Hi,

I found a problem that when adding vlans to a XL710 port only 255 vlans can be configured.

Further configuration will return with error message as "ixl0: aq_add_macvlan err -53, aq_error 16"

-53 means admin queue error and 16 should be no space available or alloc fail.

This is found on FreeBSD OS and with driver 1.4.8 and latest NVM on chip.

Could you please help confirm what issue it is?

Thanks very much.

Regards,

Jingxun

Hello and Intel Support,

I have a problem since several weeks. My LAN connection does not work any more. In the device manager I get a 'Code 10' (Not able to start device).

I have tried already several version of the available driver but this was not successful.

Attached you can find the support file of my system.

Any suggestions??

Thanks in Advance!!

BR

Steffen

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

just upgraded to windows 10.

There is a Intel pro 1000 and in windows 7 after installing Advanced Network services from Intel the VLAN feature was very helpful

after upgrading to Windows 10 the  VLANS are no longer available.

When attempting to install Advanced Network services from Intel,  the following error is displayed

there is an issue with Microsoft* windows* 10 the prevents the Intel(R) Advanced Network Services from working correctly

where can the correct drivers be found, searches have been unsuccessful thus far.

Any help appreciated.

Chris

Hi,

Having issues getting my desktop PC that has an Intel i128-v NIC to connect to my network over TP-Link powerline adaper (TL-PA8010P)

I can confirm that the Powerline adapters are working as my Laptop and HTPC both connect to the network without fault.

On my desktop i have:

Updated the Motherboard BIOS - no change

Installed latest LAN drivers from Intel website - no change

Installed latest LAN drivers from motherboard manufacturer - no change

Tried using a Static IP address - no change

Running out of ideas, hope someone can help?

Hello,

I have several 2012 R2 systems with x710 NICs that blue screen and get network disconnects in the event logs when we have offloading turned on. If we disable offloading, the system is stable. We've tried the Dell version of the drivers, as well as the Intel drivers (v20.7). It appears to be very similar to the issue in the thread x710-DA2 iSCSI Connections Keep Disconnecting .

Is anyone else having problems with offloading on these NICs?

Hello,

I bought a two-port Intel I350 NIC (vendor says "Generic i350-t2 V2"), and I wish to enable its critical session feature (also known as veto bit, keep PHY link up, and manageability session). However, it is not in a classic server installation, but rather in a simple desktop computer - I have my reasons for this, namely to be able to put the computer in suspend-to-RAM sleep state - which means there is no BMC as far as I know.

From the datasheet, I understand the BMC would send through SMBus a specific command to enable this feature. What I'd want is to send this command from the Linux kernel instead. As of now, I see two SMBus devices in '/dev', which are identified as "igb BB", so I guess they correspond to the two ports of the card, but I have not been able to send the command to either device. As a matter of fact, I cannot detect any slave on each device (using command i2cdetect). My try program results in an error, which code from errno is "ENXIO (6): no such device or address" when I send the command. Opening the device and setting the slave address to 0x49 "works" (in the sense it gives me no error).

So Tl;Dr: I want to send a manageability command through SMBus (or NC-SI if it'd work) without BMC, but rather from my Linux kernel or user-space. Is that even possible? If yes, how do I enable the NIC slave device?

Thanks for your help

Hi,

I'm doing the tuning job for XL710 VF on a FreeBSD10 VM (driver is ixlv-1.2.11) hosted on a kvm-hypervisor server (pf driver is i40e-1.4.25).

I found for each queue of the vf the interrupte rate is very high (>10000/sec).

As said in the datasheet, I tried to set I40E_VF_ITRN by the VF driver and set I40E_VPINT_RATEN by the PF driver. But this didn't work.

I have upgraded the NVM to 5.02 the latest version.

Could anybody help me resolve the issue?

Hi, Expert

I've investigated this for some days.

My purpose is launching a VM which has two VFs come from two physical NICs, and then do the bonding inside the VM.

I've tried all the 7 bonding modes, only the mode 1(active-backup) works, all the others cannot work.

From the following articles, the reason why other bonding mode cannot work is VM cannot change the VF's MAC address successfully.

https://bugzilla.redhat.com/show_bug.cgi?id=1164224

SR-IOV on NIC 82599: Are VF MAC Addresses unique?

https://communities.intel.com/message/246183#246183

From the igb driver code, I also notice the HOST refuse VM's change VF's MAC address request.

case E1000_VF_SET_MAC_ADDR:

    retval = -EINVAL;

    if (!(vf_data->flags & IGB_VF_FLAG_PF_SET_MAC))

         retval = igb_set_vf_mac_addr(adapter, msgbuf, vf);

    else

         dev_warn(&pdev->dev, "VF %d attempted to override administratively set MAC address\nReload the VF driver to resume operations\n", vf);

    break;

So, My question is

* Does Intel's SR-IOV NIC only support bonding mode active-backup? only this mode not request VM change it's interface's MAC address.

