HP OpenView System Administration Handbook [Electronic resources] : Network Node Manager, Customer Views, Service Information Portal, HP OpenView Operations نسخه متنی

اینجــــا یک کتابخانه دیجیتالی است

با بیش از 100000 منبع الکترونیکی رایگان به زبان فارسی ، عربی و انگلیسی

HP OpenView System Administration Handbook [Electronic resources] : Network Node Manager, Customer Views, Service Information Portal, HP OpenView Operations - نسخه متنی

Tammy Zitello

| نمايش فراداده ، افزودن یک نقد و بررسی
افزودن به کتابخانه شخصی
ارسال به دوستان
جستجو در متن کتاب
بیشتر
تنظیمات قلم

فونت

اندازه قلم

+ - پیش فرض

حالت نمایش

روز نیمروز شب
جستجو در لغت نامه
بیشتر
لیست موضوعات
توضیحات
افزودن یادداشت جدید


1.2 WHAT NETWORK NODE MANAGER PROVIDES


Before getting into specifics, let's look at some concepts and terminology that will be used throughout this book. Then we will delve into the requirements of a corporation in search of a network management solution.

1.2.1 Continuous Status Monitoring of Managed Nodes


A managed node is a device that responds to ICMP echo requests (ping) and is actively being polled by NNM. This device may be discovered automatically by NNM or may be added manually. Optimally, the device is running an SNMP agent.

1.2.2 User and Administrator Interfaces


There are two types of NNM users: administrators and operators. An administrator manages the NNM installation and customization. An operator uses NNM to manage the network for the organization. An

administrator accesses NNM from the native interface and requires full control of customization. An

operator can access NNM using the native interface in read-only mode, which prevents permanent customization. The operator can also access NNM from the web-based Network Presenter, which provides limited functionality and read-only capabilities. Another web access method, Dynamic Views, allows full access, provided that it is executed from the NNM station that is running the NNM processes.

1.2.3 Integration with Trouble Ticketing Systems


Many corporations today have some means of tracking problems that occur within the IT infrastructure. This includes problems involving computer systems, environmental systems, network devices, and peripheral devices such as printers. These systems are commonly known as trouble ticketing systems because they use the paradigm of a ticket to track the problem from detection to correction. Depending on the size of the organization, these systems can become elaborate and include such capabilities as dispatching a person with equipment to effect a physical repair. NNM can tie into the trouble ticketing system. In this way, the system can be made aware of external detection and correction events.

1.2.4 Third-Party Applications


While NNM does not offer everything required for enterprise management, it is easily customizable and "open" to third-party applications. You can make customizations to NNM without modifying and compiling source code. NNM provides the ability to customize the

Graphical User Interface (GUI) from within the GUI.

Many vendors develop product specific applications that integrate seamlessly into the OpenView framework. For example, CiscoWorks allows you to manage Cisco® networking devices and can be integrated into the NNM GUI. This application provides the ability to configure and manage Cisco devices from NNM.

The

HP OpenView Integrated Solutions Catalog , available via http://openview.hp.com/sso/isv/search, provides a current listing of third-party software and device vendors that have integrated their products with HP OpenView solutions.

1.2.5 The Need for Network Management


Our rapidly growing business, Sealey Inc., provides managed services to its customers, including web hosting, systems monitoring, and notification. Although the company is based in Atlanta, there are co-locations across seven states. There are system administrators and router managers in each of the seven states who will need access to the data in their locations. The administrators typically use Windows systems and the router managers use a combination of HP-UX and Solaris systems, providing connections to their local network-monitoring software. The router folks expect to be able to use CiscoWorks 2000 to configure their routers and would like for the network management solution to integrate with these applications. The company also uses Service Desk as their trouble ticketing system and would like their network management application to automatically open a trouble ticket when certain events occur.

We need a network management solution that will allow Sealey Inc. to monitor customer networks proactively. For example, when a customer's Oracle database server becomes unavailable, the customer loses millions of dollars per hour. Most of the customer's servers are fully redundant with mirrored disk drives, dual power supplies, and contingency servers. However, the customers not only want the appropriate party to be notified when any components fail, they also want the ability to access this management information. The customers want to automatically page their engineers when necessary. Sealey Inc. needs a network management solution that will provide the following capabilities:

  • Notification

    before a customer network device becomes overloaded

  • Ability to monitor performance of database and application servers

  • Capability for NNM operators to log on to the servers easily

  • Access to network maps from Windows, Solaris, and HP-UX systems

  • Access to network maps from a web browser

  • HTML-based reports for the management staff

  • Historical trending for network devices

  • Ability to monitor up to 10,000 customer devices from a central location

  • Ability to automatically open Service Desk incidents

  • Integration with CiscoWorks

  • Restricted customer access to network information

  • Password-protected customer portal for remote access


The customers want their database and application servers to be monitored. Sealey Inc. will monitor the necessary MIB variables to keep track of memory and CPU utilization. Telamon® Telalert can be used to page the database administrators when problems are encountered on the servers. If network connectivity between application and database servers fails, a local network operator will need to be paged. Sealey Inc. can define thresholds for MIB variables and monitor these values in order to notify a network operator

before the network becomes overly congested.

Sealey Inc. can give operators easy access to the database and application servers by creating buttons that will launch a telnet session to the server. NNM provides both a native (Motif) interface and a web-based interface. Anyone accessing NNM as an administrator can choose between the native ovw interface and the web-based Dynamic Views. The web-based version allows flexibility to run the application from any operating system in any location. Certain web-based tools provide read-only access and limited functionality, but can be used by operators to monitor the network. NNM also has the ability to generate and view web-based reports of devices and availability. See Figure 1-2 for the web-based General Inventory report available with NNM.

Figure 1-2. NNM provides web access to data such as General Inventory.

[View full size image]

By default, NNM does not collect MIB variable data. Data collection is easily configured and can be used for historical trending. Because Sealey Inc. manages customers that are growing rapidly, they need the network administrators to closely monitor the network activity to determine whether the networking devices are sufficient to handle the customers' growth. Data collection will need to be configured for select network devices. NNM can collect the data, graph the collected data, and configure thresholds to alarm on the data. Graphs can be used to spot trends in network activity and determine when network devices need to be upgraded. For example, the MIB variable ifInOctets measures activity coming

into a network interface. You can collect, monitor thresholds, and graph MIB variables, such as ifInOctets, as shown Figure 1-3.

Figure 1-3. NNM provides the ability to graph MIB variables. The MIB variable, ifInOctets, shown on this graph tracks interface traffic into the interface.

[View full size image]

In addition to spotting trends, the staff can also configure thresholds for automatic notification. For example, thresholds can be configured for network interface traffic. When a threshold value is exceeded, someone on the network management staff can be paged. Telamon® Telalert can be integrated with NNM to provide automatic paging capability.

NNM can perform other automatic actions such as sending email. Sending an email can be a simple Unix script that passes the node name, the threshold value, and the sampled value. Another automatic action could be to open a Service Desk incident.

Because Node Manager has a practical limit of monitoring 2,000 nodes per NNM license, Sealey Inc. would need at least five NNM stations in order to manage up to 10,000 devices. This value is based on the typical polling interval used by NNM and depends on other factors such as network reliability and the amount of data being collected. The five NNM stations can be distributed throughout the network reporting to a central NNM station.


    / 276