soundofheaven.info Religion TROUBLESHOOTING CISCO IP TELEPHONY 2ND EDITION PDF

Troubleshooting cisco ip telephony 2nd edition pdf

Sunday, March 3, 2019 admin Comments(0)

Cisco IP Telephony You can find installation, configuration, and maintenance information for Cisco IP Telephony networks in the book Cisco IP Telephony ( ISBN. Troubleshooting Cisco IP Telephony (eBook, PDF) - Giralt, Paul; Hallmark, . Erscheinungstermin: ; Englisch; ISBN ; Artikelnr. ISBNX Troubleshooting Cisco IP Telephony shows you how to break down problems to find the root cause. Descriptions of each.


Author: JOANNA HOSTETTER
Language: English, Spanish, Dutch
Country: Jamaica
Genre: Biography
Pages: 350
Published (Last): 28.10.2015
ISBN: 272-2-65633-882-5
ePub File Size: 27.38 MB
PDF File Size: 14.23 MB
Distribution: Free* [*Regsitration Required]
Downloads: 27519
Uploaded by: YADIRA

Book; ISBN ; ISBN The only complete , authoritative guide to troubleshooting Cisco IP - now updated for CallManager. Understanding of IP Telephony architecture, concepts and . Monitor and Troubleshoot on CallManager Telephony. ISBN phones, and other Cisco IP Telephony products. She is a co-author of Cisco CallManager Fundamentals (ISBN: ) and Developing Cisco IP.

Summary This chapter provided an overview of the primary components that are involved in the Cisco IP Telephony architecture. SDL uses the default root path: As you learn which trace settings are required for specific problems. You ask the CEO and her admin if all the dropped calls were inbound calls. Users can get quite irritated if you have to ask them for the same piece of information two or three times. A dedicated headset port eliminates the need for a separate.

One is the communication between CallManager and the voice gateway. The other is the communication between CallManager and the phone. So how do you determine which one is causing the problem? One important distinction to make that will become evident as you read through this book is that many problems can be narrowed down to being either signaling-related or voice packet-related. You know that the call in question was set up around 5: Chapter 3 provides more details on where to find these traces and how to read them.

Analyzing the Data As soon as you have a clear understanding of the problem you're trying to resolve.. This eliminates the CEO's phone as a cause of the problem because the. As part of the data analysis stage. If you don't know the times that the other calls were dropped.

If there were a network problem. Figure shows you've narrowed down the network to only a few devices. Network After You Continue Narrowing Down the Possible Suspects The next step is to go to the suspected gateway and try to determine why one of the calls was dropped.

It would not hurt to look through the network devices between CallManager and the voice gateway to ensure that there are no network errors. This involves turning on additional debugs on the gateway to determine if the gateway is disconnecting the call or just passing along information. Because the user indicated that there were three drops. Because CallManager received a message from the gateway telling it to disconnect the call.

Depending on what you discover on the gateway debugs. When deployed in a large enterprise. As you begin this journey. Chapter 2. IP Telephony Architecture Overview. Chapter 6 discusses these considerations in detail. Which debugs to use depends on the gateway model and the type of interface to the PSTN. This is why it is so important to narrow down the problem to a small subset of devices: You do not want to turn on debugs on dozens of gateways.

Then dig deeper into each component by breaking the problem into more manageable pieces. Many basic problems can be avoided by using a consistent troubleshooting approach. The point of this example is not to teach you how to troubleshoot a specific problem or to find out exactly why the user's calls are being dropped.

It is vital that you always follow a consistent approach to troubleshooting. Summary This chapter discussed the methodology you should employ to successfully troubleshoot problems in an IP Telephony network.

What areas are you unsure about? Are you strong in IP but weak in call processing skills? Are you familiar with the basic protocols that are used? Consider where you are now. Conclusions As this case study has demonstrated. You should become familiar with the methodologies discussed here. The same principles can be applied to almost any problem you are troubleshooting. So remember. It is to show you how to approach a problem in order to isolate it and break it into more manageable pieces.

While waiting for the problem to reoccur. Your network design must be built for high availability. Triple call processing server redundancy improves overall system availability. CallManager clustering yields scalability of up to With this overview as the starting point. The infrastructure includes switches and routers. A voice-enabled network is a quality of service QoS -enabled network that gives precedence to voice. By interlinking multiple clusters. IP phones. This chapter covers the basic components of the IP Telephony architecture in order to get a big-picture viewpoint of the system.

Video and Integrated Data includes many different components that come together to form a comprehensive architecture for voice. Multiple CallManager servers are clustered and managed as a single entity. The benefit of this distributed architecture is improved system availability and scalability. Figure shows an example of each separate site connecting via the PSTN. Multiple sites exist. Multiple-Site Deployment Model.

Figure shows an example of these components located at a single site. Centralized Deployment Model. Figure illustrates the centralized deployment model. Centralized Deployment Model A centralized call processing deployment model centrally locates CallManager. Locations-based call admission control prevents over-subscription of the WAN.

At each remote site. The centralized call processing model is really the same as the single-site deployment model with the addition of remote sites across the WAN. Figure depicts a distributed deployment model. CallManager and applications are located at each site with up to One hundred or more sites could be interconnected via H. Distributed Deployment Model In a distributed deployment model.

Distributed Deployment Model. Although infrastructure is not the primary focus of this book. Cisco Unity. Figure shows the Cisco family of IP Phones. Cisco offers several models with different functions. Table Clients consist primarily of IP phones. It includes the following features: Available features include call park.

It has the following Module features: Up to two Cisco s can be connected to a Multicolor button illumination allows you to identify which lines are ringing.

The phones can also receive power down the line from any of the Cisco inline power-capable switches or the Cisco inline power patch panel. The Expansion Module lets you add 14 buttons to the existing six buttons on the phone. A second version of the phone. The system administrator can designate separate VLANs A dedicated headset port eliminates the need for a separate.

The phone provides a mute button for the handset and headset microphones. You can attach a headset by removing the handset and using the port into which the handset cord was attached. Because these phones are largely soft key-based. The 14 buttons on each Expansion Module can be programmed as directory numbers or speed dial buttons.

The plugs into a standard RJ Ethernet connection. Cisco IP Phone This low-end phone features on-hook dialing and call monitor mode but does not include speakerphone capability. These phones feature a large pixel-based display that allows for XML-based applications on the phone. The is an IP-based. Using Skinny on these gateways allows the gateway to provide features such as message waiting indicators MWI. Each port on these gateways registers individually as a phone device.

