see the below message "requested invalid api version" on host ? Will that be an issue or is it benign ? Should I upgrade the host ixgbe to a later version ?

On the host:

[69576.199458] ixgbe 0000:88:00.1 p4p2: VF 2 requested invalid api version 3

root# lsb_release -a

No LSB modules are available.

Distributor ID:    Ubuntu

Description:    Ubuntu 14.04.4 LTS

Release:    14.04

Codename:    trusty

root@mtpnjrsv113:~#

root# ethtool -i p4p2
driver: ixgbe
version: 3.19.1
firmware-version: 0x80000811, 1.1067.0
bus-info: 0000:88:00.1
supports-statistics: yes
supports-test: yes
supports-eeprom-access: yes
supports-register-dump: yes
supports-priv-flags: no

On the guest:

root# lsb_release -a

No LSB modules are available.

Distributor ID:    Ubuntu

Description:    Ubuntu 14.04.2 LTS

Release:    14.04

Codename:    trusty

root# find / -name rte_kni.ko

/var/lib/dkms/dpdk/16.07/3.16.

0-30-generic/x86_64/module/rte_kni.ko
/lib/modules/3.16.0-30-generic/updates/dkms/rte_kni.ko

Hi,i have the following issue. i am using a workstation equipped with Intel X550-T2 card and connected both 10Gbase-T ports with a cat7 cable to switch-port 27/28. switch-port 27/28 are also 10Gbase-T capable.

only one port on the intel card will connect at full speed (10Gbps). the second port will run with 1Gbps. I tried these things:

- build up a lan-team on workstation-side in windows server 2012

- forced card- and switch-ports to 10Gbase

- build a aggregation with both switch-ports

nothing really helped. is there a compatibility problem between card and switch ?

i tried also to active and deactivate stp. i dont understand why only one port run with full speed.

i got a supermicro-mainboard based server in my office as well, two intel x550 onboard ports and the same problem occur. one port is at full speed the second port is at 100Mbps.

hi, i would like to report that the new intel nic drivers version 19.3 that just released recently has huge 6000+ dpc latency spike once in a few secs.

my specs

Intel(R) 82579LM Gigabit Network Connection

windows 7 SP1 32bit + lastest windows update

i downgrade to intel nic drivers previous version 19.1 and the problem is just gone.

Some of my servers become unresponsive during traffic spikes (e.g. small DoS/DDoS attacks, which do not max out our connection.)

The servers become unresponsive because of high CPU usage, after running some tests I saw that only one core is being utilized for 100% during a <200Mbps packet flood. The server stopped replying to ICMP echo/ping.

I think the pictures explain them self.

On this machine we use VMQ, there are 7 Vport available. But according to the Intel driver only two are actually in use.
How can we allow VMQ/network stack on the host OS to use more than 1 core?

We've seen similar issues with servers with RSS load balancing, although the load balancing works better there. (spreads over multiple cores.)
Are there any tips to improve RSS load balancing during traffic spikes? We use Intel Xeon E3 systems with 4 cores + Hyper Threading and one Intel Xeon E5 system with 6 cores and Hyper Threading.

We want to keep the server working during traffic spikes and all round keep our latency to the minimum.

Hi,

I recently bought an i350 ethernet controller. I would like to enable the "critical manageability session" of the card, but I can't find any documentation that explains how to do that. The only document related to my question is https://www-ssl.intel.com/content/dam/www/public/us/en/documents/guides/maintaining-the-ethernet-link-to-the-BMC.pdf#pag…  which says to refer to the i350 documentation. I did... but I was not able to find any concrete information in it. Any hint?

Thanks!

Just built a new computer with a I219-V ethernet device and it has extremely difficulty establishing connection upon boot. upon booting the computer, the network jumps between unplugged and searching for 10 seconds then finally stay in search. It will send about 200 packets (none will be received by the modem) until is slows to almost nothing and stay there forever, never establishing a connection. I have to keep disabling and re-enabling the network adapter for around 30 minutes until I get lucky enough for a connection to be established. Once connection is established everything works smoothly with no problems which makes me think this is a drivers issue over a hardware issue.

