I built a VM using a downloaded OVA template for CUC 9.1(0) although my CUC disk version is 9.1(1a). This OVA template is the last one uploated at cisco.com, and the installation fails with a critical stop. Please refer to the attached file for the reason of the failure.
I built this VM on a UCS C200 M2 server running ESXi 5.0. On this server I have two CUCMs that work correctly.
Any tips on how to resolve this issue? I have read a couple discussion with a similar symptoms, but for a CUCM installation on ESXi, not for CUC.
For Unity Connection VM one CPU core must be idle/unused (it is actually being used by ESXi scheduler).
This could be the issue.
Highly appreciate your promt response. OK, for that reason I have installed the second one CPU which is Intel Xeon E5649 2.53GHz, but I thought that the OVA template automaticaly allocate one vCPU for the CUC APP and 1 vCPU which puts in the idle state. As you can see in the attached pictures the OVA allocated only one vCPU and in addition I have hyperthreading enabled. Do you think this can cause the issue? When I tried to allocated two vCPUs for the CUC I receive an issue for not supported CUC.
P.S. I have affinity for that machine.
Unfortunatelly not now, 'cause it want a ESXi restart. Based on cisco docs I can see that they recommends to be enabled If available:
If "Hyper-threading" BIOS option is available (and the CPU supports hyper-threading), UC recommends enabling.
Note that the resultant "Logical Cores" do not factor into UC sizing rules for co-residency. UC still requires mapping one physical core to one vcpu core (not to one "Logical Core").
I was just reading Hyperthreading which says :
is a technology included by Intel first in their Netburst line of parts. Hyper-Threaded processors present their individual processing cores to the system as if they are two processing cores. To use Intel's parlance, that means that each
physical core appears in the operating system as two logical cores. While the OS can distinguish between a system that has two logical cores (i.e. a single physical core with Hyper-Threading enabled) and two physical cores, applications cannot.
No, that extra CPU you need for CUC is not allocated by the OVA, you need to do the math of how many cores on on VMs + 1 <= total cores on server
Sent from Cisco Technical Support iPad App
CPU Cores: 12 CPUs x 2.533 GHz
Prorcessor Sockets: 2
Cores per Socket: 6
Logical Processors: 24
Hyper-threading enabled. Currently I am using around 10 vCPUs, so I have a lot of idle vCPUs.
Just refer this POST which might help.
Unfortunately failed again.
I am going to install CUC 8.6, but I have user licesnes for CUC 9.x. Do you know if they are backward compatible? I have ELM installed on another server.
Attached you can find three screenshots:
1. CUC Info
2. CUC State before failed
3. CUC Error
The installation process continues around 4-5 hrs and faild after this process (screen shot 2).
Maybe I should open a TAC case. I have a valid service contract.
You have missed the screenshots.
Please attach them.If u have a valid contract,u can go ahead with opening of the case.
I can still find VCPU as one which should be two.
Anyway please open the case and clear our doubts.
The OVA template allocated 1vCPU, and one should be in idle state which is not configured per VM. I'm using OVA template for 1K users. When I manually select 2 vCPUs I receive an issue with "CUC is not supported".
Going to the TAC...
I just wanted to add this reference to the great help so far from
my friend Aman (+5 Aman )
Your install seems to fail at the spot identified in this bug;
Cisco Unity Connection 8.x/9.x - Fresh Installation fails
You can open a TAC case directly from this page and the engineer with be able to see
the thread which may help your cause.
"A smile relieves a heart that grieves"
Hi Rob, thank you for your information! It's a funny one :
When we use "cucservice" as the Application username while installation.
Use the Application Username other than the following usernames
I have used cucadmin/cucroot.
Aman, thank you too, dude!
Thanks for sharing the Bug info and opening SR from the thread.
Never thought this could be a bug.
Did u open the case?
Hi Aman, I'll open it tomorrow morning, çause I need to check the service contract #. Meantime I'll try to do the installation again...based on the "workaround" posted in the bug.
I'll keep you in touch with the TAC.
Thank you! Owe u a beer :-)
Dear Aman, first of all sorry for the delay in my answer, but I was on a vacation. Yes, the problem was resolved. It appeared to be a performance issue. It can happen as install and upgrade require more IOPS than normal operation. We stoped 2 of the 3 guest virtual machines running on the host server, so we can have more IOPS free when we install CUC. And that was enough to complete the installation.
thanks for updating .Just one more query : Did Cisco TAC engr. run any Command or tool to find IOPS behaviour and reached to conclusion that other VM Machines need to be stopped.