MGCP provides the following services: Although the does not accept inline power from a Cisco inline power-capable switch.

CallManager controls routing and tones and provides supplementary services to the gateway. These gateways interoperate with CallManager using various protocols. Two gateways. Compared to MGCP. Windows Cisco ER features a real-time location-tracking database and improved routing capabilities to direct emergency calls to the appropriate Public Safety Answering Point PSAP based on the caller's location.

Summary This chapter provided an overview of the primary components that are involved in the Cisco IP Telephony architecture. When troubleshooting. Personal Assistant provides rule-based call routing. Windows NT. Depending on the problem you encounter and your particular skill set.

These traces are usually reserved for Cisco development engineering use. Later chapters demonstrate the use of these tools in different scenarios you might encounter.

This chapter covers the following topics: If you are strong in IP. In addition. SDL traces describe the events occurring in the CallManager software at a code level. CCM trace files provide information about call processing events and all messages exchanged between Skinny. The resulting matrix shows when each bug was integrated or fixed if applicable. The Enhanced Q. Signaling also occurs between the respective CallManagers of each endpoint device. When trace files are collected for analysis.

This distributed architecture creates a highly available and scalable system. Time synchronization is critical. This ensures consistent timestamps regardless of which device you are looking at. CallManager Serviceability. When you synchronize the time on all involved servers. Also points you to a section in the "Introduction" with a recommended reading list. As endpoints such as IP phones and voice gateways call each other.

Time Synchronization Time synchronization is simply making sure that all the participating CallManager servers and network devices have the same exact time. A large CallManager cluster can have eight or more separate servers. In addition to CallManager servers. It also makes the troubleshooting process more involved because you have to collect traces from all participating servers to see the full picture of what happened.

The various information elements are decoded for ease of reading. If a problem occurs. Step 4. You can configure CallManager to point to specific time servers see Example To synchronize with a remote Time Server. Example Allow several minutes for the update to take place. Sample ntp. Synchronizing Time Manually on CallManager Servers Use the following steps to manually configure time synchronization. Synchronize the clock by using one of the following commands from a command prompt.

Step 3. Configure the C: Use the following steps to configure the CallManager server to automatically synchronize—and stay synchronized—with a Time Server. This file contains the list of time servers that CallManager will synchronize with. Expand the Services and Applications section. Right-click My Computer and select Manage. Verify that the NetworkTimeProtocol service is configured to launch automatically upon startup. Double-click the NetworkTimeProtocol service.

You can do this by configuring the command ntp master on the IOS device. First you must configure the time zone. You should configure the IOS device to take daylight savings time into account as well if you live in a time zone that observes daylight savings time. For Cisco IOS devices. If you are not making the device an NTP master. Once your time zone is configured properly. If you do not have a device on the network running NTP. To enable NTP.

Future chapters continue to show CCM trace examples as you learn how to troubleshoot more problems. After you learn a few tricks. CCM traces are usually the first place to look when troubleshooting most problems. Later in this chapter you also learn about the Q.

It is not a substitute for learning how to read the CCM trace. By the end of this section. You can analyze problems related to device registration. Then enable the NTP client using the command set ntp client enable. Once you have the time zone configured properly.

As you learn which trace settings are required for specific problems. Setting the Appropriate Trace Level and Flags CallManager allows you to select from a variety of different options that adjust which events are logged to the CCM trace files. For this reason. Figure shows the top half of the Trace Configuration page.

Pdf 2nd telephony cisco edition troubleshooting ip

When you are beginning to learn how to read CCM trace files. See the later "Q. If you configure the trace to collect too much information. If you know exactly what you are looking for. Unfortunately you can never predict what problems will come up. This must be selected for any of the trace settings to be available.

First thing to note is the Trace On checkbox. CallManager Serviceability provides six different levels. Because of the distributed nature of CallManager. You generally want to keep the same level of tracing enabled on all the servers in the cluster unless you are absolutely certain the problem is isolated to a single server in the cluster.

The Apply to All Nodes checkbox allows you to apply the specified trace settings to all CallManager servers in the cluster. This is useful when you are troubleshooting a problem that is occurring on more than one CallManager server. The Trace Filter Settings area allows you to specify the exact parameters of your trace.

Table describes the trace fields to choose from. This level is best suited for debugging difficult problems. It can cause performance degradation on CallManager. All system and device initialization traces are at this level.

Most are reasonably self-explanatory. Activates the logging of events related to the legacy gateways. Enable Activates a trace of miscellaneous devices. CallManager uses asynchronous tracing to reduce the impact that trace file generation has on call processing.

For CallManager releases 3. In release 3. This level includes nearly everything that is included in Detailed with the exception of KeepAlives. Miscellaneous Trace Enable Conference Activates a trace of the conference bridges. Enable DT. Other trace levels are provided. If you are not troubleshooting problems related to missed KeepAlives. Use this level to Bridge Trace trace conference bridge statuses such as.

Trace Enable Forward Activates a trace of call forwarding and all subsystems not and Miscellaneous covered by another checkbox. Activates CallManager real-time information traces used by the Time Information real-time information server. Enable All Phone Activates a trace of phone devices. Figure shows the bottom half of the Trace Configuration page.

The non-device option is a catch-all. Tracing based on specific devices is very useful when you know which devices are involved in the problem. When that checkbox is selected. Once you've specified the trace settings. For each server on which you're running trace. The Maximum No. We don't recommend going above One downside to this. You should never need to or be asked to enable this checkbox.

The trace output is then sent to the path specified in the File Name field. The new trace settings take effect immediately when you click Update. This could be servera-ccm. At these average sizes. In most cases. For traces to be logged to a file. The downside. The Enable Debug Output String checkbox sends debugging information to a Microsoft development tool useful only to Cisco development engineers.

We're going to skip discussion of the Enable XML Formatted Output for "Trace Analysis" checkbox for a moment to finish the fields related to file settings. The default of for the Maximum No. If you do not enable XML-formatted output. Set the service you want to retrieve the trace files for such as Cisco CallManager. The Trace Configuration page in CallManager Serviceability validates the filename and ensures that it has a. Do not use a filename that exists on another computer. Depending on the time period you specified.