A couple things I have found messing around with the system:

1. It only seems to work on a 10 mbps full duplex mode. If I change it from auto negation speed to anything other than 10 mbps it never reaches the searching connection status.

2. If I disable the EEE, the network will keep changing from unplugged, searching, internet established, and then drop to unplugged after 5 seconds, and repeat. I can actually access the internet for about 5 seconds the connection is established before it drops again. This is particularly weird since during stable connection it states that the EEE is not negotiated which makes me believe it is not using it anyway so why would this change the connection pattern.

I know 100% that:

the Cat5 line works properly

drivers are up to date (tried both Gigabyte downloaded drivers and Intel downloaded drivers)

Modem and router is working properly (tried two separate modems and routers)

OS is not messing it up (have tried installing both windows 7 and windows 10)

Updated MoBo bios to the latest version (F6)

GIGABYTE GA-Z170XP-SLI

Intel Core i7-6700K 8M

Actiontec C1000A Modem/router

Does the Intel XL710 4x10GE card support the processing of VLAN frames that have are tagged with STAGs (TPID == 0x88a8) instead of CTAGs (TPID == 0x8100)?   In my lab, I have tried a number of things to enable the processing of STAG frames on the XL710 board, including setting up the ethernet interface with commands such as the one below, but no luck, the interface will still not see the VLAN frames with a TPID of 0x88a8.  Frames tagged with CTAGs are processed just fine however.

$ sudo ip link add link enp1s0f0 name enp1s0f0.4004 type vlan protocol 8021.ad id 4004

Also, I have tried to setup the board in both promiscuous mode and true promiscuous mode.  Again, no joy, the board still refuse to detect/receive a VLAN tagged frame if the TPID is 0x88a8 instead of 0x8100.  By the way, for true promiscuous mode, I am using this command.

$ sudo ethtool --set-priv-flags enp1s0f0 vf-true-promisc-support on

I have the latest Intel firmware installed on this board and am using the latest Intel kernel module for this board.

$ sudo ethtool -i enp1s0f0

[sudo] password for nknuth:

driver: i40e

version: 1.5.16

firmware-version: 5.04 0x800024ca 0.0.0

expansion-rom-version:

bus-info: 0000:01:00.0

supports-statistics: yes

supports-test: yes

supports-eeprom-access: yes

supports-register-dump: yes

supports-priv-flags: yes

Alas, when I query the features of this ethernet board, it seems clear that STAG frames are not supported.

$ sudoethtool -k enp1s0f0

...

tx-vlan-stag-hw-insert: off [fixed]

rx-vlan-stag-hw-parse: off [fixed]

rx-vlan-stag-filter: off [fixed]

...

Any help is appreciated.  It really is a show stopper for my company to not be able to process STAGs on the XL710 boards.  We were using the 2 port 10GE boards that use the ixgbe kernel module and STAG frames were processed on those boards without issue.

Thanks in advance for any clarity here.

Nathan

Hi all. Few days ago I bought the motherboard MSI Night Elf (network adapter Intel I219-V).

I Have a problem with gigabit internet. It doesn't work. In properients i see 1 GBPS, but on speedtest.net i see only 100 mpbs.

It's not a problem with cable or router, because in the past i had not problem with it (before that i have a motherboard ASUS P5Q3 and saw real speed 800 mbps).

I hope to solve this problem and hope for your help.

Thanks a lot!