* If I want to use active-active bonding mode, such as mode 0/2/4/5/6, any solution/walkaround to make it work?

Thanks.

We already got working card with same chip (Selecom) and decided to purchase another one - but Intel branded XL710-QDA2.

After deploying and upgrading firmware and driver to the latests versions, we found that 40gbit/s link is up and icmp (pings) packets go perfectly, but there is no ability to set up any TCP connection.

Ethtool shows a lot of rx_crc_errors on this interface.

We've tested it with two brand new Intel cables and got the same result.

What can be reason of this?

i40e version 1.5.18

nvm version 5.0.4

I´ve got a problem with a server connection using the following combination:

IBM Server SystemX 3450 Typ:7379-K4G (PCIE x8 slot, 25w), running Windows Server2008R2

Intel Network Adapter X520-DA2

Intel TwinAxial Cable 3m (XDACBL3M)

D-Link switch DGS-1510-52

Problem:

10Gbit Link is up, but the performance is like 10Mbit. Gigabit connection with the internal adapter is much faster!

Even when i try to open the management gui of the adapter it takes several seconds to open and everything seems to be very slow.

As i don´t use them i tried to disable the IPv6 but this didn´t have any effect.

Any ideas?

Thank you in advance! :-)

best regards

Stefan Rindler

Hello,

  I had a mainboard with i218-V and I was extremely happy with it. Now as I upgraded to Skylake, I have an i219-V with new mainboard (MSI C236A). I first tried with Windows 10 and now downgraded to Win 8.1 but problem persists on both platforms.

  My problem is (as subject partly indicates) when I am having any file to download, I can't even browse Internet before download is over. So in that sense, there is no QoS at all. I checked Network details and NIC options but no success, I never had such issue before.

Thanks in advance for help.

Is it possible to configure RTTDT1C register of 82599 without setting up the NIC in VMDq or DCB mode please? The background is  - following is verbatim customer's input we are configuring the weights for the NIC Tx Queues by configuring the RTTDT1C register. We don't need VMDq or DCB mode. We are directly enqueuing CustomProto/UDP/IP/Ethernet packets from my application on to specific hardware queues (without an intermediate DPDK QOS layer) and updating the hardware queue rates, so the priority/ hardware rate limiting are the only features I need (not VMDQ, DCB). Can I go about just changing the weights as in the code above with VMDQ and DCB turned off?  For context, I'm working on developing a prototype of a high speed congestion control algorithm for 40G networks where rates for flows are calculated out of band (using control packets at a higher priority) and updated every RTT, and the flow's packets are sent out at the calculated rates. The 82599 NIC's hardware rate limiting and priorities look very promising and I would love to learn how I can use them.

我使用Intel NUC6i7KYK

bios裡面設定有開

但是無法使用wake on lan

使用android app喚醒或是從asus ac56u路由器介面去喚醒都失敗

使用wifi介面或是網路線介面都無效

請問我該如何設定?

OS: windows 10 home edision

Hello

I have a gigabyte mb GA-Q77M-D2H with the intel 82579LM Ethernet chip on board and when the system comes out of sleep state it takes a minute before the LAN starts up again.

System is Windows 10 64bit pro

i7-3770, 8gb ram and Samsung SSD.

From power off the system boots really quick and the LAN is working just fine, however put it into sleep state and Ethernet takes a minute or more to start up again after leaving sleep state. (Yellow LED blinking).

If I go into device manage and disable and then enable the 82579lm it connects immediately.

I have tweeked just about ever config option in the device options and BIOS that I think would affect the slow start up and nothing has helped.

I have also test with several different generation of the intel LAN drivers.

Interesting is that I have the same problem when I install Win 7 64bit pro in the system.

However and this is the bit that makes little sense I also have an ASUS P8Q77-M mother that is essentially the same as the gigabyte and it works just fine and connects immediately to the LAN after sleeping.

I even tried the ASUS drivers in the GB board and problem the same.

Any ideas please as I have tried everything I can think of :-)

Ian K.

The Intel drivers, PROSet, Advanced Network Service, Powershell Module, and Intel SNMP agent are *not* functioning whatsoever in Windows 2012 R2. I've tried an install on multiple machines using the latest 18.7 driver set and a after a clean install of Windows. I am able to get connectivity with limited features, using the built-in Microsoft ver. 12.6.47.1 March, 28, 2013 drivers. The error message at install states "Cannot install drivers. No Intel(R) Adapters are present in this computer." My adapter is built onto the motherboard and resides at PCI bus 4 and 5. To troubleshoot this, I did verify the Windows Device Manager Service was started. I also checked the release notes in several previous releases to see if there are any possible work arounds. No luck. While I realize R2 was just released to the public this past month, it was in beta for quite some time. I hope a fix can be found. The Intel 82574L integrated network adapter is found on *almost* every server grade system I've ever worked on. Whether a supermicro, intel, or asus motherboard - this chip is certainly always present. This is a serious problem for Windows 2012 R2 users. I did not have the same issue on the non-R2 version.

