Voice Quality: Deploying Tools for a Stable Infrastructure © 2007 Cisco Systems, Inc. All rights reserved. This document is Cisco Public Information.
White Paper Contents Introduction ..................................................................................................................................... 3 Voice Impairment Parameters ........................................................................................................ 3 Requirements of a Voice Quality Measurement Product .............................................................. 4 Components of Cisco Unified Communications ..............................................
White Paper Introduction This white paper provides insight into voice quality and the different methods to measure voice quality. Voice over IP (VoIP) has passed its infancy stage and is a mature technology that has been widely adopted by customers hoping to take advantage of the cost savings offered by VoIP in addition to a range of advanced features that improve efficiency. Voice quality is the qualitative and quantitative measure of sound and conversation quality on an IP phone call.
White Paper The network infrastructure must meet the following requirements: ● Packet loss must not be more than 1 percent ● Average one-way jitter must not be more than 30 ms ● One-way delay must be under 150 ms To help ensure good voice quality, it is imperative to keep jitter and packet loss under control by paying close attention to voice impairment factors.
White Paper It is imperative to have a rock-solid network infrastructure. The network infrastructure offers the platform for advanced technologies such as voice, video, wireless, and so on.
White Paper The sensor has two Fast Ethernet interfaces: ● Management port ● SPAN port The sensor uses IEEE 802.3af standard Power over Ethernet (PoE) from the switch to which it connects. When the Cisco 1040 Sensor boots up, it uses Dynamic Host Configuration Protocol (DHCP) option 150 to retrieve its configuration and image files on a TFTP server.
White Paper Service Monitor satisfies most quality-monitoring requirements for enterprise IP telephony. Deployment strategies include: ● Strategic monitoring: The Cisco 1040 Sensor is installed to continuously monitor IP phones at some or all locations in the managed environment.
White Paper covering segments 1 and 2. The RTP stream coming from Phone B covering segment 4, 3, and 2 is of importance; the switch on the left must be configured to span the incoming RTP stream, to span the destination port seen by the left sensor. With optimal SPAN port configuration, each sensor can monitor 100 calls. The number of sensors required depends on the busy hour call completion (BHCC) handled by the switch.
White Paper Operations Manager provides a powerful product for monitoring and troubleshooting voice quality problems. Placement of Cisco 1040 Sensors The Cisco 1040 Sensor is FCC Class B certified and can be deployed in a wiring closet or on a desk. The Cisco 1040 uses the SPAN port on the switch to monitor the RTP stream. As shown in the following diagram, the sensors are deployed as close to the IP phone as possible so that the voice quality measurement will be close to what the user experiences.
White Paper A typical enterprise customer has various organizations, such as engineering, human resources, marketing, sales, and support. The traffic patterns are different in these organizations. You can expect more calls for support, sales, and marketing groups, and the switches that house these users will have a higher busy hour call attempt (BHCA) value.
White Paper In a scenario in which the switches are not stacked but have gigabit uplink to an aggregate switch, if the number of RTP streams is below 100, then one sensor per switch is excessive. In this situation, RSPAN is useful. The configuration done on the switch with SPAN, RSPAN, or ESPAN is transparent to the sensor; the sensor functions normally as long as it sees the RTP stream.
White Paper ● Cisco Voice Transmission Quality is supported from Cisco Unified Communications Manager 4.2 or later versions. ● Cisco 7906, 7911, 7931, 7921, 7962-G, 7962-G/GE, 7942-G, 7942-G/GE, 7972-G/GE, 7940, 7960, 7941, 7961, 7970, and 7971 IP Phones support Cisco Voice Transmission Quality in SCCP and Session Initiation Protocol (SIP) mode. (You must have new firmware; the firmware can be downloaded from Cisco Unified Communications Manager 4.2 or 5.x.
White Paper Hostname It is recommended that you configure the hostname for the Service Monitor server before you start installing Service Monitor. Specify the hostname when you are installing the operating system or subsequently, using My Computer > Right Click and select Properties > Computer Name. Once Service Monitor is installed, changing the hostname is a very laborious process involving file manipulation and the execution of scripts.
White Paper 2. Use your system documentation to configure NTP on the Windows Server 2003 system where Service Monitor will be installed. Configure NTP with the time server being used by call manager in your network. You might find How to configure an authoritative time server in Windows Server 2003 at http://support.microsoft.com/kb/816042 useful.