Hi,

The sed pattern to determine the kernel version is broken. It will end up stripping too much off.

for example /lib/modules/2.6.32-504.16.2.el6.x86_64 will get stripped back to 2.6.32

This is not the intent of the sed pattern and causes install problems as /lib/modules/2.6.32 doesn't exist

Patch

@@ -69,7 +69,7 @@

if [ "%{pcitable}" != "/dev/null" ]; then

fi

-for k in $(sed 's/\/lib\/modules\/\([0-9a-zA-Z_\.\-\+]*\).*/\1/' $FL) ;

+for k in $(sed 's/\/lib\/modules\/\([0-9a-zA-Z_\.\+-]*\).*/\1/' $FL) ;

do

@@ -90,7 +90,7 @@

done

# Check if kernel version rpm was built on IS the same as running kernel

-BK_LIST=$(sed 's/\/lib\/modules\/\([0-9a-zA-Z_\.\-\+]*\).*/\1/' $FL)

+BK_LIST=$(sed 's/\/lib\/modules\/\([0-9a-zA-Z_\.\+-]*\).*/\1/' $FL)

MATCH=no

for i in $BK_LIST

do

I have just tried to update my NIC driver for an I218-V adapter built into my motherboard.  The driver downloads (allegedly version 21.0) & the install begins, but at a certain point it prompts me for a CD-ROM.  I have no idea what I am supposed to supply in the way of a CD-ROM.  I'm attaching a screen snip of the error.

I don't know what to do at this point so I just click Cancel.  That generates this:

OK.  So here I am contacting my technical support group.

HELP . . .

I have been experiencing NIC drops on multiple servers across our organization for over 6 months.  I have had number cases open with Dell and Microsoft unsure if our 10gb switches were causing the issue or the OS or the NIC hardware.  I can confidently say that the common factor is the X710 network card in all cases as we have switched out our 10gb switches to Cisco now and we are still seeing this issue. 

The system event log shows an event 27 - Intel(R) Ethernet Converged Network Adapter X710 #X    Network link is disconnected

There is no rhyme or reason as to when this happens, but I can guarantee at least 1 time per week or more I see this event.

I have a case open with Dell, again, and they had me update the firmware and drivers on the cards to:   Firmware family: 17.5.11  and  Driver Provider Intel, driver date 3/22/2016 1.3.115.0

These are 2 PCIe cards installed on a Dell PowerEdge R430.

During troubleshooting I disabled the following nic features:

* Encapsulated Task Offload

* IPv4 Checksum Offload

* Large Send Offload V2 (IPv4)

* Large Send Offload V2 (IPv6)

* TCP Checksum Offload (IPv4)

* TCP Checksum Offload (IPv6)

* UDP Checksum Offload (IPv4)

* UDP Checksum Offload (IPv6)

* Virtual machine Queues

Since making the changes to the 4 10gb nics on 1 of my 2012R2 Hyper-V hosts I have gone 3 weeks without an Event 27 NIC drop.

I did NOT make the change on my other 2 hosts, 1 hasn't dropped in 3 weeks and the other was good for 2 and now started dropping the last 4 days in a row, sometimes multiple times a day.

I have done hours and hours of research and just don't know if I need all these offloads disabled and what will happen to my virtual machines performance in my cluster if we turn off offloading.

Is there a driver update for these cards so we can enable offloading again or is this the only way these cards work?  Does the disconnect happen because of load on the NIC and that is why it is so random, it shuts itself down if it gets too much traffic?

Thanks in advance for any insight into this problem.

hi, i would like to report that the new intel nic drivers version 19.3 that just released recently has huge 6000+ dpc latency spike once in a few secs.

my specs

Intel(R) 82579LM Gigabit Network Connection

windows 7 SP1 32bit + lastest windows update

i downgrade to intel nic drivers previous version 19.1 and the problem is just gone.

I'm trying to set my Intel 82599 link speed to 1Gb/s using ethtool but it fails with "Invalid argument"

More details of the NIC I'm using- AOC-STGN-i2S 2-port 10GB SFP+ NIC (Intel 82599) and the Finisar transceiver FTLX8571D3BCL.

The NIC is connected via the transceiver and a 10G Duplex Multimode optical fiber cable to a port on an FPGA  that is currently programmed to run at 10G

I would like to do this using a kernel supported driver (ixgbe) and also when using a DPDK enabled driver (igb_uio)

My OS is Linux 4.4.0-34-generic #53-Ubuntu SMP Wed Jul 27 16:06:39 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux

Some questions..

