Guests randomly ping loss on same vswitch & host

[ogdrujahad]

We have network problem on vm guest server, try to check ping on a guest and found randomly ping loss with no network load and no packet drop(esxtop), even when ping on two guests on same host and same vswitch, still have this problem. Anyone ever face this kind of problem? I already open support with vmware but they didn't find any problem on configuration.

Evironment

Regards,

Nawa

[Stanley_]

are they two VM's also on same portgroup? if not then it still must go out to the router and then back on different vlan. and we have similar issue. and it was caused by bad ASIC chip on cisco switch.

[ogdrujahad]

Yes, it's on same portgroup.

[Stanley_]

so you are pinging from one VM to another VM on same host in same port group and you have packet loss?

if they are on same host in same portgroup, then i guess it must be something inside VM.

are the vNIC's supported for the OS?

for example "e1000" is not supported in Windows 2016 see VMware Compatibility Guide - Guest/Host Search

do you have vmtools installed?

are there any errors in VM OS log?

duplicate MAC? correct MTU?

what kind of OS is installed on both VMs?

are you able to ping somewhere without packet loss from that VM?

[ogdrujahad · Accepted solution]

Hello Stanley,

Thank you for your recommendations after consult with vmware support this is not about vmware configurations. By the way I found repeatedly messages on vmkernel.log "NMP device "naa.blablablablaxxxx" state in doubt; requested fast path state update", so I try to deep investigate on I/O path and found the root cause was a SAN switch. After replacement of SAN switch everything look good no ping loss anymore.

Thanks,

Nawa