It is a good practice to have each server use a filename format that has the server name in it. Click Update to save your settings. With Trace Analysis. If you did. Use a filename that exists on the computer running the trace.

Assuming you don't go above That way. Trace shows all events as specified in. We generally recommend that you do not use the trace collection utility for anything other than very small traces on a system that is not busy. Manually copy the traces off the server s in question and analyze them offline. It is usually quicker and easier to manually collect the traces yourself using either Terminal Services or VNC discussed later in this chapter to access the CallManager server and copy the files to another machine for analysis.

If you choose to view the text- based output in a new window. If you specify a device name. In this dialog box. Once the trace has been gathered. Trace Type Select this checkbox if you want to include the trace type in the trace output.

Information Select this checkbox if you want to include a description of what the trace found in the trace output.

Pdf 2nd cisco troubleshooting telephony ip edition

CM Node. Date and Time. Figure shows a trace file formatted as XML output and filtered to display only one CallManager host. Source IP. Date and Select this checkbox if you want to include the date and time of each Time event listed in the trace output.

Alarm shows only specific messages that meet the criteria of being an information Alarm message. IP Address Select this checkbox if you want to include the device's source IP address in the trace output. Cluster Select this checkbox if you want to include the cluster name in the trace output. Device Name Select this checkbox if you want to include the device name in the trace output. Correlation Select this checkbox if you want to include the number that correlates Tag traces with each other in the trace output.

Application Select this checkbox if you want to include the directory numbers Name DNs and other service-specific information in the trace output. The following section describes how to read a text-based CCM trace. As you can see. Reading CCM Traces This section shows you a few call flow examples and highlights key information in each example that helps you understand the CCM trace.

For that reason. So the same trace is shown in this book as StationInit: For brevity. You can click the Back to Selection link to return to the Trace Analysis dialog and filter based on different criteria. You should learn how to read the CCM traces directly from the text files to help you troubleshoot problems more quickly and accurately. In other words.

A Skinny station is any endpoint that uses the Skinny protocol to communicate with CallManager. Phone A and Phone B. The header portion of the trace line just specifies the date and time when the trace event was generated and which trace file you are looking at.

The IP address is When you first open a CCM trace file. This includes the Cisco 79xx family of IP phones. We recommend that you use the default trace settings for CCM traces except you set the trace level to either Arbitrary or Detailed. IP phones use the Skinny protocol to communicate with CallManager. For a CCM trace file. Appendix C. You can determine the IP address by working backwards: Phone A goes off-hook and you see the following line in the CCM trace: Phone A calls Phone B.

With the TCP handle. This means that if you search for the directory number of the IP phone. If you know the phone number of the calling IP phone. The OffHook message means that CallManager received a Skinny message indicating the phone went off-hook.

This signifies that CallManager is sending a Skinny message to the phone. The next message is StationD: In CallManager 3. Once you find the KeepAlive message. Again you see the same TCP handle. When searching through a CCM trace file in Notepad. The number represents the directory number of the phone. Skinny message transmission such as this between the IP phone and CallManager occurs for every action undertaken by the IP phone. Again you see that all the trace lines begin with StationD indicating that these are messages from CallManager to the IP phone and you see each line has the same TCP handle.

In particular. These are all Skinny messages sent to the IP phone. Note the callReference ID. Other messages sent to the IP phone include the following: It is best to search for the device name you are looking for and find a KeepAlive to get the TCP handle as discussed earlier.

Do not concern yourself at this point about exactly what each of the pieces in the trace mean. Chapters 4 and 5 provide additional detail relating to the Skinny messaging you see in the preceding output. This time. Each leg of a call gets its own callReference ID assigned. You will notice. A new callReference ID is created for each participant in a call and when some features are invoked.

So far you have only seen Skinny protocol messages. Now it is time to ring Phone B. Whenever digit analysis makes a match for a call. Notice that after the phone is told to stop the tone. There is. The Skinny protocol does not provide a mechanism to specify which tone to stop. This stops any tones the IP phone happened to be playing at the time. As soon as the first digit is dialed. Notice that a 3 is dialed and a tone is then stopped. That makes sense because when you pick up the phone.

So for example. Digit analysis: CallManager is constantly analyzing the digits the user dials. Chapter 9. The important concept to grasp here is that any time digit analysis makes a match. This call reference persists for the duration of the call on Phone B. Because CallManager has collected all the required digits. Also notice that the TCP handle is different than in the preceding trace output. Chapter 9 explains some of the concepts such as partitions and calling search spaces.

Notice also that the callReference value is different from the previous output. Do not be concerned about what each of the fields means at this point. As we mentioned earlier. If you look at a CCM trace. These digit analysis results are easy to spot in a CCM trace because of the white space to the right of the digit analysis results. This means that a message is sent from CallManager to the IP phone.

Future chapters go into detail about each message. Phone A gets some updated information. Now that a call is in progress. Phone B rings and the call information shows that James called Mary. Remember each call reference is only valid for one leg of the call. Once again. These are just standard messages sent by CallManager. You also see the first callReference value.

The alerting tone is the standard ringback tone you hear when placing a call. You see several messages that seem to suggest the call is being redirected. A logical channel is a unidirectional RTP stream. In a call involving a Skinny device. Each RTP stream is called a logical channel. Phone A responds first: As with all VoIP protocols. CallManager asks the IP phone for specific parameters for this connection. Until this point. All VoIP devices are blindly told to send RTP packets to an IP address and port number without knowing what type of device they are sending these packets to.

Appendix C provides a cheat sheet for conversions. This is what allows CallManager to set up calls between devices that use different signaling protocols. MGCP is not described in detail in this example. When reading CCM traces. The majority of information presented in the CCM trace file for a gateway call is in hexa-decimal notation. Also it is important to separate the signaling aspects of a call set up from the RTP media streams.

As long as the terminating device provided the correct IP address and port number and CallManager relayed this information correctly. All Q. Phone A has never sent nor received any Skinny signaling to or from Phone B. Once you understand how to decode them. The Q. This makes understanding the basic structure of this kind of trace message important.

This is because all the signaling goes through CallManager. Notice that for the duration of this call. Phone A has no idea that it is sending RTP packets to a voice gateway and vice versa. Do not be intimidated by the hexadecimal values. The only time IP phones send packets to each other is for the actual voice stream. Out Message -. When viewing an ISDN trace. Information elements are covered in detail in Chapter 6.

