0 Replies Latest reply on Oct 6, 2019 6:32 AM by eravmw

    Nested vSphere under QEMU/KVM unable to run VMs or Manage hosts

    eravmw Lurker

      Dear community,

       

      I am having troubles setting up a nested vSphere 6.7 environment under QEMU/KVM.

       

      I created my first vESXi VM where I mount an NFS share and deploy my vCenter to it. The vCenter deployment on the first vESXi host works fine and I am able to add the first ESXi host under the the nested vCenter (vCenter runs on that very host). I use vmxnet3 for the first vESXi host. If I switch to e1000 for the first vESXi host, the vCenter deployment fails.


      I created another vESXi VM but I am unable to add this vESXi host under the vCenter. The task gets stuck on 80% and packet capture shows a lot of re-transmissions. If I switch to e1000 for the second vESXi host, I am still not able to add it under the vCenter unless on the vCenter VM I disable LRO/LSO.

       

      This helps me manage the second vESXi host under the vCenter. I am able to create vDS, PortGroups. etc but unable to run any VM on the second vESXi host because of e1000 performance probably. I see the following log when I try to deploy a VM on the second vESX host. Same error in the first vESXi's log made me try to use vmxnet3 to make the vCenter deployment work but vmxnet3 won't let me manage it under the vCenter.

       

      [0x4180250e84e2]HelperQueueFunc@vmkernel#nover+0x30f stack: 0x43097b0e4768, 0x43097b0e4758, 0x43097b0

      2019-03-20T16:14:54.386Z cpu3:2097552)0x451a0889bfe0:[0x4180253081f2]CpuSched_StartWorld@vmkernel#nover+0x77 stack: 0x0, 0x0, 0x0, 0x0, 0x0

      2019-03-20T16:14:54.774Z cpu0:2100411)<4>e1000: vmnic0: e1000_phy_read_status: Error reading PHY register

      2019-03-20T16:14:54.774Z cpu0:2100411)<6>e1000: vmnic0: e1000_watchdog_task: NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None

      2019-03-20T16:15:04.332Z cpu3:2097552)IntrCookie: 2949: 0xa took 1000000us to sync

      2019-03-20T16:15:04.405Z cpu3:2097208)NetqueueBal: 5030: vmnic0: device Up notification, reset logical space needed

      2019-03-20T16:15:04.405Z cpu3:2097208)NetPort: 1580: disabled port 0x2000002

      2019-03-20T16:15:04.405Z cpu2:2097714)NetSched: 654: vmnic0-0-tx: worldID = 2097714 exits

      2019-03-20T16:15:04.405Z cpu3:2097208)Uplink: 11680: enabled port 0x2000002 with mac 0c:67:93:0d:6f:00

       

      Any help would be very much appreciated.