- Once the NIC changes to 1 G, can I expect it to do auto -negotiation and get the FPGA port also to shift to 1 G?

- The ethtool output says the NIC supports a single link mode 10000baseT/Full but the instruction in the datasheet under MAUI Link Setup Flow suggest that it could run at 1G, which one is right? (Also, I don't understand this way of configuring the NIC, and would be glad to get more pointers on this)

- There is a doc that calls the transceiver as a single rate 10G, does that mean

sudo /sbin/ethtool -s ens1f0 autoneg off speed 1000 duplex full

Cannot set new settings: Invalid argument

  not setting speed

  not setting duplex

  not setting autoneg

Here is the output from ethtool

lavanyaj@mule-1:~$ ethtool ens1f0

Settings for ens1f0:

  Supported ports: [ FIBRE ]

  Supported link modes:   10000baseT/Full

  Supported pause frame use: No

  Supports auto-negotiation: No

  Advertised link modes:  10000baseT/Full

  Advertised pause frame use: No

  Advertised auto-negotiation: No

  Speed: 10000Mb/s

  Duplex: Full

  Port: FIBRE

  PHYAD: 0

  Transceiver: external

  Auto-negotiation: off

Cannot get wake-on-lan settings: Operation not permitted

  Current message level: 0x00000007 (7)

        drv probe link

  Link detected: yes

Hi! all

I use 82599EN with 10Gbps SFP module. I measre through put and the result is 1.3Gbps.

I would like to improve through put, so I would like to use MSI-X.

I enter "insmod ixgbe.ko IntMode=2". The result is below,

This driver fail to start because  of "Failed to allocate MSI-X interrupts. Err: -28".

-----

~# insmod ixgbe.ko  IntMode=2

Intel(R) 10 Gigabit PCI Express Network Driver - version 4.3.15

Copyright (c) 1999-2015 Intel Corporation.

ixgbe: Interrupt Mode set to 2

ixgbe 0000:01:00.0: Failed to allocate MSI-X interrupts. Err: -28

ixgbe 0000:01:00.0: Disabling VMQd support

ixgbe 0000:01:00.0: Disabling RSS support

ixgbe 0000:01:00.0: PCI Express bandwidth of 16GT/s available

ixgbe 0000:01:00.0: (Speed:5.0GT/s, Width: x4, Encoding Loss:20%)

ixgbe 0000:01:00.0 eth2: MAC: 2, PHY: 15, SFP+: 5, PBA No: FFFFFF-0FF

ixgbe 0000:01:00.0: 00:a0:c9:12:34:56

ixgbe 0000:01:00.0 eth2: Enabled Features: RxQ: 1 TxQ: 1

ixgbe 0000:01:00.0 eth2: Intel(R) 10 Gigabit Network Connection

platform leds.7: Driver leds-gpio requests probe deferral

-----

Could you tell the reason which occurs "Failed to allocate MSI-X interrupts. Err: -28" ?

BR,

Taira

Hi, I connected server 10GBase-T port to switch 10GBase-T port with Cat.7 STP cable(length 1m).

Two interfaces are up at 1Gbps speed not at 10Gbps.

= Motherboard: Supermicro X10SDV-4C-TLN4F

- CPU: Intel Xeon D-1518

- LAN: 2x 1GbE via i350-AM2

            2x 10GbE via SoC(x552/x557 AT?)

Supermicro | Products | Motherboards | Xeon® Boards | X10SDV-4C-TLN4F

- OS: FreeBSD 11.0-RC1, CentOS x86_64 1511

- Driver: ix-3.1.13k(FreeBSD), ix-3.1.14(CentOS)

I tried ix-3.1.14 driver installation under FreeBSD but failed to compile.

= Switch: HPE OfficeConnect 1950 24G 2SFP+ 2XGT Switch(JG960A)

- 24x 1GbE port(1-24)

- 2x 10GbE port(27, 28)

http://www8.hp.com/us/en/products/networking-switches/product-detail.html?oid=6887535

After connecting two 10GbE port of server mother board with Cat.7 STP cable, both inferfaces are up at 10Gbps.

