Hello,
I'm currently using ESXi 6.5.0 U1 on 5th gen Intel NUC 5i5RYH. My problem is that in vmkernel.log, I can periodically see following error messages:
2017-07-31T13:02:58.620Z cpu3:65919)INFO (ne1000): false RX hang detected on vmnic0
2017-07-31T13:28:57.556Z cpu3:65919)INFO (ne1000): false RX hang detected on vmnic0
2017-07-31T13:56:36.992Z cpu3:65919)INFO (ne1000): false RX hang detected on vmnic0
2017-07-31T18:09:02.992Z cpu2:65919)INFO (ne1000): false RX hang detected on vmnic0
2017-07-31T18:54:39.258Z cpu2:65919)INFO (ne1000): false RX hang detected on vmnic0
2017-07-31T19:23:58.097Z cpu1:65919)INFO (ne1000): false RX hang detected on vmnic0
2017-07-31T19:31:25.509Z cpu0:65919)INFO (ne1000): false RX hang detected on vmnic0
I assume that it has something to do with the native NIC driver, ne1000. Since I haven't noticed any performance degradation while those errors appears, I don't know if they are safe to ignore. I would like to have some more information about what does those errors means (unfortunately, I haven't found anything relevant)? My NIC, I218-V is on HCL list for version 6.5.0 U1. I know that e1000e driver supports my NIC as well, although shouldn't ne1000 perform better?
I have also found relevant changelog information for my network adapter in here [VMware ESXi 6.5 Update 1 Release Notes ]:
Supported Hardware Issues
- Intel I218 NIC resets frequently in a heavy traffic scenario When TSO capability is enabled in NE1000 driver, I218 NIC reset frequently in heavy traffic scenario, because of I218 h/w issue. The NE1000 TSO capability for I218 NIC should be disabled. This issue is resolved in this release.
Although, when I look in ESXi Host Web Client -> Manage -> Advanced Settings, TSO related settings are following (enabled):
Net.UseHwTSO
When non-zero, use pNIC HW TSO offload if available
1
1
No
Net.UseHwTSO6
When non-zero, use pNIC HW IPv6 TSO offload if available
1
1
No
I don't really know if they are somehow related, so please enlight me