Brief Detour Service Level Relationships Between Layered Communication Services

Services such as voice generally rely on services provided at lower layers, such as datatransmission services. Therefore, there are dependencies between the levels of service that are experienced at different layers. For example, there could be a relationship between the quality of the voice service that is experienced and the link or routing capacity at a lower layer If the capacity is too low, voice packets will be lost and the quality of the voice will suffer. However, to average users, it...

Simple Modeling Example

Imagine that you are tasked with defining a simple management information model for a device. All you are interested in managing is some basic system information about the device, such as the name of the device, where it is located, who the contact is, how long it has been running, and its TCP connections. The resulting models are graphically depicted in the following figures. All three represent the same underlying domain, but each is based on a different type...

Adapting Integration Approach and Network Provider Organization

One aspect that will likely influence a network provider's approach to management integration is how its operations organization is set up. The larger the network that needs to be managed, the more important it becomes how the organization that is responsible for managing the network is structured. After all, the management organization itself is an important part of how networks are managed hence, integrated management does not stop at the technical infrastructure but needs to take the...

Advanced Alarm Management Functions

Beyond those basic alarm management functions, in any network of meaningful size, additional functions to manage alarms are required. Some of those functions provide network managers with greater flexibility in processing alarms. For example, an alarm-forwarding function might send alarms to the pager of an operator to allow for an automatic dispatch, much as a home intrusion detection system automatically calls the local police station. Another function allows network operators to acknowledge...

Auditing Discovery and Autodiscovery

Being able to configure your network is important, but not enough. You need to also be able to query the network to find out what actually has been configured you need a read in addition to the write. This is referred to as auditing. Many reasons exist for auditing devices in the network. For example, you might want to verify that the configuration of the network is indeed what you expect it to be. You might want to see if configuration commands that you sent down indeed took. Without this...

Basic Management Ingredients Revisited

Now that the notion of real resources and the distinction between the network device and the management agent is in place, we can briefly revisit our original picture of the basic management components to include that distinction. Figure 3-6 refines Figure 3-1. At the most basic level, there are really only two components, depicted at the top and at the bottom the network provider's operational support organization and the real world that it wants to manage. However, technical means are...

Benefits of Integrated Management

Having management that is integrated as opposed to management that is based on a piecemeal approach that consists of multiple management islands is important for many reasons that include the following It helps ensure that management tasks do not fall through the cracks. Management tasks that are supported by a holistic, integrated operational support environment do not need to rely as much on manual procedures and leave little to chance, compared to management tasks that are not supported by...

Build Complexity

We first explore the aspect of build complexity that is, the complexity of scaling management application development. Imagine for a moment that you were tasked to build a simple service provisioning application for example, to provision digital subscriber line (DSL) service as explained in Chapter 7, Management Communication Patterns Rules of Conversation. We assume that you have a single type of DSL access multiplexer (DSLAM) to deal with and a single type of aggregation router. Provisioning...

C

Call detail records (CDRs), 153, 306 BHCA, 3S0 completion rates, 3S0 costs, 420 management information (MIBs), 1lS-1ll CDRs (call detail records), 153, 306 Central Offices, 97 chain of command, top-down solution design, 363 metaschemata, 1S4 MIBs, 1S2 clear roles, 125 clearing alarms, 136 CLI (command-line interface), S1 overview of, 261-26S protocols, 26S-26l Show command, 109 close-session operation, Netconf, 2S3 CMIP (Common Management Information audit trails, 96 automated snapshots, 223...

Chains of Command

Another related issue concerns how components in the solution interact with the managed devices in the network. When following a TMN-like hierarchy, upper-layer systems that need to interact with the network, such as to provision a service, instruct lower-layer systems to carry out requests, such as to configure a port on a device, until finally requests reach the network element. The responses are propagated back up accordingly. A major advantage of this architecture is that a clear hierarchy...

Challenges from Conflicting Software Architecture Goals

Another difficulty arises from the fact that different management functions can impose conflicting requirements on the software architecture that the integrated management system is to be based on. At times, those requirements could be difficult, if not impossible, to reconcile in a single system. This point is a little more subtle than the previous ones yet is just as important, so let us elaborate further on it. In Chapter 1, we mentioned that one of the challenges in building management...

Chapter

Both the provider and the customer of a service should be concerned with service level monitoring. Why is this so Answer The customer might not know for sure whether he is indeed getting the service level that was agreed to. 2. Assume that you are a service customer and are about to enter an SLA with a service provider to provide you with video phone service across your enterprise network. Can you think of some service level parameters that you might want to include in service level objectives...

