Continuous 10G port flapping (enable/disable) observed with the X710 card

[sriram-sm]

Hi,     
We are currently using the Intel X710 10Gb NIC adapter as part of the HPE Proliant DL360 Gen10 server.    

Of late, we are seeing a continuous flapping of the 10G port and in the kernel logs we are observing the below - 

Aug 31 12:38:34  kernel: i40e 0000:12:00.0 ens1f0: tx_timeout: VSI_seid: 387, Q 14, NTC: 0x120, HWB: 0x1e2, NTU: 0x1e2, TAIL: 0x1e2, INT: 0x0

Aug 31 12:38:34  kernel: i40e 0000:12:00.0 ens1f0: tx_timeout recovery level 1, hung_queue 14

Aug 31 12:38:36  systemd-networkd[1436]: ens1f0: Lost carrier

Aug 31 12:38:37  kernel: i40e 0000:12:00.0: FW LLDP is disabled, attempting SW DCB

Aug 31 12:38:37  kernel: i40e 0000:12:00.0: SW DCB initialization succeeded.

Aug 31 12:38:37  kernel: irq 64: Affinity broken due to vector space exhaustion.

Aug 31 12:38:37  kernel: irq 65: Affinity broken due to vector space exhaustion.

Aug 31 12:38:37  kernel: irq 67: Affinity broken due to vector space exhaustion.

Aug 31 12:38:37  kernel: irq 70: Affinity broken due to vector space exhaustion.

Aug 31 12:38:37  kernel: irq 71: Affinity broken due to vector space exhaustion.

Aug 31 12:38:37  kernel: irq 72: Affinity broken due to vect

...

...

Aug 31 12:38:37  kernel: irq 132: Affinity broken due to vector space exhaustion.

Aug 31 12:38:37  kernel: i40e 0000:12:00.0: VF BW shares not restored

Aug 31 12:38:37  systemd-networkd[1436]: ens1f0: Gained carrier

Aug 31 12:38:37  systemd-networkd[1436]: ens1f0: Configured

   As our host operating system is Ubuntu 18.04 Linux, we checked with Canonical and they advised us to disable LLDP in the firmware (using ethtool), and we have done that.      

  We are currently using 2.22.18 of the i40e driver, our NIC firmware is at 10.54.5, and the private flags are - 

Private flags for the ens1f0 port :

MFP                                   : off

total-port-shutdown     : off

LinkPolling                       : off

flow-director-atr             : on

veb-stats                          : off

hw-atr-eviction               : off

link-down-on-close       : off

legacy-rx                          : off

disable-source-pruning  : off

disable-fw-lldp              : on

rs-fec                                : off

base-r-fec                       : off

multiple-traffic-classes: off

vf-vlan-pruning         : off

vf-source-pruning       : on

vf-true-promisc-support : off

   
   As this has become a serious showstopper for us, can someone from Intel help us here?
Regards,
Sriram

[IntelSupport]

Hi sriram-sm,

Thank you for posting in Intel Communities.

Can you please confirm if you are using pre-built Intel X710 or is it bought separately.

Please be informed that it is important to identify this as if it is pre-built, then it would be best to contact the Original Equipment Manufacturer (OEM) as they might have specifics or altered driver version to integrate with their system.

For the confirmation, you are using i40e Version 2.22.18 and NIC Firmware of 10.54.5.

Let us know if you have any questions or any further details to be provided.

Regards,

Hayat

Intel Customer Support

[IntelSupport]

Hello sriram-sm,

Regarding your thread, we are following up to find out if you were able to complete the actions, we previously recommended.

Please reply to this thread to confirm, so we can continue helping with a resolution. Looking forward to receiving your reply!

Thank You

Regards,

Hayat

Intel® Customer Support

[IntelSupport]

Hello sriram-sm,

Regarding your thread, we are following up to find out if you were able to complete the actions, we previously recommended.

Please reply to this thread to confirm, so we can continue helping with a resolution. Looking forward to receiving your reply!

Else if no reply, we will proceed to close the thread. Once the thread is closed, it will no longer be monitored by Intel.

Kindly open a new thread if you need further assistance.

Thank You

Regards,

Hayat

Intel® Customer Support