Hello, I came across a strange issue in a customer...
# citrix-vad
d
Hello, I came across a strange issue in a customer's environment. The customer has two hypervisors: • VMware • XenServer I created two separate MCS master images with the exact same operating system and software configuration. In the VMware environment, everything works as expected. However, in the XenServer environment, every provisioned MCS worker VM boots with the same hostname as the master image. As a result, the machine is not renamed and, consequently, is not domain joined. The environment is using Machine Creation Services (MCS), and the catalog provisioning process completes successfully without any reported errors. The Identity Disk is attached and visible on the worker VMs. Environment • CVAD 2507 CU1 • XenServer 8.4 The screenshot below shows the only error I can find on the affected machines. Has anyone seen this behavior before or has any ideas where to continue troubleshooting?
🤷🏻 1
l
Weird, I've never seen this. You need the logs from the Prep machine.
d
New machine catalog with the same image. No problem. Very strange
l
Very strange, only thing I can think of is network. Maybe new machines on Xen are using an incorrect interface so the prep machine is behaving differently?
d
The machine catalog runs for months. After the last software update this happens. But rollback to the old image shows the same behavior then. Hilarious. But workaround machine catalog is now the productive
l
Certainly no pattern there. Keep digging my friend.
❤️ 1
s
We migrated Citrix workloads from VMWare to XenServer. We have 900 PVS target machines and 15 MCS machines. MCS machines on initial boot gets DHCP IP and it works, after few days MCS machines loses IP and gets an APIPA address. PVS targets on the same pool and same DHCP VLAN working fine. We are not sure where the issue. Whether MCS handles DHCP lease renewals differently?
d
But they get a fresh hostname, or the same like the master ?
My topic is exact this...the MCS Worker got the same hostname like the master....and my workaround with the new MC only works once....after the first update of the new MC i see there the same issue..
s
We don't issue with hostname but the DHCP IP, not sure if it was getting the same IP as Master. Our issue also started after an image update. Do you create new MC and move the machines from old catalog or create new MC and machines?
d
I delete the old MC and machines....after that i created a new MC with new machines. But it used the same AD objects names...but created new ones
s
Will try the same and let you know how it goes.
d
this could be your topic
s
The article I shared was from 2025, not valid anymore. Our issue is still open. Cases opened with XenServer, DaaS and Microsoft.
d
It seems that I found the reason. It was caused by an incorrect CrowdStrike policy.