Chapter Review

Explain the term network management in one sentence. 2. We used a patient in intensive care as one analogy to explain network management. Can you think of areas in network management that this analogy does not capture 3. Can you think of other areas in which you would expect analogies to network management to apply 4. Give two examples of how network management can help an enterprise IT department save money. 5. Give two examples of how network management can help a service provider increase...

Conferences and Workshops

Network management conferences and workshops constitute the best source of information about the current forefront of network management technology. There are several noteworthy conference series for each, proceedings are published that contain a wealth of papers describing current research projects. (Proceedings are essentially books whose chapters consist of papers that are also presented in talks at the event.) By nature, they focus mainly on the academic and research-oriented crowd. Those...

Configuration Change Events

Maintaining an accurate database of current device and network configuration is critical to many applications. As explained in the previous chapter, many applications cache configuration information of devices for efficiency. Configuration-change events communicate the fact that a configuration change has taken effect at the device. Processing configuration-change events is an important and efficient technique to prevent the cache from going stale. Of course, the application that initiated the...

D

Common platform infrastructure, 356 configuration-change events, 239-240 costs, 416 craft terminals, 64 integration, 366 management systems, 85 MIBs, 177-178 data-collection capabilities, 414 DDOS (distributed denial-of-service) attacks, 160 decommissioning networks, 118 decomposing service-level parameters, of alarms, 139-140 of events, 305 deficits, SNMP, 202 defining management, 5-10 MIBs, 180-181 models, 110 object types, 195 SLAs, 382-388 syslog protocol extensions, 271 degradation,...

Definition of Service Level Objectives

When deciding which service level objectives to include, you need to make a number of considerations Determine which service level parameters and which service level objectives are really critical to your needs. To identify candidate service level parameters, remember that they must meet the criteria that we just discussed in the previous section of being significant, relevant, and measurable. Furthermore, you need to think about what target levels are really required. Of course, everybody...

Deployment Aspects

In addition to distributing the processing task, management hierarchies can reduce requirements for management communication bandwidth. With a management hierarchy, it might no longer be necessary to deploy all management functionality centrally in a NOC. Instead, it is possible to deploy subordinate management systems geographically close to the equipment that they are supposed to help manage for example, a particular branch location of an enterprise. This can help make more efficient use of...

Device Managers and Craft Terminals

Craft terminals, sometimes also referred to as device managers (not to be confused with element managers, discussed shortly), provide a user-friendly way for humans to interact with individual network equipment. Craft terminals are used to log into equipment one device at a time, view its current status, view and possibly change its configuration settings, and trigger the equipment to execute certain actions, such as performing diagnostic self-tests and downloading new software images....

Example A SyslogtoSNMP Management Gateway

We return to the example of a management gateway that is supposed to convert syslog messages from a syslog agent into SNMP traps for an SNMP manager. One way this could be accomplished is as follows A simple syslog mediation MIB is defined, as shown in Figure 9-13. The basic idea behind this MIB is that it provides a notification type that is used to carry a syslog message. The different fields of the syslog message are conveyed through corresponding variable bindings in the SNMP trap message...

Example An SNMPtoOO Management Gateway

A second, more complicated example involves mediating between SNMP on the agent side and a management interface with an object-oriented (OO, for short) information model on the manager side. Object-oriented information models model the managed domain in terms of objects for example, a port, a connection, and a card on a device might all constitute objects, each representing a corresponding real-world counterpart. The definition of the information model specifies each kind of object that can...

Fault Diagnosis and Troubleshooting

Alarm management is a significant aspect of fault management so significant, in fact, that the two terms are often used synonymously. However, there is more to fault management than alarms. One other aspect concerns fault diagnosis and troubleshooting. Network diagnosis is conceptually not much different from medical diagnosis. The difference, of course, is the type of patient. To reach a medical diagnosis for a set of symptoms (for example, a rash), the doctor might want to take a look at...

How This Book Is Organized

This book is intended to be read cover to cover because later chapters build on concepts and principles that earlier chapters introduce. Nevertheless, many chapters are relatively self-contained, which should make it fairly easy to move between chapters. The chapters of this book are grouped into four parts Part I, Network Management An Overview, provides an overview of what network management is about and why it is relevant. It also conveys an informal understanding of the functions, tools,...

Inside the Network Operations Center

One important aspect of the management support organization concerns where it is physically located. This might not be a consideration for a small business running a few routers in one or two locations, but it does matter for a service provider with a global presence, interconnecting thousands of sites. The place from which large networks is managed is generally termed the Network Operations Center (NOC). From here, the bulk of management-related activities is carried out, from monitoring the...

Management Hierarchies

As indicated earlier, a single system is generally not sufficient to manage a network. Instead, the work needs to be distributed. Let's look at a real-life analogy. Consider a person who owns and runs a small business. As the business grows, the business owner might no longer be able to manage the business single-handedly. So she gets help. She still wants to be in charge of running the overall business, but she distributes certain tasks across her people. Eventually, she starts building an...