You also know the call reference from decoding the first few bytes of the IsdnMessageData2. In this example. The direction is from the perspective of CallManager. You might be wondering how a4e81cac is converted to an IP address in dotted decimal notation. Whenever an H. Ignore the first two numbers usually This is a hexadecimal representation of the IP address.

PriSetupMsg -. You can figure out the IP address by working backwards. They are equivalent. As with the IP phone. The next two numbers are the call reference length PriCallProceedingMsg -. QChannelIdIe -. Notice that the message that follows is an In Message. These are just some of the tricks that help you follow call flows through the CCM trace.

The most-significant bit MSB is set on the call reference value. Half the battle in reading a CCM trace is knowing which pieces of the trace file to ignore so that you can focus on the important messages in the trace.

In Message -. When searching. To the average CallManager administrator. Chapter 6 goes into detail about how this works. The first bit. This bit determines if this message is the originating or terminating side.

This represents the directory number that is being called. SDL trace files are far too detailed for normal practical use. As you read through the following chapters.

All you need to know is to search for the call reference minus the first digit. The call reference for a call remains the same for the duration of a call. After the name follows the data contained in that information element.

In this case the call reference is 00 If you search through the CCM trace. Subsequent chapters provide additional trace examples as we investigate other troubleshooting scenarios. The format of the data that follows is dependent on the particular information element. Now you can follow the call's events. Fortunately for most day-to-day activities you do not need to reference the Q. You need to know how to configure the trace files to capture the right information for TAC.

SDL generates two types of files: Although the SDL trace files are generally not used for troubleshooting purposes. SDL traces can span multiple servers. An index file allows SDL logging to start where it left off each time an application is restarted. Log filenames are composed of the following: The number of files determined by the CallManager service parameter SdlTraceTotalNumFiles and the number of lines per file determined by the CallManager service parameter SdlTraceMaxLines governs how long it takes to overwrite old log files.

This mechanism is supported by the use of SDL links. SDL maintains a circular queue of files to log information. Over time. In cases like this. These signals are just messages from one piece of software to another internal to CallManager. SDL is what allows CallManager clustering to work by allowing processing for tasks to run on any node in the cluster. All columns in a log entry are delimited by a vertical bar character 0x7c.

The common prefix of a trace line is as follows: But what if an IP phone registered to one CallManager in the cluster places a call out a gateway registered to a different CallManager in the cluster? The prefix always has the same format.

These signals can be sent to a piece of code on an entirely different CallManager. CallManager functions by sending signals from one process to another. Prefix Component Detail Component Every log entry contains a prefix. A log entry is broken into two components. Because CallManager is a distributed architecture.

The SDL trace lets you see this happening. The best way to see this is by looking at the log entry type field at the beginning of an SDL trace file. This is the reason why having the clocks synchronized on all servers is so vitally important.

You do not need to understand what the various process names signify. The third column shows the destination Db 1. The two columns to examine are the third and fourth columns after SdlSig-O.

So how do you know which node the signal was coming from or destined to? After the SdlSig-O or SdlSig-I message you should see several columns of text separated by the vertical bar character. These describe the process sending the signal and the process to which the signal is being sent.

The important part is the four comma-separated numbers that follow the process name. Without time synchronization. Now you need to understand what you are looking at when you see Db 2. We have condensed the trace to fit on the printed page. You might be wondering why you need to know all this. Db is the process name. When you see an unexplained event in the CCM trace. The default is Bit masks are usually represented as strings of hex digits.

A bit mask is a string of bits that each represent a particular trace setting. See Table for details. If insufficient disk or CPU bandwidth exists to write the trace file. Each 1 in the bit mask indicates that a particular trace should be enabled. There is rarely any reason to change this value. Table shows the parameters you can adjust. If this parameter is set to False. The port is by default. The default is 5. SDL trace entries are flushed to disk when there is spare disk bandwidth not being used for call processing.

Detailed SDL tracing consumes a lot of disk space and affects the processor on the CallManager server. Setting this parameter to True. The default is True. The default is False.

If this path is not defined or is defined incorrectly. The default is files. SdlTraceTotalNumFiles This value indicates the maximum number of files that can be created for logging purposes. You should set this to False during normal operating conditions. SDL uses the default root path: SdlTraceMaxLines This value indicates the maximum number of lines written to a log file before a new log file is created. You can calculate the size of your trace files by figuring the average CCM trace at Set this flag to True to turn tracing on or False to turn tracing off.

SdlTraceTypeFlags This field indicates the bit mask value for collecting the trace type flag of choice. The default is bytes. SdlTraceDataSize For signal types. Determine how much free disk space you have. This information appears in the freeform information column at the end of each line in the SDL trace file.

Be sure to monitor free disk space by using other tools mentioned in this chapter such as PerfMon or CCEmail. Pretty print adds tabs and spaces in a trace file without performing post processing.

See Table for more details. Not used — 0x This bit is not used. TranslateSdlToIsdnRes traces. It contains logging facilities to take snapshots of any counter at user-defined intervals. PerfMon Advantages.

CallManager includes several performance objects that let you monitor various counters related to the operation of the CallManager services and associated devices. The RTMT. An example of a standard performance object is the Processor object. It is colloquially referred to as PerfMon. PerfMon allows you to look at real-time statistics.

If statistics are disabled. PerfMon can be used to monitor a variety of performance objects. Statistics are enabled by default. A performance object is a set of counters reported by a process or application running on the system that can be monitored.

A third-party tool called CCEmail discussed in the next section allows you to configure alerts in PerfMon. The application looks similar to Figure You'll read more about the additional capabilities of the RTMT later in this chapter. An example of how to view some real-time statistics with PerfMon will better familiarize you with the tool. This kind of information can easily be obtained through PerfMon.

In addition to performance-monitoring capabilities. You can configure specific counters and save the configurations in RTMT. This CSV file can then be imported into a spreadsheet application for further analysis.

Three buttons on the toolbar shown in Figure are used to switch between the different formats. Click the Add button on the toolbar. Click the View Report button on the toolbar to gray out the area below. You see a dialog box similar to Figure For viewing real-time statistics. PerfMon has three formats to display data: You can see that the counters are updated every second or so. Below the object selection you can choose which counter to monitor.