White Paper TFTP Server Cisco Unified Communications Manager 5.x or 4.2 can be used as the TFTP server. If you use Cisco Unified Communications Manager as a TFTP server, Service Monitor cannot copy configuration files to Cisco Unified Communications Manager due to security settings on the latter. You will need to manually upload the configuration file. After uploading the configuration file, reset the TFTP server on Cisco Unified Communications Manager.
White Paper Note: If Service Monitor is already using Cisco Unified Communications Manager and you are changing the cluster name, then you have to delete and re-add the Cisco Unified Communications Managers in Service Monitor for it to reflect the new cluster name. If Operations Manager is already managing Cisco Unified Communications Manager and you are changing the cluster name, then the cluster names in the service-level view will not reflect the new cluster name.
White Paper keep from the C:\Program Files\CSCOpx\data folder. Failover and Redundancy Service Monitor supports failover for only sensor functionality. A primary and secondary Service Monitor server can be configured to provide redundancy and failover support to Cisco 1040 Sensor. In case of a primary Service Monitor server going down, the probe would automatically switch over to the secondary Service Monitor server.
White Paper Figure 4. Cisco Failover Scenario Cisco 1040 Failover Deployment Preparing for Failover The first step is to have two Service Monitor servers available for configuration. One server acts as the primary server and the second as a secondary server. Please refer to the Service Monitor installation guide for the hardware specification of these servers. It is recommended that these servers connect to the network through redundant paths.
White Paper Figure 5. The Setup Dialog Box Setting Up Failover for a Specific Sensor Go to the primary Service Monitor server, select the Configuration tab, the Sensors option, and Management from the TOC. The Cisco 1040 Details dialog box opens showing a list of any previously defined or registered Cisco 1040s. Select Add to create a specific configuration for a Cisco 1040. The Add a Cisco 1040 dialog opens (Figure 6).
White Paper Viewing Failover Status Enter http:// in your browser where IP address is the address of your Cisco 1040. The Current Service Monitor field will show the Service Monitor to which the Cisco 1040 is sending data; this could be a primary or secondary Service Monitor server. Go to the primary Service Monitor server or secondary Service Monitor server if failover took place, select the Configuration tab, the Sensors option, and Management from the TOC.
White Paper In Service Monitor standalone servers, it is recommended to use the Common Services backup and restore process only when the database is less than 6 GB. For large Service Monitor databases, it is recommend to do backup manually (saving database password and copy database files) as documented in the user guide. Configuring Low-Volume Schedule and Database Purging Service Monitor needs 8 hours of low-volume time during a day.
White Paper Data purge need not run continuously for 4 hours. You can configure more than one data purge period as long as: ● The total time adds up to 4 hours ● Data purging occurs during low-volume schedule ● No data purging occurs from midnight through 2 a.m. Here are some examples: datapurge-Mon=2-5;22-23; datapurge-Tue=2-3;4-6;23-24 Note: Do not edit the properties files using Wordpad as it introduces a carriage return. Use Notepad to edit the file instead.
White Paper Q. A few rows in the sensor diagnostic reports do not show directory numbers What could be the issue? A. Make sure that the call is over. CDRs and CMRs are generated at the end of the call only, whereas sensors send data even while the call is in progress. ● Make sure that the corresponding call in the Cisco Voice Transmission Quality diagnostic report shows directory number for both the caller and callee.
White Paper ● For call manager 4.x : Both HTTP/HTTPS and CDR database credentials CDR database: Is the Microsoft SQL Server CDR database credentials on 4.x versions. HTTP is for device name correlation. ● For call manager 3.3 : Both the device database and CDR database credentials The device DB is the Microsoft SQL Server CCM030X database source credentials on 3.3 versions. This is for device name correlation ● HTTPS gets the device type for an endpoint involved in a call from the call manager..
White Paper Step 3. Read the notice to Cisco employees and click Continue. Step 4. From the navigation menu at the top-left corner of the page (above the Subscriptions link), select the Marketing Collateral link. From the Marketing Collateral navigation menu, select Network Management Evaluation Kits, and then select the desired evaluation kit. Step 5. Use Add to cart and Checkout to place the order for the desired kit, using your Access Visa or personal credit card.
White Paper Printed in USA © 2007 Cisco Systems, Inc. All rights reserved. This document is Cisco Public Information.
