Hyper-V Guest - RDP Connections cause high latency.

Question

Hi all,
In a test environment we have a HP ProLiant DL360p Gen8 server with Windows Server 2012 R2 - Standard installed with the Hyper-V Role and no others.
On the server there is a Switch Independent network team configured with 2 of the 1GB NICs.
There is a virtual Swtich (External) configured on the Host linked to the Team with allow management selected.
The Guest VM has a single NIC (not a legacy setup) connected to the virtual Switch
The issue we are having is that once an RDP session is connected to a Guest OS, for example another 2012 R2 Server, it experiences high latency (which makes the RDP session appear to run in a very jerky way)
Using Hyper-V VMConnect, the server has no issues at all although the window is smaller.
We can verify this is being caused by an RDP session by observing ping times:
Reply from 10.101.3.152: bytes=32 time<1ms TTL=127
Reply from 10.101.3.152: bytes=32 time<1ms TTL=127
Reply from 10.101.3.152: bytes=32 time<1ms TTL=127
Reply from 10.101.3.152: bytes=32 time=1ms TTL=127
Reply from 10.101.3.152: bytes=32 time<1ms TTL=127
Reply from 10.101.3.152: bytes=32 time<1ms TTL=127
Reply from 10.101.3.152: bytes=32 time=75ms TTL=127 <RDP SESSION STARTS>
Reply from 10.101.3.152: bytes=32 time=52ms TTL=127
Reply from 10.101.3.152: bytes=32 time=63ms TTL=127
Reply from 10.101.3.152: bytes=32 time=121ms TTL=127
Reply from 10.101.3.152: bytes=32 time=135ms TTL=127
Reply from 10.101.3.152: bytes=32 time=141ms TTL=127
Reply from 10.101.3.152: bytes=32 time=146ms TTL=127
......... Snipped ........
Reply from 10.101.3.152: bytes=32 time=119ms TTL=127
Reply from 10.101.3.152: bytes=32 time=2ms TTL=127
Reply from 10.101.3.152: bytes=32 time=8ms TTL=127
Reply from 10.101.3.152: bytes=32 time=14ms TTL=127
Reply from 10.101.3.152: bytes=32 time=22ms TTL=127
Reply from 10.101.3.152: bytes=32 time=72ms TTL=127
Reply from 10.101.3.152: bytes=32 time=75ms TTL=127
Reply from 10.101.3.152: bytes=32 time<1ms TTL=127 <RDP SESSION CLOSES
Reply from 10.101.3.152: bytes=32 time<1ms TTL=127
Reply from 10.101.3.152: bytes=32 time<1ms TTL=127
Reply from 10.101.3.152: bytes=32 time<1ms TTL=127
Reply from 10.101.3.152: bytes=32 time<1ms TTL=127
Reply from 10.101.3.152: bytes=32 time<1ms TTL=127
This issue does not affect RDP sessions to the Host.
I have seen posts about disabling VMQ on the host/guest servers as there may be an issue with this model - this has had no effect at all, the issue persists after changing VMQ and restarting both the guest and the host OS.
I have seen another fix for a similar issue by running these commands: bcdedit.exe /set USEPLATFORMCLOCK on and this also has not resolved the issue.
I have also set the Guest VM to use 256Mbps minimum (512maximum - also tried with no maximum) bandwidth allocated, but again this does not resolve the issue.
There are no backup jobs or reporting jobs running on these servers (they are in test).
Is this caused by the specific hardware we have chosen or is there further diagnostics that can be performed? Any help would be appreciated.

Answer 1

Hi DalamarUK83,
"I have seen posts about disabling VMQ on the host/guest servers as there may be an issue with this model - this has had no effect at all, the issue persists after changing VMQ and restarting both the guest and the host OS."
Did you disable the feature in advanced settings of physical NICs ?
I would suggest to re-create virtual switch after disabling VMQ .
What is the make of that NIC ?
Best Regards
Elton Ji
We
are trying to better understand customer views on social support experience, so your participation in this
interview project would be greatly appreciated if you have time.
Thanks for helping make community forums a great place.