Jump to content
Welcome to our new Citrix community!
  • 0

Host connections in PVS with XenServer



I´m used to VMware and host connection in XenDesktop.


What I see in this Citrix environment,  with XenServer and PVS, is that under host in PVS console there are a lot of
hots connections that are similar to each other but configured differentently.

As in this example we have two xen104 connections, one with (fake ip-addres in this example)
-(please see screenshot) and one host connection with only xen104.
If you look further you also can see that a bunch of connections are confiured with host name (xenxxx) and
other with http://FQDN.
Some connections are doubble, but with different configuration with host name or FQDN.
I´ve tried all the connections and they are working.
- see screenshot


My question is what is best practice of how to configure these host connections. 
Can it be som problems with these doubbled host connections?

Because we have som problems in our Citrix environment, but we dont know for yet

what causing these problems


Screenshot is with fake ip-address



Local storage




-Really appreciate your answer

Link to comment

5 answers to this question

Recommended Posts

  • 1

Host connections observed in the PVS console are usually only created and used during PVS target device creation.

They would also be used when some operations are completed from PVS console, like updating the bdm drive.
However PVS host connections would not normally be used during production (all power management etc would be handled by the DDC via host connections configured in DDC studio).


What specific problems do you have in the citrix environment?

  • Like 1
Link to comment
  • 1

Host connections in PVS console would not affect target device boot times.

Boot times, especially bios based VMs, is especially impacted by latency.

  • Vdisk storge latency, or network latency
  • Network latency can be observed by high ping times (less than 5ms to and form PVS server to targets would be expected), but can also be introduced by network saturation.
  • Also potential for latency introduced if CPUs are busy
  • Boot storms, starting many PVS targets at once (25+ or mor targets per PVS server) can increase all 3 areas of latency (depending on underlying hardware)

Also of relevance is reliablity of the network, packet drops between PVS server and target can significatly increase boot times.
On PVS targets which are fully booted, you would observe a large amount of retries (100s) in PVS tray tools. But note that for some environments, a certain amount of retries are expected, or within their acceptable toelrances for the environment.

  • Like 1
Link to comment
  • 0

We have packet loss 12-30 % with ping - t from pvs to XenServer host

Request timeout 3 of 10 pings.

What we can see are that guest VM who is booting, it hangs and no or very little traffic i going on from VM to PVS. VM (vDisk) hangs,

from time to time, with the same symptom.

Our network guy cant fin anything particular in our Network. Even with WireShark.
And we have adjusted our MTU from 1506 down to 1500





Link to comment
  • 0

The packet loss occurs with ping would indicate the problem is independant of PVS components.
Even a heavily utilised network would not be expected to have that volume of packet loss.

But if needing to investgate PVS network traffic and retries, opening a case with citrix technical support would be the way forward.

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Create New...