Select this counter and click Add. The dialog box remains open so that you can continue selecting and adding counters. When you are done adding counters. Chapter 1. Troubleshooting Methodology and Approach It's 5: You recognize the phone number— it's your CEO's administrative assistant.

As the administrator of the company's phone IP Telephony network, you assume there's a big problem. You rush into work and find the CEO's administrative assistant, who states that several calls for the CEO have been disconnected in the middle of the call, including a call from a very important customer.

Where do you start? Troubleshooting a Cisco IP Telephony network can be a daunting task. Rather than describing step-by-step how to solve specific problems subsequent chapters provide that information , this chapter focuses on teaching a good troubleshooting methodology: A typical IP Telephony network consists of—at the very least—one or more of the following components: These components are in addition to the data network infrastructure that supports voice over IP VoIP traffic.

More-complex installations can have dozens of servers for different services and redundancy, each server running a variety of applications, as well as hundreds or thousands of IP phones and a large number of voice gateways. The Skinny protocol is covered in greater detail in Chapter 5. Understanding the Skinny protocol is essential to understanding how the phone operates and how to troubleshoot problems with it. Before exploring the myriad of tools. Some clues lead you to additional clues.

As soon as you've got all the clues. In addition to the information in this book. On the other hand. In the case of a service-affecting problem during production hours. In contrast. The following list is a general guide for steps to take when troubleshooting an IP Telephony problem: Step 1. Gather information from the end users. Use topology information to isolate the problem. During a new install or scheduled outage window.

Identify and isolate the problem. Verify IP network integrity. Gather data about the problem: Note that although percent CPU of a high-level process can cause sluggish behavior or delayed dial tone. Use deductive reasoning to narrow the list of possible causes. After you restore service. Analyze the data you collected about the problem: For example. Production Versus Nonproduction Outages Troubleshooting a problem can occur in one of two timeframes: Troubleshooting a problem can be broken down into two stages: The downside of this approach is that you might not be able to further troubleshoot the problem when the process is restarted.

Determine the problem's timeframe. CallManager provides many diagnostic traces if they are enabled prior to the problem that you can reference after a problem has occurred to see what was happening on CallManager at the time of the problem. Step 2. Determine the proper troubleshooting tool s. Identifying and Isolating the Problem Half the battle in troubleshooting a problem is determining which piece of the puzzle is the source of the problem.

One of the first lines of defense is possessing current topology information. If you encounter an event where you are unable to determine the root cause due to insufficient information. To help you visualize the big picture. You must also determine which parts of the problem are symptoms and which are the root cause of the problem.

When multiple problems occur simultaneously. Troubleshoot the dropped-call problem first because keeping calls connected is more critical than removing the occasional echo on an active call. You must always remember to look at the big picture when searching for the root cause and not let the symptoms of the problem lead you in the wrong direction.

You've encountered a problem. As of CallManager 3.

Pdf ip 2nd edition troubleshooting cisco telephony

So although the symptom is a phone reset. With so many different pieces composing an IP Telephony network. The network diagram should include network addressing information and the names of all the devices. In this case. Look at the percent CPU as a possible symptom but not necessarily the root cause. This information will prove invaluable when you try to isolate which components are involved in a particular problem.

One of the most important pieces of topology information is a detailed network diagram usually created using Microsoft Visio or a similar application. It should also clearly show how the devices are interconnected and the port numbers being used for these interconnections.

The first thing to do is gather as much information about the problem as possible. Step 1: In fact. Using Topology Information to Isolate the Problem You can take many proactive steps to help make the troubleshooting process easier. Because Figure is a high-level diagram. For a small network. Figure shows a typical high-level topology diagram for a large enterprise IP Telephony network. Knowing where a call is. For medium. You also need documentation of your dial plan.

This makes troubleshooting easier by allowing you to quickly look up devices to access them. Be sure to keep this information in a secure location. Figure For larger deployments. Some deployments. In addition to the network diagram.

Notice that device names and IP addresses are listed in the diagram. Many customers keep all this topology information on a web server as well. If the user's phone is connected to Access Switch 1A. Gathering Information from the User Information the user provides can be vital to your ability to correct a problem. Use all the topology information you have to narrow down which pieces of the network might be involved in the problem you are trying to troubleshoot.

Try to gather as much detail as possible on exactly what the problem is. Assume that the network you are troubleshooting looks like Figure The more detail about the problem you can gather before you begin troubleshooting.

If a topology drawing is not available. IP addressing for all network devices routers. Without a network diagram.

If you're going to keep topology information highly recommended. When your topology information is complete. What is worse than not having topology information? Having incorrect topology information can lead to countless hours heading down the wrong path. A general topological understanding of the network or at least the piece of the network in question helps when you're trying to differentiate the problem from its symptoms. If any patch panels exist between devices. Often when troubleshooting a problem.

To further isolate the problem. It's necessary when you're trying to isolate the problem to a particular part of the network. Users can get quite irritated if you have to ask them for the same piece of information two or three times.

Many users report that they get a busy signal when dialing into their voice mail. Phone numbers for all parties involved in the problematic call or calls.. Clearly in this example you need more voice mail ports or servers to handle call volume. When relying on end users to give "when" information about a problem. Also point out to the user the importance of letting you know immediately after a problem occurs. Sometimes the proper diagnostic tools are not enabled when the problem occurs. It is important to know whether they are attempting to do this at 9: This might change the problem from a troubleshooting issue to a load-balancing or equipmentexpansion issue.

Determining the problem's earliest occurrence can help correlate the problem with other changes that might have been made to the system or other events that occurred around the same time. This includes text messages displayed on the phone or recorded announcements. Be sure to turn on tracing or debugs before making the request so that when the problem occurs again.

User observations. Here is some general information to collect from users: Sometimes the information provided by an end user is not enough to even begin troubleshooting. As another example. Information about the user's device. You can use this as search criteria if you need to look through traces.

As long as you have the time on your CallManagers and network devices synchronized. Determining the Problem's Timeframe In addition to what the problem is. You check the voice mail system and notice that at the time the problem was reported.

The phone's time is synchronized with the clock on the CallManager to which the phone is registered. Actions performed by the user when the problem occurred. This includes what buttons were pressed and in what order. TIP Although it is important to use information about when the problem started happening. The user probably won't know the answer. In some cases. If so. The more you can isolate the problem. He might think the problem happens only on his phone.

