SUB on VM, disk size not matching

[goranpilat]

Hi,

I am having more or less the following issues:

I have CUCM cluster on 5x 7835 MCSs (Pub + 4 SUBs)

I am installing additional SUB into cluster, and this new one will be on VMWare.

Using cucm_9.1_vmv8_v1.7.ova, node for 2500 users which gives me 60GB disk

My MCSs have 2x146 HDD each, and I am getting the error in middle of installation (something along the lines that SUB and PUB HDDs are not the same size). I changed the disks manually, installation went on, but I got the following error on CLI and on GUI homepage : ERROR-UNSUPPORTED: Partitions unaligned

Just like on the link http://docwiki.cisco.com/wiki/Troubleshooting_and_Performance_Monitoring_Virtualized_Environments#VMware_and_VM_Configuration

So I am not sure what how to proceed,

Anyone had the similar issues?

Another strange issue with the server is the following: I installed it (ok, with above error), put it into cluster, I can open its service from PUB, started them and so on. However I noticed that at some point my phones restarted by themselves. I found about the bug CSCub12922 which causes phones to reset because of ITL certificates. But my phones restarted at least three times (later two times when I reset the new SUB, once it went down and once it came up). This new SUB is in no CUCM group or anything. So I am afraid it will cause the phones to reset every time it (new sub) is reset. I am not sure if it has anything to do with the above unsupported/partitions unaligned issue

Any suggestion will be appreciated

tx a lot

Goran Pilat

[Manish Gogna]

Hi Goran,

The phones will reset when you add a new subscriber to the cucm cluster. This new feature was introduced in 8.x and higher versions. Its covered under bug id CSCub12922. You may check the following post

https://supportforums.cisco.com/document/12563121/clarity-bug-cscub12922-adding-new-subscriber-causes-device-reset

Manish

- Do rate helpful posts -

[goranpilat]

Hi Manish,

Thanx for the reply. I saw the bug/feature, but as I said this happened again twice during the restart of the server (once when it went down and once when it got back up), so at least three times that I saw. So I am afraid it will be doing so each time server for any reason becomes unavailable/is restarted.

But as Jaime suggested I think I will reinstall the server for the HD size reasons and see if the behavior will improve somehow

Anyway, thanks again

regards,

Goran

[Jaime Valencia]

You're not supposed to change the HDD or the OVA settings at all, yes, the size between MCS and VMs is different, and it's completely expected to be!

You need to deploy the OVA with the specs it has, and DO NOT modify it, then use it for the CUCM 9.1 (which I assume you're using as per the OVA)

[goranpilat]

Hi Jaime,

tnx for the answer, as always... I cannot say 100% if the disk discrepancy was only a warning or the installation wouldn't proceed, because the installation was done by the customer. I will try to re-install with different disk sizes (this is what you are suggesting, right?). Will let you know next week.

Regards

Goran

[goranpilat]

Hi,

Just to follow up, It went without problems with the same ova file (cucm_9.1_vmv8_v1.7.ova), but used 7500 user node. Although HD sizes still don't match, it didn't complain this time

CUCM 7500 user node:
	Cisco Unified Communications Manager (CUCM) configuration that
	supports up to 7500 users per node.
	Details:
	Red Hat Enterprise Linux 5 (32-bit)
	CPU: 2 vCPU with 3600 MHz reservation
	Memory: 6 GB with 6 GB reservation
	Disk: 1 - 110 GB disk with pre-aligned disk partitions