cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1349
Views
0
Helpful
1
Replies

CUCM VMware Virtual Appliance keeps writing all the time

gariup.guido
Level 1
Level 1

Hi,

   Does anybody know why the Cisco Unified Communications Manager VMware Virtual Appliance keeps writing to disk all the time. I'm seeing in the ESXi's performance charts that the CUCM Virtual Machine is always issuing write commands to its disk, even when it is in idle state. When you power on the VM, after the boot process is finished you see the VMs keeps writing. It issues about 40 write IOPS, constantly.

Thanks in advanced,

Guido.

1 Accepted Solution

Accepted Solutions

dakeller
Cisco Employee
Cisco Employee

Guido,

I can think of a couple of configurations that might trigger this behavior, but you are missing some key information.  Are you observing the same behavior on all nodes, or just a specific node?  Also, is the behavior on call just a single processing node or publisher, the TFTP sever or all nodes?   But here is a more important question.  Why do you believe that this IOPS value is not normal?

Please review UC Virtualization Storage System Design Requirements - DocWiki for information on IOPS. 

There are a couple of configurations on a UCM that could trigger this behavior.  The first is if you have detailed trace files turned on and you have a busy system, that will cause higher IOSP.  Another problem could be with the OVA.  If the vdisks are not aligned, then there is extra time required to handle the misalignment.  Another possibility is running a continual loading of CDR's for CAR reporting. 

Although the IOPS rate depends on quite a few configuration items, Baseline for a multi node system with 12K BHCA is estimated at 50 IOPS.  So depending on you config, 40 is not too far out of line.

Thanks,

Dan Keller

Technical Marketing Engineer

View solution in original post

1 Reply 1

dakeller
Cisco Employee
Cisco Employee

Guido,

I can think of a couple of configurations that might trigger this behavior, but you are missing some key information.  Are you observing the same behavior on all nodes, or just a specific node?  Also, is the behavior on call just a single processing node or publisher, the TFTP sever or all nodes?   But here is a more important question.  Why do you believe that this IOPS value is not normal?

Please review UC Virtualization Storage System Design Requirements - DocWiki for information on IOPS. 

There are a couple of configurations on a UCM that could trigger this behavior.  The first is if you have detailed trace files turned on and you have a busy system, that will cause higher IOSP.  Another problem could be with the OVA.  If the vdisks are not aligned, then there is extra time required to handle the misalignment.  Another possibility is running a continual loading of CDR's for CAR reporting. 

Although the IOPS rate depends on quite a few configuration items, Baseline for a multi node system with 12K BHCA is estimated at 50 IOPS.  So depending on you config, 40 is not too far out of line.

Thanks,

Dan Keller

Technical Marketing Engineer

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: