Transcript need

1
Chapter 11
MAINTAINING
A NETWORK
INFRASTRUCTURE
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
2
OVERVIEW
 Use System Monitor to view computer performance
statistics in real time.
 Use Performance Logs And Alerts to capture counter
information to log files.
 Use Network Monitor to capture and analyze
network traffic.
 Describe the elements of a backup strategy and
their functions.
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
3
OVERVIEW (continued)
 Understand the difference between full,
incremental, and differential backup jobs.
 List the steps involved in creating a backup plan.
 Configure Windows Server 2003 Remote
Assistance.
 List the security features protecting computers that
use Remote Assistance.
 Configure Windows Server 2003 Remote Desktops.
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
USING THE PERFORMANCE CONSOLE
 System Monitor
 Displays real-time performance data collected from
performance counters
 Performance Logs And Alerts
 Records data from performance counters over a
period of time and executes specific actions when
counters reach a certain value
4
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
USING SYSTEM MONITOR
5
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
MODIFYING THE GRAPH VIEW
6
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
USING THE HISTOGRAM AND REPORT VIEWS
7
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
ADDING COUNTERS
8
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
CREATING AN EFFECTIVE GRAPH DISPLAY
 Limit the number of counters.
 Modify the counter display properties.
 Choose counters with comparable values.
 Adjust the vertical scale.
9
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
10
USING PERFORMANCE LOGS AND ALERTS
 Counter log Captures statistics for specific counters
to a log file
 Trace log Records information about system
applications when certain events occur
 Alert Performs an action when the counter reaches
a specified value
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
CREATING A COUNTER LOG
11
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
CREATING A TRACE LOG
12
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
VIEWING A COUNTER LOG
13
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
CREATING ALERTS
14
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
15
NETWORK TRAFFIC MONITORING
 The Performance console can measure all network
traffic on a specific interface or on a per-protocol
basis for the following:
 IPv4
 ICMP
 TCPv4
 UDPv4
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
16
LOCATING SYSTEM BOTTLENECKS
 A bottleneck is a component that is not providing an
acceptable level of performance compared with the
other components in the system.
 Bottlenecks can appear for a variety of reasons,
including:
 Increased server load
 Hardware failure
 Changed server roles
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
17
MONITORING PROCESSOR PERFORMANCE
 Processor: % Processor Time—Should be <85%
 System: Processor Queue Length—Should be <10
 Server Work Queues: Queue Length—Should be <4
 Processor: Interrupts/Sec—Will vary, depending on
configuration
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
18
MONITORING MEMORY PERFORMANCE
 Memory: Page Faults/Sec—Should be <5
 Memory: Pages/Sec—Should be <20
 Memory: Available Bytes—Should not fall below 5%
of the system’s total physical memory
 Memory: Committed Bytes—Should always be less
than the physical RAM in the computer
 Memory: Pool Non-Paged Bytes—Should be a stable
number that does not grow without a corresponding
growth in server activity
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
19
MONITORING DISK PERFORMANCE
 PhysicalDisk: Disk Bytes/Sec—Should be equivalent




to the levels established in the original baseline
readings or higher
PhysicalDisk: Avg. Disk Bytes/Transfer—Should be
equivalent to the levels established in the original
baseline readings or higher
PhysicalDisk: Current Disk Queue Length—Should be
<2 per disk spindle installed in the storage
subsystem
PhysicalDisk: % Disk Time—Should be <80%
LogicalDisk: % Free Space—Should be >20%
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
20
MONITORING NETWORK PERFORMANCE
 Network Interface: Bytes Total/Sec—Should be
equal to baseline readings or higher
 Network Interface: Output Queue Length—Preferably
0, <2 acceptable
 Server: Bytes Total/Sec—Should be <50% of the
total bandwidth capacity
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
21
USING NETWORK MONITOR
 Protocol analyzer application included with
Windows Server 2003
 Provides the ability to view header and payload data
for common network protocols
 Installed through Add Or Remove Programs tool in
Control Panel.
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
22
NETWORK MONITOR VERSIONS
 The version of Network Monitor included with
Windows Server 2003:
 Can only view information sent to and from the system
on which it is installed
 Is limited in the statistical information it can report
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
CAPTURING NETWORK TRAFFIC
23
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
USING CAPTURE AND DISPLAY FILTERS
24
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
PLANNING A BACKUP STRATEGY
 Performing regular backups is one of the most
important functions of the system or network
administrator.
 When developing a backup strategy, ask and
answer two questions:
 Why do we back up data?
 Why do we need to back up data?
25
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
BACKUP HARDWARE
 Capacity
 Speed
 Cost
26
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
27
UNDERSTANDING MAGNETIC TAPE
COMPRESSION
 Most backup hardware and software support
compression to increase the capacity of a medium.
 Most manufacturers quote uncompressed/
compressed statistics at a ratio of 2:1.
 With some types of files, compression rates of up
to 8:1 are possible.
 Files that are already compressed will yield little or
no further compression.
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
USING AUTOCHANGERS
 Reduce the need for manual intervention in
backups
 Allow you to back up more data
 Preserve investment in media
28
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
29
MEDIA COSTS
 When evaluating backup hardware, consider the
cost of media.
 Media costs can sometimes be greater than the
cost of the actual drive.
 Media costs are ongoing: tapes must be periodically
replaced.
 Costs can be computed on a per-megabyte or
per-gigabyte basis.
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
30
BACKUP SOFTWARE
 Backup software provides mechanisms and
features that allow backup tasks and routines
to be configured and performed.
 Backup software for Windows Server 2003 includes:
 Windows Server 2003 Backup
 Third-party software, such as Veritas Backup Exec and
BrightStor ARCserve Backup
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
CREATING A BACKUP PLAN
 What data will be backed up
 When backups will occur
 Which tapes will be used and when
31
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
SELECTING BACKUP TARGETS
32
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
UNDERSTANDING BACKUP JOB TYPES
 Full backup
 Copies all the selected files to the backup medium
and resets the archive bits for all the copied files
 Incremental backup
 Copies only the selected files that have archive bits
and resets those archive bits
 Differential backup
 Copies only the selected files that have archive bits
without resetting those archive bits
33
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
34
SCHEDULING BACKUP JOBS
 Scheduling allows backups to be performed during
off-hours.
 Scheduling reduces potential for administrative
error.
 Media rotations use a specified number of tapes,
telling you which tape to insert in the drive
each day.
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
35
PERFORMING RESTORES
 Backup software should provide the following basic
restoration options:
 File selection
 Restore location selection
 Overwrite options
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
USING VOLUME SHADOW COPY
36
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
PROVIDING SECURE NETWORK
ADMINISTRATION
 Remote Assistance
 Remote Desktops
37
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
38
USING REMOTE ASSISTANCE
 Remote Assistance allows administrative personnel
to connect to a remote system in order to provide
help to a user.
 Experts can be restricted to viewing the remote
system or be granted full access.
 Remote Assistance can be enabled on a single
system by accessing Control Panel, selecting
System, and then Remote, or on multiple systems
via group policy.
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
CREATING AN INVITATION
39
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
40
SECURING REMOTE ASSISTANCE
 Invitations must be issued before a remote
assistance session can be established.
 The user must be present to accept a remote
assistance session.
 Clients can break the Remote Assistance
connection at any time by pressing ESC.
 Experts can be granted permission to view only the
user’s desktop, or to take control of the system.
 Firewalls should be configured to block port 3389.
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
USING REMOTE DESKTOPS
 Enables users to access computers from remote
locations with no interaction required at the
remote site
 Allows you to operate the remote computer as
though you were sitting at the console
 Allows up to two simultaneous Remote Desktops
connections on Windows Server 2003 without the
need for a separate license
41
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
ACTIVATING REMOTE DESKTOPS
42
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
USING THE REMOTE DESKTOPS CLIENT
43
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
44
CHAPTER SUMMARY
 System Monitor is an MMC snap-in that monitors
specific computer resources in real time.
 Performance Logs And Alerts is an MMC snap-in
that uses System Monitor’s performance counters to
capture information to log files over a period
of time.
 Network Monitor is a protocol analyzer that can
capture packets from the network and analyze their
contents in detail.
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
45
CHAPTER SUMMARY (continued)
 Server performance degradations are often caused
by a bottleneck, that is, a slowdown of a component
that prevents the system from performing to its
potential.
 A network backup solution consists of backup
drives, backup software, and a backup plan.
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
46
CHAPTER SUMMARY (continued)
 Automated System Recovery enables you to restore
an entire computer without reinstalling the
operating system.
 Volume shadow copy is a Window Server 2003
feature that enables users to access multiple copies
of files they have accidentally deleted or damaged.
 Remote Assistance is a Windows Server 2003 and
Windows XP feature that enables users to request
assistance from an expert at another location.
Chapter 11: MAINTAINING A NETWORK INFRASTRUCTURE
47
CHAPTER SUMMARY (continued)
 Remote Assistance cannot easily be abused
because users must request help before experts
can connect to their computers.
 Remote Desktops enables administrators to
connect to distant computers that are unattended,
and to work with them as though they were seated
at the system console.