Also connecting two switch 10GbE port with same cable, both interfaces are up at 10Gbps.

But connecting server port and switch port, all are up at 1Gbps.

It seems auto-negotiation not work well.

Customer looking for available of EPS for Fisher Flat adapter card model XL710-QDA2.

Host OS: Ubuntu 16.04 LTS

Hypervisor: Xen 4.6.0

Kernel version: 4.4.0-22-generic

Driver version: i40e-1.5.18

NVM version: 5.02

Guest OS: Windows Server 2012 R2

Guest Driver version: 20.7.1

I'm trying to forward the ports of the card to a guest Windows VM.  An example of forwarding just one port:

# echo 1 > /sys/class/net/ens2f0/device/sriov_numvfs

# /sbin/ip link set ens2f0 vf 0 mac 02:00:00:00:00:00

# xl pci-assignable-add 02:02.0

Added to Xen VM .cfg:

pci = ['0000:02:02.0']

All of that works and the VM starts fine and sees the card.  Xen reports that it added it:

[26326.586765] i40e 0000:02:00.0 ens2f0: adding 68:05:ca:33:0a:59 vid=0

[26326.586774] i40e 0000:02:00.0 ens2f0: adding 68:05:ca:33:0a:59 vid=1

[26326.595995] i40e 0000:02:00.0: Allocating 1 VFs.

[26326.697778] pci 0000:02:02.0: [8086:154c] type 00 class 0x020000

[26326.697973] pci 0000:02:02.0: Max Payload Size set to 256 (was 128, max 2048)

[26329.893582] i40e 0000:02:00.0: Setting MAC 02:00:00:00:00:00 on VF 0

[26329.965339] i40e 0000:02:00.0: Reload the VF driver to make this change effective.

[26366.200750] pciback 0000:02:02.0: seizing device

[26366.200815] pciback 0000:02:02.0: enabling device (0000 -> 0002)

[26472.867101] xen_pciback: vpci: 0000:02:02.0: assign to virtual slot 0

[26472.867643] pciback 0000:02:02.0: registering for 4

I install the drivers and Windows 'see' the card.  The problem is the reported state goes into a cycle:

Connected @ 10GB

Connected @ 40GB

Disconnected

When this happens, in dmesg on the host:

[26624.137415] i40e 0000:02:00.0: VF 0 failed opcode 11, retval: -10

[26635.075640] i40e 0000:02:00.0: VF 0 failed opcode 11, retval: -10

[26645.903762] i40e 0000:02:00.0: VF 0 failed opcode 11, retval: -10

[26656.669642] i40e 0000:02:00.0: VF 0 failed opcode 11, retval: -10

[26667.904191] i40e 0000:02:00.0: VF 0 failed opcode 11, retval: -10

[26679.013381] i40e 0000:02:00.0: VF 0 failed opcode 11, retval: -10

[26689.966331] i40e 0000:02:00.0: VF 0 failed opcode 11, retval: -10

[26701.028906] i40e 0000:02:00.0: VF 0 failed opcode 11, retval: -10

[26712.372668] i40e 0000:02:00.0: VF 0 failed opcode 11, retval: -10

[26723.154159] i40e 0000:02:00.0: VF 0 failed opcode 11, retval: -10

[26733.935320] i40e 0000:02:00.0: VF 0 failed opcode 11, retval: -10

[26744.685130] i40e 0000:02:00.0: VF 0 failed opcode 11, retval: -10

[26756.075844] i40e 0000:02:00.0: VF 0 failed opcode 11, retval: -10

This same issue happens using the built in kernel i40e and also i40e-1.5.18. Am I doing something wrong?

I need to upgrade an old Win2000 Advanced Server to 10Gb Ethernet with Cat 6A wire.

The motherboard is a Tyan bi-Opteron 2216 with PCI Express connectors.

Has anyone a solution ?

Thanks for the return

Emmanuel