VMware Cloud Community
CarlosGMat
Contributor
Contributor
‎12-03-2018 04:27 AM

vmnic issue with ESXi 6.0u3 (Build 9239799) & Intel Corporation 82599EB 10-Gigabit adapter

Hi,

I´m having issues with Intel Corporation 82599EB 10-Gigabit adapter after installing the HPE-ESXi-6.0.0-Update3-iso-Gen9plus-600.10.3.0.21 image, which changes the driver from vmklinux ixgbe to the native ixgben driver.

In chronological order:

  • After deploying ESXi image, driver ixbgen 1.6.5 , the problem was: after vMotion, the source host lost communication over all networks (including Management), and a mgmt network restart in DCUI was requiered. Same network issues shutting down/restarting VM´s. It seems that both 10 GBe adapters are disabled, and not enabled. (Very similar to https://community.spiceworks.com/topic/2125714-cannot-live-vmotion-vm-on-esxi-6-0-0-7967664-without-...)
  • Upgrade ixgben driver to 1.7.10: communications are not lost anymore, but adapters are disabled/enabled quickly, so vCenter alarm "Network Redundancy Lost"/"Network Connectivity Lost" is issued in source vMotion host.
  • Install patch in https://kb.vmware.com/s/article/52032 ​. No adapters disabled/enabled during vMotion, but immediately after VM shutdown.

         

2018-11-30T13:05:31.069Z cpu53:4627406)VSCSI: 6784: handle 8306(vscsi0:0):Destroying Device for world 4627407 (pendCom 0)

2018-11-30T13:05:31.069Z cpu53:4627406)VSCSI: 6784: handle 8307(vscsi0:1):Destroying Device for world 4627407 (pendCom 0)

2018-11-30T13:05:31.069Z cpu53:4627406)VSCSI: 6784: handle 8308(vscsi0:2):Destroying Device for world 4627407 (pendCom 0)

2018-11-30T13:05:31.069Z cpu53:4627406)VSCSI: 6784: handle 8309(vscsi0:3):Destroying Device for world 4627407 (pendCom 0)

2018-11-30T13:05:31.084Z cpu172:33194)Net: 3654: disconnected client from port 0x4000009

2018-11-30T13:05:31.097Z cpu137:33196)WARNING: SP: 1524: Smashing barrier dealloc-Barrier.

2018-11-30T13:05:31.522Z cpu46:4651893)CpuSched: 596: user latency of 4651893 helper24-1 0 changed by 4651893 helper24-1 -6

2018-11-30T13:05:31.522Z cpu110:33190)NetPort: 1782: disabled port 0x4000004

2018-11-30T13:05:31.523Z cpu110:33190)Uplink: 7468: enabled port 0x4000004 with mac 48:df:37:2d:8a:b8

2018-11-30T13:05:31.620Z cpu110:33190)NetPort: 1782: disabled port 0x4000002

2018-11-30T13:05:31.621Z cpu110:33190)Uplink: 7468: enabled port 0x4000002 with mac 48:df:37:2d:a9:29

No network communication lost, but alarm is issued. Any ideas?

Regards,

Carlos

Reply
0 Kudos
4 Replies
dbalcaraz
Expert
Expert
‎12-03-2018 07:09 AM

I didn't see you mentioned anything about compatibility, did you check the HCL from VMware?

-------------------------------------------------------- "I greet each challenge with expectation"
Reply
0 Kudos
CarlosGMat
Contributor
Contributor
‎12-04-2018 06:31 AM

Hi dbalcaraz,

Yes, I checked HCL for this Intel 10 GBe and the driver ixgben (version 1.7.10), and are supported.

Thanks.

Carlos

Reply
0 Kudos
Dave_the_Wave
Hot Shot
Hot Shot
‎12-05-2018 03:48 PM

The whole point of HPE builds is to stop grey hairs.

Do yourself a favor and treat yourself a nice Hp nic from your popular friendly online auction website. You'll even find many with free shipping.

Yes, That Hp NIC might even be of an identical Intel chip, but it'll contain Hp "branded guts".

In a few days, all your problems will be solved.

I'm really happy that Hpe ESXi works error free with all my Hp "stuff".

Sure, you may be able to get that generic nic working eventually, how long will that take?

Keep us posted.

Reply
0 Kudos
vzav
Contributor
Contributor
‎05-03-2019 07:55 PM

Hi Carlos, I have a similar problem, the host reconnects after shut/no shut off the switch interface. I have 3 Host HP BL460c G9 connected to a nexus fex N9K.

I have open tickets in HP, VMware, Cisco and I still do not have a clue to the problem.

How did it go to you? Could you solve it?

Reply
0 Kudos