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

Having an Issue with Retries in new PVS Setup 7.15 LTSR Citrix Hypervisor 8.1


Mark Federico

Question

So I am having an issue with Retries on some new PVS server I am trying to setup 

 

I deployed them on Citrix Hypervisor 8.1 

 

I have disabled IPv6, Disabled Task Offload V2 via the XenNet Driver, and also set the registry value to disable TaskOfflload done this on both the PVS Server and the Guest itself where I am trying to image into PVS.

 

The Guest VMs and PVS servers are on the same Citrix Hypervisor Cluster and I even went as far as booting the VM where I am taking the image from on the same Hypervisor that the PVS its imaging to is running off of and I still get a ton of them.

 

Since I have done all the normal tweaks kind of stuck looking for a solution 

Link to comment

5 answers to this question

Recommended Posts

  • 0
19 hours ago, Koenraad Willems said:

HI mark,

 

Where are the images sitting in terms of storage location? Could it be the connection there? Are you in the possibility to use local storage, to rule that out?

Have you checked out if UDP packets are dropped on the switches and/or set QoS for the PVS traffic?

 

Best,

 

Koenraad

Koenraad,

 

Already using local storage local to the new PVS servers, will check via packet capture today if UDP packets are dropped but dont think that is the issue and we are not using QoS on the switches that this traffic is going through. 

 

Here is exactly what we have tried so far:

  • Making sure DisableTaskOffload Turned on
  • For Testing Making my vDisk store on local SSD storage right from Hypervisor
  • Uninstalling all AV and no firewall on Guest or New PVS Server for testing
  • Booting Guest on same Host that the PVS Server lives on to cut out the network
  • Made Sure XenTool were completely up to date with no Ghost Devices present
  • Made sure there were no MTU Fragmentation issues between guest and PVS
  • Tested Citrix Hypervisor Networking in both openvswich and linux bridge mode
  • Disabled TCP Offload on the Citrix Hypervisor PIF level
  • Taken off any network bonding and just running to one single Hypervisor Network and yes the PVS and Target on are the same VLAN.
  • I have opened a case with Citrix and we went though my config with a fine tooth comb and from what he could see everything looked good.
  • Tested latency and throughput looked good nothing off in WireShark traces taken from both the Guest and PVS Side
Link to comment
  • 0

Hi Mark,

 

Could you provide more info regarding retires count? Most number of retries are noticed during boot or after boot? When a target is up for 10 min after boot, what is the count of retries?

 

How's the performance of the target device? Is it sluggish and slow or normal? If the performance is normal then the retires would be a result of no response from PVS server in a timely manner for a heartbeat sent by the target instead of an actual network issue while streaming the Vdisk

 

Sandeep.

Link to comment
  • 0
1 minute ago, Sandeep Sandeep said:

Hi Mark,

 

Could you provide more info regarding retires count? Most number of retries are noticed during boot or after boot? When a target is up for 10 min after boot, what is the count of retries?

 

How's the performance of the target device? Is it sluggish and slow or normal? If the performance is normal then the retires would be a result of no response from PVS server in a timely manner for a heartbeat sent by the target instead of an actual network issue while streaming the Vdisk

 

Sandeep.

 

Sandeep,

 

I haven't got a full image of my new target yet since these retries have occurred during the image creation process most of the time the retries start after the machine has already booted however I have seen times where the machine gets stuck at the PVS IP information during the boot process but that is rare.  

 

Another piece of valuable is I tried an interesting test I took my new VM, on my new hardware, and did the image process back to one of my old PVS servers on my old hardware during the image creation there isn't even 1 retry happening. so at least this proves its not a Hypervisor or Network issue that its something on PVS itself. so that at least narrows down where I need to look.

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...