Management Information MOs MIBs and Real Resources

In general, many aspects of a network device (such as a router or a switch) are important for its management. For example, the device has a network address, it is of a certain type, and it has software installed of a certain revision. If the device is a router, it might be running a variety of routing protocols. The device might consist of a rack-mountable chassis with a fan for cooling, a central processor module, and a set of expansion slots. Furthermore, the device might contain a set of...

Management Layer Its a Device No Its a Service No Its a Business

Network management is not just a multidimensional but also a multilayered problem space. At one layer, the concern is with managing individual devices. For example, each device must have the right software patch installed and must be monitored to make sure that it is running properly. These tasks apply regardless of what devices are actually used for in the network for example, whether they route IP traffic in the core of the network, whether they connect end users to the network, or whether...

Management Platforms

Management platforms are general-purpose management applications that are used to manage networks. The functionality of management platforms is generally comparable to that of element managers. However, management platforms are typically designed to be vendor independent, offering device support for equipment of multiple vendors. Typically, the primary task of a management platform is to monitor the network to make sure it is functioning properly. Therefore, it was also the main tool that Chris...

Management Subject What Were Managing

As mentioned and depicted in Chapter 1, Setting the Stage, in Figure 1-4, there are different kinds of networked systems that require management. Network management is often categorized into different subdisciplines to reflect that distinction Network management, in a narrower sense, deals with the management of communication networks and the resources in the network that are required to establish end-to-end communications. For example, this includes the routers and switches in a network, or...

Management System and Manager Role

The terms manager and management system are often used synonymously. Strictly speaking, this is not quite correct, and, in general, care should be taken to distinguish a manager (the role) from a management system (the application). This is because, for various reasons, it might make sense for the same system to play both agent and manager roles. For example, one network element might act as a management proxy to another. In this case, the network element plays the agent role in interacting...

Management Transactions

Sometimes management applications would like not having to issue a request response pair for each configuration operation or management action, but instead be able to group several commands together and have them execute together as one unit. This is often the case when services need to be provisioned over a network. Consider the simplified example of a service provider that wants to provision a digital subscriber line (DSL) service, as Figure 7-14 illustrates. Figure 7-14 Provisioning a DSL...

Manager Initiated Interactions Request and Response

Let us now turn to the way in which actual interactions between managers and agents, or management applications and managed devices, take place. Here we take a look at how management operations are used to conduct effective management communications. We start with interactions that are initiated by the manager. Interactions that are initiated by the agent are the subject of the next section. The patterns of interactions between managers and agents that are described are largely independent of...

Managing the Management

The management support organization ultimately is responsible for making sure that the network is being run effectively and efficiently. It needs to perform such tasks as were presented in the previous chapter, including but not limited to these Monitoring the network for failures Diagnosing failures and communication outages if they occur, and planning and carrying out repairs Provisioning new services, and adding and removing users to and from the network Keeping an eye on performance of the...

Mediation of Management Information at the Semantic Level

Mediating management information without the limitations of the syntactic transformation approaches requires a semantic understanding of the management information involved. This means that custom translation rules need to be crafted, mapping the mediated management information to the target information model. For example, in the case of mediation from syslog (agent) to SNMP (manager), it would be necessary to determine which specific syslog messages should trigger which specific SNMP traps,...

Modeling Management Information

We mentioned several times that the management information that an agent exposes across its management interface constitutes an abstraction of the managed device. This abstraction is based on a model of the real world, and information in the MIB is an instantiation of this model. Because it is used for management purposes, the model includes aspects that are relevant for management and omits aspects of the real world that are not it abstracts them away. For example, the software revision that...

Netconf Operations

At this point, we can finally turn toward the guts of Netconf, the operations layer. As mentioned, Netconf is built around the notion that management information in general, and specifically configuration information, can be thought of as being contained in a conceptual datastore. In the case of configuration information, this datastore is a configuration file, in short referred to as config. Not coincidentally, this resembles how things are handled using CLI on a router. Different examples of...

On the Job with a Network Manager

This chapter presents a number of scenarios to give an impression of the types of activities that are performed by people who run networks for a living. We refer to them collectively as network managers, although they perform a wide variety of functions that have more specialized job titles. In fact, strangely enough, the term network manager is rarely used for the people involved in managing networks. Instead, terms such as network operator, network administrator, network planner, craft...

Organization and Operations Challenges

Small networks, such as those deployed by small businesses, might be run by a single person or network administrator as a part-time job. In those cases, how to run the network isn't much of an organizational issue The network administrator is in charge, and if problems arise that the network administrator cannot solve (or if the network administrator is out sick), customer support by a third party, by the equipment vendor, or by a consultant is only a phone call away. In addition, many...