Step 2: Analyzing the Data Collected About the Problem Now that you have collected data from a variety of sources. Does the problem happen only between IP phones. What numbers are being called when the problem occurs? The answer to this question helps determine which parts of the system are being used when the problem occurs.

Using Deductive Reasoning to Narrow the List of Possible Causes The next part of your fact-finding mission is to identify the various components that might be involved and to eliminate as many components as possible.

If you have information about when. Although not all of the following apply to every problem. The network is always a consideration when you encounter certain problems. You should use the tracing and debugging facilities available in CallManager and other devices to determine exactly what is happening.

You would then check Layer 3. Use your topology information to help obtain this information. Network health is especially important during the discussion of voice quality problems in Chapter 7. As an example.

Check your physical layer connectivity cables. Upon further investigation.

A degraded network or a network outage can cause a wide range of problems. If you fix the problem for that one user. Additional tools and traces are covered in the chapter associated with diagnosing certain types of problems. Most components have multiple troubleshooting tools available to help you. Determining the Proper Troubleshooting Tool After you narrow down the appropriate component s causing a problem and have detailed information from the user s experiencing the problem.

Taking the layered approach. Chapter 3. Then make sure you have Layer 2 connectivity by checking for errors on ports. Continue working your way up the stack until you reach the application layer Layer 7. When you reach this layer. Chapter 6. You head into the CEO's office and look at the list of received calls and placed calls for the morning.

Sometimes the problem description you have is not detailed enough to determine which tool to use. You notice that a call was received at 5: Please help ASAP! He states that at various times during the previous day and one time this morning. The following case study shows how this troubleshooting methodology works in a real-world scenario. Eager to resolve the problem.

This is the extent of the information he remembers. This step is the most demanding on your troubleshooting skills because you analyze the detailed information provided in the various tools and use it to search for additional clues using other tools. Case Study: You must gather the data before you can begin the analysis. The only data you have is the page you received at 5: You notice that the second call was placed to the same area code and prefix as the call that was received. This case study applies the methodology previously described.

The CEO has a phone that stores information locally about missed calls. Most users do not pay attention to specifics like this unless they have been instructed to. Because CallManager is central to almost all problems. Gathering the Data As part of the data-gathering stage.

Now you know that the problematic call was received at approximately 5: The telephone company has set up the inbound calls so that the 32 gateways are redundant whereby if one of the gateways is down. This lets you isolate which CallManager in the cluster is involved in the signaling for this phone. You refer to your topology diagram to isolate the components that are involved. She also confirms that the first call that was received was the dropped call. All outbound calls prefer the first gateway.

Armed with this information. While you are looking at the CEO's phone. Figure 12 shows a high-level diagram of the network topology. She remembers that she was on the phone with a customer for about 15 minutes when the call was disconnected. You ask the CEO about the two calls.

She immediately called the customer back. Two gateways at each remote site used for both inbound and outbound calls. With just the information you have so far. As shown in Figure You know that the call this morning was during a time of day where there is little phone activity. You then look at the configuration for the two gateways at Remote Site 2 and note that they are both configured to send incoming calls to CallManager Subscriber 3 as their preferred CallManager and CallManager Backup 1 in case CallManager Subscriber 3 fails.

Remember that all inbound calls to the remote site come in through Primary Rate Interfaces PRIs connected to the remote voice gateways and that inbound calls to the site prefer the second gateway. It is unlikely that all the channels on the first PRI were in use during a time of low call volume. You also know that the problematic call was an inbound call. With the information you have so far. As far as they can remember. You ask the CEO and her admin if all the dropped calls were inbound calls.

So far you know that the problematic call was to the CEO. At this point. Now that you know the problem is related to inbound calls. Keep in mind that you haven't eliminated the possibility that the problem is on CallManager or is networkrelated. Determine whether these calls all come directly to the user or if the call flow has any intermediate steps.

Armed with this knowledge. If the problem is more widespread than this one user.

Troubleshooting Cisco IP Telephony (2nd Edition)

For the sake of this example. This is not to say that other users are not experiencing similar problems. After combing through the trace file. The phone and voice gateway never directly exchange signaling data. Now it is time to start analyzing the data. You can then analyze the trace files to discover the device that disconnects the call from CallManager's perspective—in other words. Analyzing the Data As soon as you have a clear understanding of the problem you're trying to resolve.

As part of the data analysis stage. You know that the call in question was set up around 5: The other is the communication between CallManager and the phone. The CCM traces discussed in Chapter 3 indicate which gateway the calls are coming from. This concludes the data-gathering piece of your investigation. One is the communication between CallManager and the voice gateway. The trace includes all the messaging between CallManager and both the phone and the gateway.

Because nearly all signaling for a call must go through one or more CallManager servers. So how do you determine which one is causing the problem? One important distinction to make that will become evident as you read through this book is that many problems can be narrowed down to being either signaling-related or voice packet-related.

All signaling goes through CallManager. After you isolate the problem. Chapter 3 provides more details on where to find these traces and how to read them. This eliminates the CEO's phone as a cause of the problem because the.

A call between the CEO's phone and the voice gateway has two distinct signaling connections. It would not hurt to look through the network devices between CallManager and the voice gateway to ensure that there are no network errors.

Figure shows you've narrowed down the network to only a few devices. Because the user indicated that there were three drops.

Because CallManager received a message from the gateway telling it to disconnect the call. This involves turning on additional debugs on the gateway to determine if the gateway is disconnecting the call or just passing along information.

Network After You Continue Narrowing Down the Possible Suspects The next step is to go to the suspected gateway and try to determine why one of the calls was dropped.

If you don't know the times that the other calls were dropped. If there were a network problem. The point of this example is not to teach you how to troubleshoot a specific problem or to find out exactly why the user's calls are being dropped.

When deployed in a large enterprise. While waiting for the problem to reoccur. So remember. Which debugs to use depends on the gateway model and the type of interface to the PSTN.

The same principles can be applied to almost any problem you are troubleshooting. Depending on what you discover on the gateway debugs. It is to show you how to approach a problem in order to isolate it and break it into more manageable pieces. You should become familiar with the methodologies discussed here. What areas are you unsure about?

Troubleshooting Cisco IP Telephony Full

Are you strong in IP but weak in call processing skills? Are you familiar with the basic protocols that are used? Consider where you are now. As you begin this journey. IP Telephony Architecture Overview.