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,

my Supermicro X10SRL-F has two onboard I210 NICs. Two days ago, after having it passed one through to a PFSense VM with KVM, they now show up under Linux as

07:00.0 Ethernet controller: Intel Corporation Device 1531 (rev 03)

08:00.0 Ethernet controller: Intel Corporation I210 Gigabit Network  Connection (rev 03)

Only the second one is detected, the first one is missing in the BIOS, too. I've had the problem before and solved it by re-flashing the firmware which was provided by Supermicro support. I'm not certain about the circumstances back then, but it's possible that it was related to pass-through, too. However, it does not seem to be easily reproducible as the same setting has worked flawlessly several times before. Before trying to re-flash the firmware as advised by Supermicro, I'd like to ask if

- this is a known problem (I couldn't really find related reports) and if yes how it can be avoided

- you could use a firmware dump to get an idea of the problem / cause

Of course, any other helpful information is appreciated! The NIC was running with firmware Rev 1.24 (05/21/2013), they've sent me Rev. 1.27A (8/27/2015) to test.

With regards

N0T3P4D

Hi,

While running our application in windows 2008 R2 server some random BSOD are happening.

On analyzing the dump it points to a call from elr62x64.sys and then fails in tcpip.sys. Since tcpip.sys is a system file, we would like to understand if there is any known bug from NIC card driver.

the NIC card details:

Intel R Gigabit 2P I350-t Adapter.

The current driver version  details in the server machine where BSOD's surfaced :

1) erl62x64.sys - 12.13.27.0 (Date : 6/5/2015)

2) tcpip.sys - 6.1.7601.22950 (Date: 2/5/2015)

3) NDIS.sys - 6.1.7601.23235 (Date : 10/13/2015)

4) ad.sys - 6.1.7601.23237 (Date: 10/14/2015)

Also the call stack obtained from dump file is as follows:

STACK_TEXT: 

fffff800`03650158 fffff800`01e80e69 : 00000000`0000000a 00000000`00000000 00000000`00000002 00000000`00000000 : nt!KeBugCheckEx

fffff800`03650160 fffff800`01e7fae0 : 00000000`00000000 00000000`00000000 fffffa80`481a2000 fffff800`03650640 : nt!KiBugCheckDispatch+0x69

fffff800`036502a0 fffff880`01a6606b : 00000000`00000716 fffff800`03650640 fffffa80`2138dcf0 fffff800`03650640 : nt!KiPageFault+0x260

fffff800`03650430 fffff880`01a643c6 : 00000000`71f8da89 fffffa80`2aaced70 00000000`00000000 fffffa80`2aacee38 : tcpip!TcpSegmentTcbSend+0x17b

fffff800`03650530 fffff880`01a685d9 : fffffa80`2a9d5360 fffff880`01a8c89b fffffa80`1d81ad30 00000000`000025bc : tcpip!TcpBeginTcbSend+0xa66

fffff800`036507b0 fffff880`01a69450 : fffffa80`000000ba 00060000`00000006 00000000`00000000 00000000`00004800 : tcpip!TcpTcbSend+0x1d9

fffff800`03650a30 fffff880`01a681a8 : 00000000`00000000 00000000`00000000 fffff800`03650c50 fffff800`03650d30 : tcpip!TcpEnqueueTcbSendOlmNotifySendComplete+0xa0

fffff800`03650a60 fffff880`01a6836b : fffffa80`00000004 fffffa80`1cc0a924 00000000`00004800 fffff800`03650ba0 : tcpip!TcpEnqueueTcbSend+0x258

fffff800`03650b10 fffff800`01e8e1f8 : fffff800`03650bd8 00000000`00000000 fffffa80`1d1ed7b8 fffff800`03650bd8 : tcpip!TcpTlConnectionSendCalloutRoutine+0x1b

fffff800`03650b40 fffff880`01a6922a : fffff880`01a68350 fffffa80`1d81af20 00000000`00000502 fffff880`01a68e00 : nt!KeExpandKernelStackAndCalloutEx+0xd8

fffff800`03650c20 fffff880`0436b7be : fffffa80`3934f3e0 fffffa80`1fd537e0 fffffa80`1fd537e0 fffffa80`1fd537e0 : tcpip!TcpTlConnectionSend+0x7a

fffff800`03650c90 fffff880`0436f7b4 : fffffa80`1ffa4b50 fffff880`01a70000 fffffa80`389cc280 fffff800`03650f00 : afd!AfdTLTPacketsSend+0x47e