Remote Operations

The Remote Operations layer offers three distinct functions that complement and perform important services for the Management Operations layer on top association control, remote operations support (in Figure 7-1, this is depicted a little simplified as RPC for remote-procedure call), and encoding of payload data. In many cases, those functions are provided not by a dedicated protocol, but by the management protocol that also provides the functionality of the Management Operations layer on top....

Scale

Parents of young children should be able to relate to the following scenario Try babysitting a toddler for a few hours. When she is hungry, she requires something to eat you should make sure she drinks enough so she doesn't get dehydrated perhaps she needs her diaper changed once in a while and a little entertainment to keep her occupied, so you read her a story and offer her some Legos. Doable. Now imagine a toddler birthday, with 20 toddlers and no one there to help you, and things become a...

Security of Management

Security of management deals with ensuring that management operations themselves are secure. A big part of this concerns ensuring that access to management is restricted to authorized users. For example, access to the management interfaces of the devices in the network needs to be secured to prevent unauthorized changes to network configurations. Also, the management network needs to be secured to prevent disruption to management traffic. In addition, access to the management applications...

Service Level Statistics Its Fingerpointin Good

In addition to monitoring and ensuring service levels, it is also important to keep a record of the service levels that are experienced over the course of time. Why Because such a record can prove of tremendous value if a dispute between service provider and customer occurs. That's why historical service level statistics are often also referred to as fingerpointing data. If a customer complains to a service provider about service level violations, the service provider might simply admit to it,...

Service Provisioning Systems

Service provisioning systems facilitate the deployment of services over a network, such as Digital Subscriber Line (DSL) or telephone service for residential customers of large service providers. Service provisioning systems translate requests to turn on or to remove a service into a series of steps and configurations that are then driven into the network. Service provisioning systems are typically very complex applications that can be found only in operational support environments of large...

Set Request

A manager uses a set request to write to a MIB that is, to set a MIB object to a particular value. The structure of the set request is exactly the same as with get and get-next, except that, in this case, the object values in the variable bindings are not set to null, but contain the values to set the respective objects to. The same restrictions related to message size apply as before. Set requests are used in several ways. The first, most obvious, and most common use of set requests is to...

SNMPv2 SNMPv2c

As SNMPv1 gained widespread support, it turned out that certain aspects about it were perhaps a little too simple. SNMPv1 is notoriously inefficient at retrieving large amounts of management information, knowing no concept of scoping or bulk requests. It offers only minimal security, making it vulnerable to security threats, which effectively prevents SNMPv1 from being used to change the configuration of managed devices in many cases, the risk of compromising the integrity of the network is...

Solution Philosophy and Challenges

Contrary to applications that run on a management platform and rely on the management platform's infrastructure, each of the systems and applications that is integrated in a solution incorporates sufficient infrastructure to be capable of running on its own. They constitute a set of loosely coupled solution components, which coordinate with each other as required to exchange data and provide an integrated management experience. Where a management platform provides off-the-shelf integration,...

The Basic Ingredients of Network Management

Chapters 1, Setting the Stage, and 2, On the Job with a Network Manager, explored what network management does, why it is important, where its challenges lie, and what kinds of activities and tools are associated with it. But what does network management, at a very basic level, really consist of First, there is, of course, the network that is to be managed, consisting of a multitude of interconnected devices that collectively shuffle data (for example, web pages, e-mails, voice packets of phone...

The Management Network

Now we understand that managers and agents refer to different roles in which management systems and network elements communicate with each other for management purposes. But how do they communicate The answer is, over a network, of course. At the end of the day, network management is just another distributed application. The different systems that need to communicate in this case just happen to involve management systems and the network elements that they manage. Managing systems and managed...

The Motivation for Service Level Agreements

In the context of this chapter, we often refer to the providers of networking services as service providers. However, as in previous chapters we do not mean this term to be restricted to traditional service providers it can include also enterprise IT departments or even organizations running a data center. Service providers run networks or data centers not just for the heck of it, but so that benefits can be derived from the services that run over it. In the case of an enterprise's IT...

The Relationship Between MIBs and Management Protocols

You will find that the term MIB is often associated with SNMP, the Simple Network Management Protocol. SNMP defines a particular communication protocol that is often used between managers and agents it is discussed in detail in Chapter 8. SNMP requires management information in a MIB to be represented according to the rules of a particular specification language, known as Structure of Management Information (SMI), introduced later in this chapter. This particular representation, not just the...