Many basic problems can be avoided by using a consistent troubleshooting approach. Conclusions As this case study has demonstrated. Summary This chapter discussed the methodology you should employ to successfully troubleshoot problems in an IP Telephony network.

Chapter 6 discusses these considerations in detail. It is vital that you always follow a consistent approach to troubleshooting. This is why it is so important to narrow down the problem to a small subset of devices: You do not want to turn on debugs on dozens of gateways.

Chapter 2. Then dig deeper into each component by breaking the problem into more manageable pieces. The infrastructure includes switches and routers. The benefit of this distributed architecture is improved system availability and scalability.

By interlinking multiple clusters. IP phones. With this overview as the starting point. Multiple CallManager servers are clustered and managed as a single entity. Your network design must be built for high availability. Triple call processing server redundancy improves overall system availability. Video and Integrated Data includes many different components that come together to form a comprehensive architecture for voice. A voice-enabled network is a quality of service QoS -enabled network that gives precedence to voice.

This chapter covers the basic components of the IP Telephony architecture in order to get a big-picture viewpoint of the system. CallManager clustering yields scalability of up to Multiple sites exist.

Multiple-Site Deployment Model. Figure shows an example of each separate site connecting via the PSTN. Figure shows an example of these components located at a single site. The centralized call processing model is really the same as the single-site deployment model with the addition of remote sites across the WAN.

At each remote site. Centralized Deployment Model. Locations-based call admission control prevents over-subscription of the WAN. Centralized Deployment Model A centralized call processing deployment model centrally locates CallManager. Figure illustrates the centralized deployment model.

Distributed Deployment Model. Figure depicts a distributed deployment model. Distributed Deployment Model In a distributed deployment model. One hundred or more sites could be interconnected via H. CallManager and applications are located at each site with up to Cisco Unity. Although infrastructure is not the primary focus of this book. Figure shows the Cisco family of IP Phones.

Cisco offers several models with different functions. Table Clients consist primarily of IP phones. It includes the following features: Available features include call park. An expansion module for the Cisco IP Phone that provides 14 additional line or speed dial buttons.

It has the following features: The Expansion Module lets you add 14 buttons to the existing six buttons on the phone. The plugs into a standard RJ Ethernet connection. Cisco IP Phone This low-end phone features on-hook dialing and call monitor mode but does not include speakerphone capability.

The 14 buttons on each Expansion Module can be programmed as directory numbers or speed dial buttons. The system administrator can designate separate VLANs Because these phones are largely soft key-based. You can attach a headset by removing the handset and using the port into which the handset cord was attached. The phones can also receive power down the line from any of the Cisco inline power-capable switches or the Cisco inline power patch panel. A second version of the phone. Up to two Cisco s can be connected to a These phones feature a large pixel-based display that allows for XML-based applications on the phone.

A dedicated headset port eliminates the need for a separate. The phone provides a mute button for the handset and headset microphones. Multicolor button illumination allows you to identify which lines are ringing. CallManager controls routing and tones and provides supplementary services to the gateway.

Compared to MGCP. The is an IP-based. MGCP provides the following services: Two gateways. Although the does not accept inline power from a Cisco inline power-capable switch. Using Skinny on these gateways allows the gateway to provide features such as message waiting indicators MWI.

These gateways interoperate with CallManager using various protocols. Each port on these gateways registers individually as a phone device. Windows NT. Personal Assistant provides rule-based call routing. Cisco Emergency Responder— An application that allows emergency agencies to identify the location of callers and eliminates the need for any administration when phones or people move from one location to another. Cisco ER features a real-time location-tracking database and improved routing capabilities to direct emergency calls to the appropriate Public Safety Answering Point PSAP based on the caller's location.

Personal Assistant— A software application that selectively handles calls and helps you make outgoing calls. Windows When troubleshooting. These traces are usually reserved for Cisco development engineering use.

This chapter covers the following topics: In addition. SDL traces describe the events occurring in the CallManager software at a code level. CCEmail— Details the third-party alerting tool that can be used in conjunction with PerfMon to configure alerts for the performance counters. Later chapters demonstrate the use of these tools in different scenarios you might encounter.

Depending on the problem you encounter and your particular skill set. If you are strong in IP. Event Viewer— Briefly explains the function of another built-in Windows tool that plays a key role in troubleshooting CallManager. CCM trace files provide information about call processing events and all messages exchanged between Skinny.

This ensures consistent timestamps regardless of which device you are looking at. CallManager Serviceability. The various information elements are decoded for ease of reading. Cisco Bug Toolkit formerly Bug Navigator — Describes the web-based tool that allows you to find known bugs based on software version.

If a problem occurs. Sniffer traces— Discusses when and why to use a network packet-capture tool. When you synchronize the time on all involved servers. When trace files are collected for analysis. As endpoints such as IP phones and voice gateways call each other.

The resulting matrix shows when each bug was integrated or fixed if applicable. The Enhanced Q. Signaling also occurs between the respective CallManagers of each endpoint device. It also makes the troubleshooting process more involved because you have to collect traces from all participating servers to see the full picture of what happened.

Voice Codec Bandwidth Calculator— Describes how to use the Voice Codec Bandwidth Calculator to determine the bandwidth used by different codecs with various voice protocols over different media. This distributed architecture creates a highly available and scalable system.

Time Synchronization Time synchronization is simply making sure that all the participating CallManager servers and network devices have the same exact time. Time synchronization is critical. Also points you to a section in the "Introduction" with a recommended reading list. In addition to CallManager servers.

A large CallManager cluster can have eight or more separate servers. To synchronize with a remote Time Server. Step 3. This file contains the list of time servers that CallManager will synchronize with.

Synchronize the clock by using one of the following commands from a command prompt. You can configure CallManager to point to specific time servers see Example Synchronizing Time Manually on CallManager Servers Use the following steps to manually configure time synchronization. Step 4. Sample ntp. Verify that the NetworkTimeProtocol service is configured to launch automatically upon startup. Use the following steps to configure the CallManager server to automatically synchronize—and stay synchronized—with a Time Server.

Configure the C: Expand the Services and Applications section. Allow several minutes for the update to take place. Double-click the NetworkTimeProtocol service. Right-click My Computer and select Manage. Example For Cisco IOS devices.

