05-20-2013 11:00 PM - edited 03-16-2019 05:25 PM
With Harmit Singh
Welcome to the Cisco Support Community Ask the Expert conversation. This is an opportunity to learn and ask questions about Cisco Unified Communications Manager, ranging from install / upgrade / migration paths to general best practices to configuration and troubleshooting of the product
This is a continuation of the live Facebook Forum.
Harmit Singh is a technical leader with the High Touch Technical Services (HTTS) Unified Communications team based in Bangalore. He has broad experience in Cisco Unified Communications infrastructure solutions. He has 10 years of experience working with large enterprise and service provider networks. He also holds a CCIE certification (20012) in Voice as well as Red Hat and VMware certifications.
Remember to use the rating system to let Harmit know if you have received an adequate response.
Harmit might not be able to answer each question because of the volume expected during this event. Remember that you can continue the conversation on the IP Telephony community IP Telephony shortly after the event. This event lasts through Friday, May 31, 2013. Visit this forum often to view responses to your questions and the questions of other Cisco Support Community members.
05-24-2013 06:30 AM
Hi Aman,
How are you? Long time. For your question, I'll share the same links I shared for the previous question here:
Translator X:
Triple Combo:
http://www.employees.org/~tiryaki/tc/
You will see that both tools are extremely helpful to filter only the call in question and analyze it irrespective of the voip protocol being used. The Triple Combo can even be used to help analyze debugs taken from voice gateways.
The above tools come in handy, whether you are new to the SDI/SDL trace world or a pro. If you ask me, I use a combination of tools such as --
1> Agent Ransack (for grepping through MBs or GBs of traces)
2> Translator X (to help me filter the problematice call)
3> Raw trace files when I want to track the various call legs, call IDs, CUCM interfaces which are involved in that specific call / feature, etc.
To be able to read the raw trace files, it's important you familiarize yourself with the CUCM software architecture, else it could be like finding a needle in a haystack :-)
Hope this helps!
--
Regards,
Harmit.
05-25-2013 07:25 PM
Hi Harmit,
I am fine.
Thanks a lot for all the questions being asked and asnwered by you in the forum.
I have downloaded the tools and would use them in future.
In case of any queries, I would let u know.
regds,
aman
05-25-2013 11:59 PM
Hi Aman,
Wonderful! Always a pleasure to help!
--
Regards,
Harmit.
05-24-2013 05:10 AM
Hi Tenaro,
My pleasure! For your query, there are several options, depending on what exactly you are looking for and to what extent you want to troubleshoot an issue.
1> Assuming that there is a CUCM involved in the troubleshooting scenario, detailed CUCM SDI traces will surely come in handy. One of the best tools out there is Translator X. You can download it from the following location:
The features are nicely described there. Another great tool that you can use is the Triple Combo. You can download it from the following location:
http://www.employees.org/~tiryaki/tc/
2> Then there is always a packet capture, which also does the trick as and when needed. To be able to capture packets on a CUCM server, you can use the following procedure:
https://supportforums.cisco.com/docs/DOC-11599
To be able to capture packets on an IP Phone, you can use the following procedure:
https://supportforums.cisco.com/docs/DOC-11735
Hope the above helps. Do let me know if you have any follow questions.
--
Regards,
Harmit.
05-28-2013 05:06 AM
OVA file for CUCM gives you choice to select proper VM requirements based on number of users. What bothers me is that requirements for 1000 users are MORE demanding on virtual machine than requirements for 2500 users.
To be precise, if you use cucm_9.1_vmv8_v1.6(2).ova or cucm_8.6_vmv7_v1.5.ova you'll notice that for 1000 user node VM details are:
CPU: 2 vCPU with 800 MHz reservation
Memory: 4 GB with 4 GB reservation
Disk: 1 - 80 GB disk
while for VM node that will handle 2500 users details are following:
CPU: 1 vCPU with 800 MHz reservation
Memory: 4 GB with 4 GB reservation
Disk: 1 - 80 GB disk
I thought it has something to do with CUCM and UCxn coresidency inside SAME virtual machine but as far as I know that is not possible anymore.
I'll appreciate if you can explain the reason.
Cheers,
Tenaro
05-28-2013 09:41 AM
Hi Tenaro,
That's a good question. If you refer to the Readme file for 8.6 OVA file, you will notice the following:
http://www.cisco.com/web/software/283088407/51125/cucm_8.6_vmv7_v1.5.ova.README.txt
CUCM 1000 user node - C200 (incl BE6K):
Cisco Unified Communications Manager (CUCM) configuration that supports up to
1000 users per node. For use on C-200 (including BE6000) only.
Details:
Red Hat Enterprise Linux 5 (32-bit)
CPU: 2 vCPU with 800 MHz reservation
Memory: 4 GB with 4 GB reservation
Disk: 1 - 80 GB disk with pre-aligned disk partitions
CUCM 2500 user node:
Cisco Unified Communications Manager (CUCM) configuration that
supports up to 2500 users per node.
