Citrix Provisioning Server 5.6 Excessive Retries

We have an odd situaion.  VMware 5.0 Hyper.  PVS6.1 of Server 2008R2.


Provisioned servers show excessive retries ~11/minute.


Pings between PVS and provisioned servers ~30% loss.


Ping from PVS to any other server (virtual or physical) including other PVS servers or network hard ware 0% packet loss


Ping from provisined servers to anything else, other provisioned servers, other PVS servers, all 0% packet loss.


Excessive retries makes provisioned servers very hesitant.  They run full speed for ~20 seconds then freeze for 5-10 seconds over and over.


Does not matter if all aspects of provisioning are on the same physical host or scattered accross different hosts.


Other items in the enterprise are fine.


This was funtioning fine until about two weeks ago.


Obvious question is, "what changed?"  I wish I could think of something.
Who is Participating?
whoamConnect With a Mentor Author Commented:
Finally found an answer, or at least an effective work around.

Power down the affected device, the provisionED server.

Note the MAC address (if you're using reservations) and remove the virtual nic (ESXi 5.0)

Add back in a new virtual nic, using the same MAC address.

Push the affected device back into production, network works fine

Repeat as needed, usually last me ~2-3 weeks per device.
Daniel BorgerSenior Citrix Engineer- CCEECommented:
Please review the following best practices. pay special attention to the Network settings and the Reg entries.

CTX124185 - Provisioning Services Anti Virus Best Practices - Citrix Knowledge Center


CTX117374 - Best Practices for Configuring Provisioning Server on a Network - Citrix Knowledge Center


CTX119469 - Understanding Write-Cache in Provisioning Server - Citrix Knowledge Center

Also.. Is the ia virtual PVS server? if so there are other recommendations for that as well.
whoamAuthor Commented:
Thank you I'll take a look at these articles. Yes both PCS servers are virtualized.
7 new features that'll make your work life better

It’s our mission to create a product that solves the huge challenges you face at work every day. In case you missed it, here are 7 delightful things we've added recently to monday to make it even more awesome.

Daniel BorgerSenior Citrix Engineer- CCEECommented:
I just closed an issue with excessive retires and boot issues. it was resolved by hard coding the port speed on the switch the PVS servers were using.
whoamAuthor Commented:

Yep, we had already adjusted all the offloads to 'off' in the nic properties.  

The whole PVS infrastruture is virtual, so I don't think it is a prot negot issue because the issue happens if all invovled are on one physical host.  We cache on the device HD.  

A very interesting finding: if we boot only one server from a vDisk, all is fine.  

We can run a server in standard and another in maintencance mode off the save vdisk fine.

However, if we run two or more servers off one vDisk in standard mode, that's when retries go to 10+/min and pings between PVS and PVS'd servers go 30% loss.
Dirk KotteSECommented:
check the MAC and IP from the running system(s).
possible one is not unique.
nprignanoTechnical ArchitectCommented:
Sounds like it could be a driver issue too, are you using vmxnet3?  In my experience, issues like this are almost always something impacting the PVS server or targets (or both) network communication.  Figuring out where is the fun part.  I'd suggest loading Wireshark on your PVS server and capturing traffic while this issue is occurring, then filter on your PVS server and one of the affected targets and look for transmission errors.
whoamAuthor Commented:
No other solution was effective.
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.