To enable NTP. If you do not have a device on the network running NTP. First you must configure the time zone. If you are not making the device an NTP master.

Once your time zone is configured properly. You should configure the IOS device to take daylight savings time into account as well if you live in a time zone that observes daylight savings time. You can do this by configuring the command ntp master on the IOS device. CCM traces are usually the first place to look when troubleshooting most problems.

It is not a substitute for learning how to read the CCM trace. Then enable the NTP client using the command set ntp client enable. After you learn a few tricks. You can analyze problems related to device registration. By the end of this section. Once you have the time zone configured properly. Future chapters continue to show CCM trace examples as you learn how to troubleshoot more problems.

Later in this chapter you also learn about the Q. Setting the Appropriate Trace Level and Flags CallManager allows you to select from a variety of different options that adjust which events are logged to the CCM trace files. When you are beginning to learn how to read CCM trace files. For this reason. Figure shows the top half of the Trace Configuration page. If you know exactly what you are looking for. If you configure the trace to collect too much information. As you learn which trace settings are required for specific problems.

Unfortunately you can never predict what problems will come up. See the later "Q. You generally want to keep the same level of tracing enabled on all the servers in the cluster unless you are absolutely certain the problem is isolated to a single server in the cluster.

CallManager Serviceability provides six different levels. First thing to note is the Trace On checkbox. This must be selected for any of the trace settings to be available. The Apply to All Nodes checkbox allows you to apply the specified trace settings to all CallManager servers in the cluster. Because of the distributed nature of CallManager. The Trace Filter Settings area allows you to specify the exact parameters of your trace. This is useful when you are troubleshooting a problem that is occurring on more than one CallManager server.

In release 3. This level includes nearly everything that is included in Detailed with the exception of KeepAlives. Special— Provides traces for all informational. Most are reasonably self-explanatory. Arbitrary— Provides low-level debug traces. It can cause performance degradation on CallManager.

State Transition— Provides traces for call processing events or normal events traced for the subsystem signaling layers. Use this level to trace conference bridge statuses such as. Activates a trace showing H. This level is best suited for debugging difficult problems.

For CallManager releases 3. Activates a Layer 3 trace of Q. All system and device initialization traces are at this level. Activates a trace of miscellaneous devices. If you are not troubleshooting problems related to missed KeepAlives.

Activates a trace of the conference bridges. Table describes the trace fields to choose from. Other trace levels are provided. Significant— Provides traces for media layer events. CallManager uses asynchronous tracing to reduce the impact that trace file generation has on call processing.

Enable DTActivates the logging of events related to the legacy gateways. Activates a trace of phone devices. Trace Enable Forward and Miscellaneous Trace Activates a trace of call forwarding and all subsystems not covered by another checkbox. Activates a trace of media termination point devices and transcoders. Enables tracing of call detail record CDR processing. When that checkbox is selected. Tracing based on specific devices is very useful when you know which devices are involved in the problem.

The non-device option is a catch-all. Figure shows the bottom half of the Trace Configuration page.

Troubleshooting Cisco IP Telephony Full | Voice Over Ip | Osi Model

Description Figure Use a filename that exists on the computer running the trace. The new trace settings take effect immediately when you click Update.

For traces to be logged to a file. With Trace Analysis. For each server on which you're running trace. Click Update to save your settings. If you did. Do not use a filename that exists on another computer.

We're going to skip discussion of the Enable XML Formatted Output for "Trace Analysis" checkbox for a moment to finish the fields related to file settings. The trace output is then sent to the path specified in the File Name field. One downside to this. Depending on the time period you specified.

That way. The Trace Configuration page in CallManager Serviceability validates the filename and ensures that it has a. It is a good practice to have each server use a filename format that has the server name in it. The Enable Debug Output String checkbox sends debugging information to a Microsoft development tool useful only to Cisco development engineers.

In most cases. Once you've specified the trace settings. You should never need to or be asked to enable this checkbox. The default of for the Maximum No. We don't recommend going above At these average sizes. This could be servera-ccm. Assuming you don't go above The downside. If you do not enable XML-formatted output. The Maximum No. Set the service you want to retrieve the trace files for such as Cisco CallManager. If you choose to view the textbased output in a new window.

We generally recommend that you do not use the trace collection utility for anything other than very small traces on a system that is not busy. If you specify a device name. Choose ALL or specify the device's name. Manually copy the traces off the server s in question and analyze them offline. Choose ALL. In this dialog box. It is usually quicker and easier to manually collect the traces yourself using either Terminal Services or VNC discussed later in this chapter to access the CallManager server and copy the files to another machine for analysis.

Once the trace has been gathered. Select this checkbox if you want to include the number that correlates traces with each other in the trace output. Alarm shows only specific messages that meet the criteria of being an information Alarm message. Figure shows a trace file formatted as XML output and filtered to display only one CallManager host.

Date and Time. Select this checkbox if you want to include the CallManager node IP address or host name in the trace output. Select this checkbox if you want to include the trace type in the trace output. Select this checkbox if you want to include the device name in the trace output. Select this checkbox if you want to include the date and time of each event listed in the trace output.

Select this checkbox if you want to include a description of what the trace found in the trace output. CM Node. Source IP. Select this checkbox if you want to include the directory numbers DNs and other service-specific information in the trace output. Select this checkbox if you want to include the device's source IP address in the trace output. You should learn how to read the CCM traces directly from the text files to help you troubleshoot problems more quickly and accurately.

You can click the Back to Selection link to return to the Trace Analysis dialog and filter based on different criteria. Reading CCM Traces This section shows you a few call flow examples and highlights key information in each example that helps you understand the CCM trace. For that reason. The following section describes how to read a text-based CCM trace. For brevity. So the same trace is shown in this book as StationInit: As you can see. We recommend that you use the default trace settings for CCM traces except you set the trace level to either Arbitrary or Detailed.

The header portion of the trace line just specifies the date and time when the trace event was generated and which trace file you are looking at. The IP address is In other words. This includes the Cisco 79xx family of IP phones.

Appendix C. A Skinny station is any endpoint that uses the Skinny protocol to communicate with CallManager. Phone A calls Phone B. Phone A goes off-hook and you see the following line in the CCM trace: IP phones use the Skinny protocol to communicate with CallManager. You can determine the IP address by working backwards: For a CCM trace file.