The Role of Standards

For managers and agents to interoperate, quite a few elements need to be aligned In addition to being interconnected, they need to speak the same management language that is, protocol. The manager needs to understand precisely which functions the agent supports and to interpret the results that are returned. Furthermore, manager and agent need to be on the same page concerning the management information carried in the management messages. Otherwise, to pick up on the earlier example, those...

The Service Provider Perspective

Management integration takes on a whole new level of complexity for a telecommunications service provider that has to manage literally thousands of different types of systems, devices, and applications supplied by dozens of vendors, used to provide not just one type, but a great variety of different services. In such an environment, many different tools have to be integrated that are provided by different suppliers and equipment vendors each of which might be fully integrated from their limited...

Tracking Service Level Objectives

In defining service level objectives, you need to also be clear about how those will be verified and how the underlying service level parameters will be measured. This is important not only to avoid disputes later it can also help identify any misunderstandings about what the service level parameters entail. Here are considerations to take into account Be clear about where parameters are measured. For example, if you have a service level objective for an application hosting service by a data...

Trouble Ticket Systems

Trouble ticket systems are used to track how problems in a network (such as those that are indicated by alarms) are being resolved. Note that this is different from managing the alarms themselves. Trouble ticket systems are used to capture information about problems that were observed in the network and to track the resolution of those problems. In many cases, trouble tickets are generated by users of the network who experience a problem, although they might also be created proactively by an...

Warning and Disclaimer

This book is designed to provide information about network management. Every effort has been made to make this book as complete and as accurate as possible, but no warranty or fitness is implied. The information is provided on an as is basis. The authors, Cisco Press, and Cisco Systems, Inc., shall have neither liability nor responsibility to any person or entity with respect to any loss or damages arising from the information contained in this book or from the use of the discs or programs that...

Of Pyramids and Layered Cakes

As mentioned previously, management reference models serve as conceptual frameworks for organizing different tasks and functions that are part of network management. The emphasis is on the word conceptual. In reality, reference models are, in many cases, not literally followed management systems and operational support environments can be structured in different ways that, for various reasons, do not reflect the same breakdown in functionality as suggested by any particular reference model....

How It All Relates and What It Means to You Using Your Network Management ABCs

With so many functional reference models to choose from, which one is the best Ultimately, it comes down to a matter of preference and to which model suits your network best. These reference models are, after all, virtual the way in which you actually organize your management functions may look different altogether. You can cut things diagonally, horizontally, or vertically. The result in each case should be that you have partitioned the task of managing your network into smaller chunks that...

Network Monitoring Overview

Network monitoring includes functions that allow a network provider organization to see whether the network is operating as expected, to keep track of its current state, and to visualize that state. This functionality is fundamental to being able to recognize and react to fault conditions in the network as they occur. The most important aspect of network monitoring concerns the management of alarms. Alarms are unsolicited messages from the network that indicate that some unexpected event has...

CLI Overview

Command-line interface (CLI) was conceived to make it easy for human operators and administrators to interact with networking equipment in particular, data-networking devices. It is reminiscent of the character-based command-line interfaces used with computer operating systems such as UNIX. This is actually not surprising at the end of the day, a router is nothing more than a special-purpose computer with a set of networking interfaces and a special-purpose operating system. In fact, the first...

Web Resources

The Web has many excellent sources of information. Readers are encouraged to conduct their own Internet searches. Just try a Google search on terms such as network management, command-line interface, SNMP, and service-level management. Searches on the various topics discussed in this book are sure to yield a wealth of information, as are the websites of vendors engaged in management technology. Here is a very short list of links that you may want to try. http www.simpleweb.org contains a wealth...

The Management Support Organization Noc Noc Whos There

The ingredients that we have introduced so far network elements and management agents, management systems, and management network are all that is required to make network management work from a technical perspective. However, if we really want to successfully run a network, we are not quite finished. Missing is the organization that will be responsible for running the network ultimately, the people who use all that management technology. Unless you are a small business that has few devices that...

Management by Objectives and Policy Based Management

The idea behind management by objectives is that a management system establishes certain goals for a subordinate system, and the subordinate system translates these goals into the required lower-level actions to ensure that those goals are met. This way, the upper-layer management system can focus simply on setting overall policy for management of the network that establishes the management objectives. The subordinate system handles translating the policy into actions. The subordinate system...

Integration Scope and Complexity

As you have just seen, the scope of what needs to be integrated as part of management depends on the perspective. It can be observed that as the scope of integration grows, its complexity grows in two dimensions simultaneously horizontally and vertically. Horizontally, the number of devices and network technologies that need to be integrated increases. However, with growing network size and complexity, it is no longer sufficient to merely integrate and add more systems while still providing...

Limitations of Syntactic Information Mediation

As simple as the previous examples were, they show the limitations of syntactic mediation. We turn first to the syslog-to-SNMP gateway. Of course, the gateway does allow an SNMP manager to receive syslog messages without needing to understand and parse syslog. However, the traps produced by the mediator likely look a little different than what would have been emitted by a native SNMP agent at the device. To distinguish those traps, we use the terms mediated traps and native traps, respectively....

The Equipment Vendor

Equipment vendors are primarily in the business of selling networking equipment, not network management applications. Hence, traditionally equipment vendors have shown a tendency to limit investment in management application development. In general, they have been willing to settle for the minimum management capabilities that customers would allow them to get away with. That means that generally they would provide just the level of management capabilities needed to not inhibit equipment sales....

Of Schema and Metaschema

As mentioned in the previous section, the model that underlies the management information in a MIB is specified in a MIB definition. Some people call the model the schema, reminiscent of a database schema that constitutes the definition of the database tables. The underlying real world that is being abstracted by the model is often called the domain because it constitutes the subject domain that the model is all about. In the example of the previous section, the domain of the model is that of...

Common Platform Infrastructure

Like a computer operating system, a management platform provides a set of infrastructure that management applications commonly need and that developers of applications for the platform consequently do not have to build themselves. This infrastructure goes beyond building blocks that are incorporated into application code in the form of software libraries. Instead, it involves common services that live in their own process space during runtime. Applications that need to use those services...

Netconf Datastores

Netconf picks up on the notion that the configuration information of devices can be thought of and handled as being contained in a datastore (one word, per the Netconf spelling) that can be handled like a file. In essence, a configuration datastore corresponds to a device's config file the set of configuration statements that need to be executed to bring the device into its desired configuration state. As a protocol, Netconf provides the operations that are necessary to manage those datastores....

Structure of Management Information Overview

In SMI, MIB definitions are specified as MIB modules. A MIB module generally serves a particular purpose, such as to define management information related to a device's communication interfaces or to a voice-mail server feature that is embedded on a particular type of device. Accordingly, the MIB of any particular device instantiates multiple MIB modules, each of which represents one aspect of the managed device, as Figure 6-11 illustrates. Again, the term MIB is often used synonymously with...

Instantiation in an Actual MIB

So far, we have described how a model to represent management information is defined in SMI and how different object types are identified through their OIDs. We have also mentioned that some of the object types can be instantiated once, others multiple times. But how are those instances identified during runtime in an actual MIB This is one of the stranger aspects of SNMP and a little counterintuitive at first. Object instances in a MIB are considered to be conceptually part of the same object...

Mediation Between Management Protocols

Mediation at the management protocol level involves translating management messages of one protocol into management messages of another one specifically, mapping between the management primitives. From a naive point of view, this should be relatively straightforward if the capabilities between the protocols that are involved are equivalent. In this case, it should be sufficient to essentially perform a syntactic transformation that can occur during runtime. For example, for mediation from SNMP...

The Managed Device as a Conceptual Data Store

A MIB is best thought of as a conceptual data store. Managers can retrieve management information from the MIB by directing corresponding requests at the management agent for example, using a get operation. In many cases, they can also manipulate and modify the information that is contained in the MIB for example, using a set, a create, or a delete operation. The MIB, of course, is not the same as a database. The MIB does not store information about the real world (the actual managed device) in...

Syslog Protocol

For a long time, there was no true standard that syslog messages had to follow. As mentioned, syslog originated from messages that were logged by the UNIX operating system. However, syslog has been treated just as a loose recommendation and was never rigorously specified as a standard. Consequently, over time, different variations of syslog message formats proliferated across different vendors and device types. For example, they differed on details such as the precise format that is used to...

Anticipating Problems Before They Occur

Being able to monitor current service levels is nice, but when it is detected that service level objectives are violated, the damage is already done. Therefore, it would be nice to also anticipate a problem before it occurs. This way, it might still be possible to take countermeasures to avert service level objective violations. Accordingly, what is needed is a service level forecast, not unlike a weather forecast of course, in this case, the course of forecast events can still be influenced. A...

Netconf and XML

One of the distinctive features of Netconf is the fact that it uses XML as encoding for its management operations. XML is a cornerstone of web technology it is a language that allows for the representation of information in a structured way. Obviously, a discussion of XML goes way beyond the scope of this book there is a rich set of literature for the interested reader. We just briefly discuss some of the most fundamental aspects as far as they are relevant for the remainder of the discussion....

Management Agent

To be managed, a network element must offer a management interface through which a managing system can communicate with the network element for management purposes. For example, the management interface allows the managing system to send a request to the network element. This could be, for example, a request to configure a subinterface, to retrieve statistical data about the utilization of a port, or to obtain information about the status of a connection. Likewise, the network element can send...

The Role of Standardization and Information Models

In Chapter 4, we discussed the role of standards in network management. It should be mentioned that standardization also plays a big role in integration among different components. After all, the intent of network management standards is to facilitate interoperability between different systems, which facilitates their integration. Standardization concerns not only interfaces between managing and managed devices, but also interfaces between different management systems. We discussed this in...

The Impact of the Metaschema on the Schema

In the fine arts, the media that an artist uses has a great influence over the type of artwork that results. For instance, the character of a painting is different if the artist uses water colors, oil colors, crayons, or a pencil. The difference is even more dramatic if clay is used, resulting in a sculpture instead of a drawing or a painting. Of course, each medium can be used to model the same aspect of the real world, such as a person. The resulting model is called a portrait. In network...

Chris Network Administrator for a Medium Size Business

Together with a colleague who is currently on vacation, Chris is responsible for the computer and networking infrastructure of a retail chain, RC Stores, with a headquarters and 40 branch locations. RC Stores' network (see Figure 2-4) contains close to 100 routers typically, an access router and a wireless router in the branch locations, and additional networking infrastructure in the headquarters and at the warehouse. The company has turned to a managed service provider (MSP) to...

Use of CLI as a Management Protocol

Strictly speaking, CLI is not a management protocol at all. It is a command-line interface, intended for human users who interact with the device directly, not through a management application that abstracts away the details of how the communication with the device takes place. Therefore, it is perhaps a bit unfair to point out some challenges that are associated with CLI related to uses that it was not designed to support. However, management applications are faced with the problem of how to...

Subcontracting Management Tasks

In a management hierarchy, certain management tasks are subcontracted to different systems. In effect, the subcontracted systems constitute management proxies To the subordinate system, it appears that the proxy is the manager. However, the manager proxy is really only a conduit, acting on behalf of another management system that is invisible to the managed system in the agent role. You've already seen one example of management hierarchies the specialization of applications along the different...

Layer Example

Netconf Protocol Layers

The transport protocol layer provides for the underlying communication transport. Different transports are possible and can be used for example, Secure Shell (SSH) and Block Extensible Exchange Protocol (BEEP). Those protocols are specified elsewhere and are not specific to management. What Netconf does is specify the requirements that a protocol must meet so that it can be used. (It also specifies bindings for a few transports, including the ones mentioned.) For example, the transport must...

The Third Party Application Vendor

Third-party management software application vendors fill the management application gap that equipment vendors leave open. For one, management application software developed by an equipment vendor tends to support only equipment of that particular vendor. Even if multivendor support is provided, preferential treatment is given to the vendor's own equipment, in terms of both available features and the timeline at which the support becomes available. At the same time, as stated previously, in...

Management Process and Organization Of Help Desks and Cookie Cutters

Management interoperability, management function, and management layers capture different technical aspects of network management. However, network management also involves a nontechnical dimension how to organize the management. This includes the processes that are required to ensure that the networks are run smoothly and reliably, as well as the structure of the support organization. Those nontechnical aspects are the topic of the management process and organization viewpoint. The management...

Alarm and Event Filtering

Let us first turn our attention to filtering, not just of alarms, but of events in general. To focus an operator's or a management application's attention on those events that really matter, it is important to block out as many irrelevant or less important events as possible. This is analogous to the way in which the human brain is able to deal with the massive flow of data that it is constantly exposed to, such as sounds, visual images, and sensory data. To focus, it filters out massive...

Basic Alarm Management Functions

We start the discussion of alarm management with the more basic functions collecting alarms, maintaining accurate and current lists of alarms, and visualizing alarms and network state. The most basic and, at the same time, most important task the task that everything else builds on consists of simply collecting alarms from the network and making sure that nothing important is missed. This includes receiving the alarms and storing them in memory so that they can be further processed by an...

Management by Delegation

Management by delegation involves an upper-layer management system delegating certain tasks to lower-layer systems in some cases, the managed systems themselves. This is a very common theme that can be found across the various management functional areas. In many cases, tasks suitable for delegating are routine tasks that do not require interaction with a management operator or administrator. They involve a relatively low level of intelligence but often require sifting through a large amount of...

Other Material Related to This Book

There Is No Such Place As America Stories. Delacorte Press, 1970. This book contains the short story A Table Is a Table, mentioned in Chapter 6. Brown, A., A. Keller, J. Hellerstein. A Model of Configuration Complexity and Its Application to a Change Management System. In Proceedings of the 2005IEEE IFIP International Symposium on Integrated Network Management, p. 631-644, 2005. This paper contains an analysis of operational complexity that is associated with configuration...

