Re: Intel X710  on Huawei server-malicious driver detection appear in the log

Hello,

We have the same issue. We have 30 Huawei servers with Intel X710 card and vsphere 6.5 u1 patch2. VM network connections on a Distributed Virtual Switch seem to freeze after this error appears in vmware logs:

"2018-03-30T07:35:50.216Z cpu5:66200)i40en: i40en_HandleMddEvent:6495: Malicious Driver Detection event 0x02 on TX queue 0 PF number 0x00 VF number 0x00

2018-03-30T07:35:50.216Z cpu5:66200)i40en: i40en_HandleMddEvent:6521: TX driver issue detected, PF reset issued

2018-03-30T07:36:20.213Z cpu48:70776)WARNING: NetPort: 1934: failed to disable port 0x300000d on DvsPortset-0: Busy

2018-03-30T07:36:20.213Z cpu48:70776)netschedHClk: NetSchedHClkPortQuiesce:4918: vmnic1: received a force quiesce for port 0x300000d

2018-03-30T07:36:20.213Z cpu48:70776)netschedHClk: NetSchedHClkHashQuiesceHierarchyIter:396: vmnic1: dropped 501 pkts from queue netsched.pools.vm.50331661 while quiescing port 0x300000d"

After this, multiple Dvs ports fail and packets are dropped (vmnic1: dropped 501 pkts from queue).

This has happened on 3 ESXi servers already, causing downtime on VMs.

The logs show the same sequence of events on all 3 servers:

1. Malicious Driver Detection event
2. failed to disable port ___ on DvsPortset-0: Busy
3. vmnicx: dropped ___ pkts from queue

Only way to recover the VM networking is to vMotion to a different host.

I am attaching the vmkernel log file for this server. Issues start at 2018-03-30T07:35:50.216Z

We are really DESPERATED... We are working in big migration project and we are totally frozen!!!

We have tried to installed the last drivers "1.5.8" and the issue happends again.

NEED URGENT HELP!!!