fffff800`03650e20 fffff880`0436fe83 : fffffa80`48c0b110 00000000`00000001 fffffa80`1c78a380 fffffa80`1ffa4cf8 : afd!AfdTPacketsSend+0x64

fffff800`03650e90 fffff880`01a4e9c0 : 00000000`00010000 fffffa80`1b4930a0 fffffa80`1f705e90 fffff800`03651158 : afd!AfdCommonRestartTPacketsSend+0x173

fffff800`03650ec0 fffff880`01a64fa8 : 00000000`00000000 fffff800`03651158 fffffa80`1b4930a0 fffff880`00000000 : tcpip!TcpCompleteTcbSend+0x40

fffff800`03650ef0 fffff880`01a6299a : fffffa80`2138dcf0 fffffa80`38854000 fffffa80`1b0a3400 fffffa80`1b6202a0 : tcpip!TcpTcbReceive+0x3ec

fffff800`03651000 fffff880`01a637fc : fffffa80`1cc457ac fffffa80`1b4930a0 fffffa80`1b4930a0 00000000`00000000 : tcpip!TcpMatchReceive+0x1fa

fffff800`03651150 fffff880`01a4667c : fffffa80`1b4946f0 00000000`00000000 fffffa80`00000004 fffffa80`1c745740 : tcpip!TcpPreValidatedReceive+0x49c

fffff800`03651220 fffff880`01a57712 : fffffa80`1c7c3010 fffffa80`1c783ba0 fffffa80`1c780006 00000000`00000006 : tcpip!IpFlcReceivePreValidatedPackets+0x5bc

fffff800`03651380 fffff800`01e8e1f8 : 00000000`00000000 00000000`00004800 fffff800`0200ecc0 00000000`00000000 : tcpip!FlReceiveNetBufferListChainCalloutRoutine+0xa2

fffff800`036513d0 fffff880`01a57e42 : fffff880`01a57670 fffff880`01a6959a 00000000`00000002 fffffa80`1b0c0d00 : nt!KeExpandKernelStackAndCalloutEx+0xd8

fffff800`036514b0 fffff880`00ee20eb : fffffa80`1c773010 00000000`00000000 fffffa80`1c6bd1a0 00000000`0000f800 : tcpip!FlReceiveNetBufferListChain+0xb2

fffff800`03651520 fffff880`00eabad6 : 00000000`00000000 00000000`00000000 00000000`00000000 00000000`00000000 : NDIS!ndisMIndicateNetBufferListsToOpen+0xdb

fffff800`03651590 fffff880`00e25ac1 : fffffa80`1c6bd1a0 00000000`00000002 00000000`00000004 00000000`00000000 : NDIS!ndisMDispatchReceiveNetBufferLists+0x1d6

fffff800`03651a10 fffff880`05382f33 : fffffa80`481a2000 00000000`00000004 fffffa80`1a00caa0 fffffa80`481a2dc0 : NDIS!NdisMIndicateReceiveNetBufferLists+0xc1

fffff800`03651a60 fffff880`05383101 : 00000000`00000001 fffffa80`1a00caa0 fffffa80`1c6c3e20 fffffa80`481a2000 : e1r62x64+0x23f33

fffff800`03651aa0 fffff880`053746e6 : 00000000`00000000 fffffa80`481a2040 00000000`00008000 ffff821b`00000000 : e1r62x64+0x24101

fffff800`03651b20 fffff880`053744a0 : fffffa80`1b8b7730 ffff0001`00000000 ffff0001`00000000 fffff880`00eab488 : e1r62x64+0x156e6

fffff800`03651b90 fffff880`05376628 : 00000000`00000000 ffff0001`00000000 00000000`00000000 00000000`00000000 : e1r62x64+0x154a0

fffff800`03651c00 fffff880`00e25951 : 00000000`001bbf1e 00000000`00000000 00000000`00000000 fffff800`0364b080 : e1r62x64+0x17628

fffff800`03651c40 fffff800`01e8d1dc : fffffa80`1cbb9918 00000023`00000000 00000000`00000000 fffff800`02000e80 : NDIS!ndisInterruptDpc+0x151

fffff800`03651cd0 fffff800`01e795ca : fffff800`02000e80 fffff800`0200ecc0 00000000`00000000 fffff880`00e25800 : nt!KiRetireDpcList+0x1bc

fffff800`03651d80 00000000`00000000 : fffff800`03652000 fffff800`0364c000 fffff800`03651d40 00000000`00000000 : nt!KiIdleLoop+0x5a

Please let us know if we were using older version of driver which leads to this error. If so please specify where to get the latest.

Also from the windows event viewer we got event as Microsoft.WIndows.Kernal.General with event ID as 12 sometimes and 41 sometimes and even 1 sometimes.

thanks.

Muralidhar

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