Pat A Network Operator for a Global Service Provider

Pat works as a network operator at the Network Operations Center (NOC) of a global service provider that we shall call GSP. She and her group are responsible for monitoring both the global backbone network and the access network, which, in essence, constitutes the customer on-ramp to GSP's network. This is a big responsibility. Several terabytes of data move over GSP's backbone daily, connecting several million end customers as well as a significant percentage of global Fortune 500...

Backup and Restore

If you are a PC user (and, chances are, you are), you are aware of the importance of protecting your data by performing regular backups. You never know when your hard disk will bite the dust or whether your PC will contract a virus that could destroy your PC's file system. Having a backup of your data in such cases enables you to recover. With backups in place, contracting a virus or needing to replace your hard drive is still annoying, but it beats by far simply being wiped out. Likewise, the...

Standards and Industry Recommendations

Schoenwaelder, J. Case, M. Rose, S. Waldbusser. Conformance Statements for SMIv2. IETF RFC 2580, April 1999. McCloghrie, K., D. Perkins, J. Schoenwaelder, J. Case, M. Rose, S. Waldbusser. Structure of Management Information Version 2 (SMIv2). IETF RFC 2578, April 1999. McCloghrie, K., D. Perkins, J. Schoenwaelder, J. Case, M. Rose, S. Waldbusser. Textual Conventions for SMIv2. IETF RFC 2579, April 1999. These above three references specify SMIv2. RFC 2578 is the...

Decomposing Service Level Parameters

In many cases, a service is realized by a combination of multiple pieces. We already encountered the example of a hosted application service depicted in Figure 11-4. In this example, an application service provider hosts a set of applications in a data center that is accessible over a network. Multiple pieces are involved in providing this service A networking piece is used to provide connectivity to the host from the edge of the data center, with traffic possibly traversing a content switch...

Get Next Request

A manager uses a get-next request to retrieve management information from an agent, just as with a get request. However, contrary to an ordinary get request, the OIDs in the variable bindings do not specify the objects that are to be retrieved directly. Instead, for each OID specified in the request, the agent is requested to return the object with the OID that comes in lexicographical order right after that OID. An OID supplied with a get-next request can be but does not have to be an OID of...

An Example MIB2

Let us now consider an excerpt from MIB-2. For brevity, portions of the definition are omitted. The symbol is used to indicate where information is omitted within the definition excerpts. We start by taking a look at the header of the MIB module. mib-2 OBJECT IDENTIFIER mgmt 1 This definition establishes mib-2 as a new node underneath a supernode called mgmt inside the Internet object identifier tree. (mgmt is imported from another standard it identifies the subnode that is reserved for...

Management Styles

We have now seen that management tasks can be distributed across many systems and that distributed management functionality can be deployed in different ways. The remaining question is how to best make use of these capabilities. A lot of business administration literature discusses different business management approaches. Actually, scaling the task to manage networks is not all that different from scaling the task to manage other business functions, such as managing people in an organization....

About the Author

Alexander Clemm, Ph.D. is a Senior Architect with Cisco Systems. He has been involved with integrated management of networked systems and services since 1990. Alex has provided technical leadership for many network management development and engineering efforts from original conception to delivery to the customer. They include management instrumentation of network devices, turnkey management solutions for packet telephony and managed services, and management systems for Voice over IP...

Stateful Mediation

Ideally, management mediation follows a simple pattern The management gateway receives a request message from a manager and translates it into an equivalent request message for the agent. When a response or event message is received, the gateway translates it into an equivalent response or event message that it sends back to the manager. The pattern is very straightforward and clean management mediation involves not much more than transforming a message that the gateway can forget about after...

Networking for Management

One way in which network elements can be connected to a management system is through the network element's management port. For most routers, this is a serial interface. It is possible to connect a terminal, such as a notebook computer, directly to that serial interface using a serial cable, as illustrated in Figure 3-10. The terminal thus connected to a network device is typically referred to as a craft terminal, in reference to the craft technician who typically uses it. The craft terminal...

Function Viewpoint What Can I Do for You Today

The function viewpoint establishes what functions are supported that is, what services a manager can expect from an agent. This includes the type of requests that a manager can make and that the agent supports. It also includes capabilities that an agent has to send event messages to notify a manager of certain event occurrences. At this point, we've covered the need to establish connectivity, as well as the need for rules for the exchange of management messages. Some additional aspects that...

Sandy Administrator and Planner in an Internet Data Center

Sandy works in the Internet Data Center for a global Fortune 500 company, F500, Inc. The data center is at the center of the company's intranet, extranet, and Internet presence It hosts the company's external website, which provides company and product information and connects customers to the online ordering system. More important, it is host to all the company's crucial business data its product documents and specifications, its customer data, and its supplier data. In addition,...