Details:
Red Hat Enterprise Linux 5 (32-bit)
CPU: 1 vCPU with 800 MHz reservation
Memory: 4 GB with 4 GB reservation
Disk: 1 - 80 GB disk with pre-aligned disk partitions
The CUCM 1000 user node is specifically for UCS C200 series / BE6000 setups. Here are some reference links for BE6000:
http://www.cisco.com/en/US/products/ps11369/index.html
http://docwiki.cisco.com/wiki/Cisco_Unified_Communications_Manager_Business_Edition_6000
http://docwiki.cisco.com/wiki/Cisco_Business_Edition_6000_OVA/OVF_Templates
http://docwiki.cisco.com/wiki/Supported_System_Capacities
Supported Hardware (latest) summarizes newest hardware supported by Cisco Collaboration applications.
• UC on UCS Tested Reference Configurations (TRCs) summarizes support for newest TRCs. Green if the application supports the same set of newest TRCs as Cisco Unified Communications Manager. Yellow if it supports some of the newest TRCs - see details page for which. Red if supports none of the newest TRCs.
• UC on UCS Specs-based or 3rd-party Specs-based summarizes support for latest Intel Xeon CPUs and Specs-based policy "rules". Green if the application has the same support policy and Intel Xeon CPU options as Cisco Unified Communications Manager. Yellow if partial - see details page. Red if the application either does not support Specs-based at all, or has a different support policy from Unifed CM and the core UC apps.
• UC apps do not support CPU oversubscription. One VM vCPU must map to one physical core (not one logical core).
• "One vCPU to one physical core" applies for all supported CPU architectures with equivalent performance for UC. E.g. one physical core on E5-2600 at 2.53+ GHz and one physical core on E7-2800 at 2.4+ GHz are equivalent for purposes of UC VM OVAs, sizing and co-residency planning.
• Each VM OVA is only for use on certain CPU architectures.
For Specifications-based deployments (non TRC deployments), the CPU core count AND the CPU core speed correspond to physical CPU cores and physical CPU core speeds. Both are required to be met to support a full load VM OVA.
Please also refer to the UC Virtualization Supported Hardware link for more on this:
http://docwiki.cisco.com/wiki/UC_Virtualization_Supported_Hardware#Processors_.2F_CPUs
Here is the corresponding UC Virtualization Supported Applications link:
http://docwiki.cisco.com/wiki/Unified_Communications_Virtualization_Supported_Applications
The CUCM 1K user OVA is only for use on E5-2609/2.4 GHz (only found in UCS C220 M3S TRC#2) and E5506/2.13GHz (only found in UCS C220 M2 TRC#1).
All other OVAs (such as CUCM 2.5K user, 7.5K user, or 10K user) are only for use on CPUs with minimum core speed of 2.53 GHz, such as E5-2643/3.3 GHz or E5640 / 2.66 GHz.
Here is the OVA Template Details for UCM 8.6:
http://docwiki.cisco.com/wiki/OVA_Template_Details_for_Unified_CM_Release_8.6
Hope this helps. Do let me know if you have any follow up questions.
--
Regards,
Harmit.
05-29-2013 02:16 AM
Hi Harmit,
I had a question regrarding BE 6000 UCS Hardware.
Can we use the UCS hardware UCSC-C220-M3SBE for BE6000 installations only or for other installations as well?
Suppose if customer wants to increase the number of users from 1000 to 1200, can we uninstall BE 6000 and proceed with CUCM 8.6 on the same above quoted hardware.
Please advice.
regds,
aman
05-29-2013 05:18 AM
Hi Aman,
To answer your question, here are some details on the UCS C220 M3 server for TRC#3:
++ 16 cores @ 2.4 GHz, 32GB RAM, w/ 8 disks.
++ Can be used with BE6K deployment and/or UC on UCS deployment
++ In the BE6K program, it supports up to 8 VMs.
++ For UC on UCS, the VM policy follows the existing rules (1 vCPU to 1 physical core and no RAM oversubscription)
Additionally, C220 M3S TRC#2 - is ONLY supported for use with Cisco Business Edition 6000, and it is quoted as UCSC-C220-M3SBE in CMBE6K-UCL-K9 or CMBE6K-UWL-K9.
http://docwiki.cisco.com/wiki/Unified_Computing_System_C220_Hardware
It's important to cross check all the documentation available for the various models of UCS available as well as which TRC you are looking to use. Here are a few docs:
http://docwiki.cisco.com/wiki/Unified_Communications_Virtualization_Sizing_Guidelines
It can be a confusing topic but hope this helps :-)
--
Regards,
Harmit.
05-31-2013 12:30 AM
Hi Harmit,
Thanks for the reply.
regds,
aman
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide