feature: Device Discovery Details - NSCC`s TWiki

Download Report

Transcript feature: Device Discovery Details - NSCC`s TWiki

EtherScope™ Series II
Network Assistant
Next generation network
analyzer helps first responders
solve problems fast
ECA02132007
Outline
Introduction to EtherScope LAN
• Introduction to EtherScope WLAN
• Additional EtherScope WLAN Features
• RFC 2544 Testing
•
– (Throughput, Latency, and Frame Loss Rate)
2
Introduction to EtherScope LAN
LAN analysis and performance testing
10/100/1000Mbps RJ-45 twisted pair interface
1000Mbps LC fiber optic interface
Components
Stylus
Serial Port (can be used to configure
network devices)
Audio Ports (MIC, Headphone)
USB (optional mini-keyboard, mouse)
Power
Kensington Lock
LAN Test Port
Specifications:
• OS: Embedded LinuxTM (X-scale based)
• Touch Screen: 640x480 TFT (Active)
• Custom network testing hardware
SFP Fiber Transceiver
WLAN Card (PCMCIA)
Compact Flash (CF2)
• 256Mb RAM, 32Mb ROM
• 64 Mb Compact Flash Card Included (for
saved test results/reports)
•Expandable
• Weight: .82 kilograms or 2 pounds
• Dimensions: 19.1 x 15.2 x 4.4 centimeters or
7.5 x 6 x 1.75 inches
4
Components
Status LEDs
• Link
• Utilization
• Collisions
• Errors
• Transmit
Power On/Off
•Two modes:
Shut Down or Suspend
Protective Rubberized “Boot”
Bright Active Matrix Touch Screen
Battery:
• Lithium-Ion Rechargeable
• Approx. 4 hours use / charge
• Optional extra battery and charger
stand
5
Home Page: Test Results and Navigation
Touch icon for
instant link to
specific tests
“Menu”-based
navigation with
highlights of
selected test
shown at left
Blue text
indicates
hyperlinks to
more detail
“Details”
drills in on
selected test
for more
results
Easy navigation:
• Back
• Home
• Tools
• Help
Tool Bar – a convenient and consistent
navigation and information platform.
6
Instrument Settings
Password
to control
community
string and
Remote U/I
access
Get an IP
Address via
DHCP, or
manually DHCP Server
Log appears
after hitting
“Apply”
Save time with
‘Fast Connect
Mode’
• Check for Link
• Check Address
• Go to next drop
Note the
default
community
strings
Add management
VLAN devices here
Drill in on Connection
to reach Instrument
Settings, where you
can configure TCP/IP,
Ethernet, Security, and
General instrument
settings. Add
community strings and
security passwords,
change Ethernet
negotiation
characteristics and
interfaces.
Full control of
your
connection
settings
7
Cable Verification
8
Cable Verification Details
Changes the color
code order in
wiremap results
Launches onboard ANALOG
Toner
9
IntelliTone Toner
•
Generate analog or IntelliTone tones to help locate copper
cabling
– choice of two tones
•
Click Cable Verification, Details and Cable Toner
choice of
modes
and tones
10
Fiber test kit (DSP-FTK) support
•
•
•
Measure the optical power or the loss of an optical fiber link using the
DSP-FTK fiber test kit
– DSP-FTK = DSP-FOM and FOS-850/1300
Click Cable Verification, Details, then select the Fiber button
Connect DSP-FOM to LAN port
auto
wavelength
detection
(FOM)
power,
loss and
margin
set reference for loss
measurement
11
Signal Verification
Checks for
presence and
amplitude of link
pulse and data
signals
Scans for DC
voltage levels
and
over voltage
conditions.
Perfect for
troubleshooting
Power Over
Ethernet!
Supported signaling and
link partner-EtherScope
advertised autonegotiation
12
Traditional switched network
Switch
Floor 3
Switch
Floor 2
Switch
Floor 1
Broadcast
domain
Router Switch
13
Switched network with VLANs
Engineering
VLAN
Either
dedicated
cables or a
VLAN “trunk”.
A trunk link can
carry traffic
from multiple
VLANs.
Router
Switch
Floor 3
Switch
Floor 2
Layer 3 Switch
Switch or
Router
Accounting
VLAN
Sales
VLAN
Broadcast
domain
Floor 1
14
EtherScope network visibility boundaries
From a single connection, EtherScope can see the
•
•
Collision Domain - local statistics
Broadcast Domain – device, network, VLAN discovery
Collision domain statistics
Broadcast domain discovery
(SNMP)
15
EtherScope local VLAN statistics
•
•
•
If the ES is connected to a trunk port, multiple VLANs
will be present
Use “VLAN Statistics” to identify the VLANs that are
active on the port (includes the Native or untagged
VLAN traffic)
If you select a VLAN ID that is
unconfigured on the port to which the
EtherScope is connected, it might
not be able to communicate with the
network (e.g. DHCP fails)
– Try configuring the ES for the VLAN with
the highest packet count
16
EtherScope VLAN discovery
•
•
The number of VLANs reported by VLAN Statistics and
VLAN Discovery will frequently differ
VLAN Statistics shows VLANs detected by monitoring the
local network segment while VLAN Discovery is using
active (SNMP) discovery
17
IEEE 802.1Q
•
•
•
Standard methodology for constructing a VLAN trunk
A “VLAN tag” is added to the Ethernet frame by the switch
that is trunking traffic to identify the VLAN
The last trunk switch will strip off the tag before forwarding
the frame to the destination
VLAN tag
(includes CoS)
18
EtherScope on a trunk port
•
•
•
•
•
Tap “Enable 802.1Q” checkbox to turn on VLAN tagging
Use the “VLAN ID” field to designate the VLAN number that
the ES will use
Use the “Priority” field to set the user priority (CoS) for the
frames generated by the ES
Active discovery will be limited to
the devices on the selected VLAN
(broadcast domain)
Can only enable 802.1Q if the port
of the link partner has .1Q enabled
19
Class of service (CoS) priority
•
CoS provides a means of prioritizing which traffic the
switch will process first
– Example: give VoIP traffic higher priority than web traffic
– 802.1p is an subset of 802.1Q VLAN tagging
•
•
Layer 2 intranetwork priority mechanism (switches and
MAC addresses)
802.1Q and 802.1p CoS support
added to EtherScope to facilitate
observation of layer 2 CoS
priority on link performance
– User can configure the priority of
the traffic EtherScope generates
20
Type of Service priority
•
Like CoS, provides a means of prioritizing traffic
– Layer 3 internetwork traffic (routers and IP addresses)
– Examples: VoIP traffic over web traffic , critical application traffic
over email
•
Two versions
– TOS with IP precedence
• 4 attributes (delay, throughput, reliability, cost)
– TOS with differentiated services code point (DSCP)
• 8 attributes for more flexible control
•
Support added for IP type of service (TOS)
Layer 2 Addressing
Dest
MAC
Source
MAC
Type
0800
Layer 3 Addressing
Source
IP
Dest IP
Data
FCS
TOS Byte of IP header
21
Local Statistics
Notice the
detailed preview
information –
giving you a
quick look so
you can decide
at a glance
whether you
want to drill-in
further…
Something
looks suspicious
here, don’t you
think? –
Just click on
‘Details’
22
Local Statistics, Details
Note that you can
also change the
source, and the
interface to trend a
remote device as
well. This helps
you follow the
scent of problems
and enables you
to track them
down and fix them
quickly!
23
Local Statistics, Details
So you click on the
Details button or
on the Local
Utilization link to
see what’s
happening.
Note that you can
also change type
of information
displayed. Here
you see Error
Details. Oversize
and Undersize
frames
24
Remote Statistics, Details
Choose a different
source (switches
only!) and
interface to
monitor
On selected
device and
interface,
EtherScope
reads only the
Layer 2
interfaces.
It Pulls MIB II
information, but
no RMON history,
so single port
statistics are
shown
25
Top Protocols
Notice the summary
protocol statistics,
listed in BLUE – that
means they are
‘clickable’ to drill in for
additional details
26
Protocol Statistics Detail
Throughout EtherScope’s User
Interface, data can be sorted by
headings, and selected by type
Top sources of the
selected packet type
are listed in order.
27
Top Talkers
At the top level, see
top talkers at a
glance. Drill in to see
details of this
particular bandwidth
hog
28
Top Talker Detail
Select the radio buttons to list senders of Errors,
Broadcasts and Multicast traffic
Detailed information
available on top
talkers. Select a
particular talker and
tap the ‘Details’ button
to obtain still another
level of drill-in.
Click to see
Layers of
protocols
available
Note that where appropriate, applicable buttons will appear, incontext… Here you can ‘Clear’ the packet counts, or generate and
store a ‘Report’
29
Top Talker IPV4 Protocol Detail
Detailed
information
available on the
distribution of
IPV4 Protocol
30
Device Discovery
Summary level
information about
the devices
discovered. Again,
you can drill in to
see a list of all
devices, or a
particular type of
device you are
interested in, just by
selecting it
Note that while the green ‘checkmark’ indicates that EtherScope has
completed its first pass on Device
Discovery - it continues to listen to
traffic to identify and report on new
network devices it discovers.
31
Device Discovery, Details
Drilled-in on All
Devices, see the
Name, IP Address,
Switch Slot/Port,
VLAN, and any
associated problems
Use these radio
buttons or use the
scroll-bar to navigate
through the table to
see additional
details on the
devices.
32
Device Discovery Report
A sample
report listing
pertinent
information on
all devices.
Column
headings can
be sorted here
as well.
Document the network!
Reports saved to Compact Flash, accessible via Web interface
MUST have CF installed to save reports!
33
Device Details by Type - Switches
Overview of device
configuration
information is shown
here
Individual Device
Traffic shown here.
Select a device and
tap the ‘Details’
button to again to
obtain still another
level of drill-in.
34
Specific Device Details
Device-specific links
and appropriate
troubleshooting tools
are now available.
Note that you can
drill in on the switch
interfaces to see
who’s connected on
each port
Quick launch
troubleshooting tools
Detailed information
about this particular
switch is shown in
the main screen
35
Switch Interface Details
Here we have
drilled-in on the
switch Interface
Detail link which
shows the status
and configuration of
all the ports,
including who’s
connected to each!
You can even select
one of the hosts and
drill in further on it…
Notice the problem icon showing Host
EVTPC1272… time for some investigation.
You can sort on that column to see all the
problems discovered on this switch port…
Problem discovery speeds troubleshooting.
36
Trace Switch Route
Trace Switch Route
shows you the Layer
2 connection path
between
EtherScope and the
selected device.
The “from” device is
always the
EtherScope. The
“destination” is
selected in the drop
down box.
Want to see the interface details or monitor utilization on the connecting switch port? Just drill in on the
BLUE highlighted Slot/Port to go there.
37
Network Discovery Overview
EtherScope lists all
the networks
discovered, and
provides quick
information about
the number of hosts
on each
38
Network Discovery Details
Select and expand
the IP Subnets to
see detailed
information on how
the network is
organized, and
which hosts are
members
By now, you
recognize that you
can select and drillin on any subnet or
device to get
additional details…
39
VLAN Discovery
See VLANs, and
associated switch
ports and hosts.
Trunks are not
shown.
Discovery of all VLANs
within connected
broadcast domain only
(discovered switches).
Finds by way of reading
switch configs
Lists hosts by VLAN
40
Nearest Switch
EtherScope locates
the nearest switch
and monitors it for
you
41
Switch Scan
Monitoring these
switches for high
utilization and errors.
Automatically shows
nearest switch and
one other userselected switch.
42
Switch Scan Details
See the Average
and Peak Utilization
on all the active
ports of two selected
switches.
Visibility of traffic IN
and OUT.
Average and Peak
reading results
shown since Switch
Scan was activated.
Select another switch to monitor from the dropdown list
43
Switch Scan Trending
Select a port from
the previous then
click Trend for
Utilization History
44
Key Devices
EtherScope
automatically checks
on selected Key
Devices when it
starts up, and again
whenever you select
‘Start Test’
45
Problem Log Details
EtherScope monitors
network conditions and
automatically alerts you
to detected problems.
Errors, Warnings, and
Info Messages are
listed. Intermittent
Problems are reported
and listed as Resolved,
so even if they go away
you know about them
You can delete selected or resolved problems
(or ‘undo’ if deleted by accident)
46
Documenting the LAN
•
Save results as a Report to document wired LAN attributes
– click on Report button found on most Detail screens
•
Use Report Viewer and File Manager applications to view
and manage reports on-board
open to
view
open,
rename or
delete
47
Remote Access Via Web Browser
Access real time or
saved reports
Direct link to support resources
at www.flukenetworks.com
Launch a remote
session; can require
password – default
is blank (nothing)
Access the embedded Online Help file
48
Performance testing
•
•
•
•
•
How do you know that a newly installed network link or
network device is really delivering the performance he
expects?
Is the performance sufficient for successful data transfers,
email, web-based seminars, VoIP and other applications?
Are user complaints of a slow network true?
How will adding additional users to a segment of your
network impact network performance?
The RFC 2544/ITO Option helps answer these questions.
49
RFC 2544/ITO Option
•
•
•
•
•
This option includes
– RFC 2544 performance testing
– a traffic generator
– the proprietary Throughput Test found on other Fluke Networks
analyzers like the OneTouch Network Assistant and the OptiView
Integrated Network Analyzer
Select the RFC 2544 tests to characterize the performance of a
network link using the well-accepted IETF recommendation
Select the proprietary Throughput Test for quick verification that a link
can carry a specific, user-defined data rate for a set interval of time
Test at up to full line Gigabit rates
Over twisted pair or optical fiber (SX, LX, ZX) cabling
50
RFC 2544 Tests
•
Characterize network performance by
selecting one or more of the three most
valuable RFC 2544 performance metrics
– throughput
– latency
– frame loss rate
•
For each test, you can use the default
frame and test defaults, or customize
the configuration
– When testing a trunk port for example,
you can identify a VLAN ID, turn on
802.1Q VLAN tagging and override
the default 802.1Q priority setting
– You can override the Type of Service
(TOS) IP Precedence and DSCP
parameter
– You can also control the time it takes
to complete a test by adjusting the
accuracy and failure thresholds.
51
Maximize efficiency and timesavings
•
Select multiple target devices for
each RFC 2544 test to maximize
testing efficiency and timesavings
– Add the same target device, with
differing test configurations,
multiple times to the same test to
see how traffic differences affect
link performance
– Add different target devices to
the same test to measure
multiple links within your network
– Devices and configurations can
vary by test
– Once you construct your test
suite, click the Start button once
to commence testing to all
device
Main
Network
Remote
52
Save, print, email or store test results
•
•
•
•
Tests run in both upstream and
downstream directions, saving
time by eliminating the need to
physically swap source and
target devices
You can view the results by
device and by test
See test results in table view or
graph view mode
Save, print, email and store the
results to document link
performance.
53
ITO Traffic Generator
•
•
•
Use a single EtherScope to generate traffic up to full line rate at
10/100/1000 Mbps speeds
Simulate the addition of new users on a segment by generating
broadcast, multicast or unicast traffic
Select the frame type and size and the test duration.
can also
access from
Tools
54
Introduction to EtherScope WLAN
802.11a/b/g wireless LAN analysis
What are the problems with wireless
networks
•
Don’t know what network types are active
– feature: Network Discovery
Don’t know who is using the WLAN
• Need to verify device configurations
• Can’t connect or login
• Not getting the expected performance (“too slow”)
• Security concerns (e.g. rogue clients)
• WLAN isn’t documented
•
56
Discovering the types of wireless networks
•
•
click “Network Discovery”
Four network types are discovered
– an infrastructure network is comprised of systems and mobile
clients connected through an Access Point (AP) which is connected
to wired Ethernet network. Most common type.
– an adhoc network is comprised of mobile clients transmitting
directly to any other; No AP needed
– a bridge network is comprised of two APs or dedicated 802.11
bridges that communicate to provide link between two physically
separate networks
– a IP subnet is comprised of WLAN devices for which IP addresses
have been discovered including EtherScopes
57
Network discovery
•
Click Details to see the wireless network hierarchy
discovered by EtherScope at its present physical location
insert screen shot
of Network
Discovery details
58
What are the problems with wireless
networks
•
Don’t know what network types are active
– feature: Network Discovery
•
Don’t know who is using the WLAN
– feature: Device Discovery
Need to verify device configurations
• Can’t connect or login
• Not getting the expected performance (“too slow”)
• Security concerns (e.g. rogue clients)
• WLAN isn’t documented
•
59
Discovering who is using the network
•
•
•
•
•
•
•
Click “Device Discovery” then “Details”
see the number of devices on the network by type
an access point is the “connecting” point used to coordinate wireless
traffic and provide connectivity to the LAN.
a bridge node is a device that provide a wireless connection between
wired networks. This feature can be built into Access Points.
a mobile client is an end point or node on the wireless network. These
are typically general purpose computers but might also include other
dedicated wireless devices.
an adhoc client is a node that comprises a peer-to-peer network that
requires no access point.
a host is a special category that only includes This EtherScope.
60
Device discovery
•
Click Details to see a comprehensive list and summary
device information
insert screen shot
of Device
Discovery details
61
What are the problems with wireless
networks
•
Don’t know what network types are active
– feature: Network Discovery
•
Don’t know who is using the WLAN
– feature: Device Discovery
•
Need to verify device configurations
– feature: Device Discovery Details
Not getting the expected performance (“too slow”)
• Security concerns (e.g. rogue clients)
• Can’t connect or login
• WLAN isn’t documented
•
62
Configuration information presented
•
SSID is Service Set ID. Typically, the wireless network comprised of an
AP and wireless stations is given an SSID name. This uniquely identifies
a WLAN and is used when configuring security options.
•
MAC is a medium access control address. 802.11 devices have unique
MAC addresses that are used to identify devices.
•
Channel is a common frequency used to direct communication between
an 802.11 radio card and an access point.
•
Signal is the transmission of 802.11 frames by radiating RF energy at a
given strength. The stronger the signal the better the coverage and
connection.
•
Noise is unwanted, interfering RF energy that disrupt normal system
operations. Noise level impacts connectivity and performance.
•
Security represents the level of 802.11 packet encryption and network
authentication for which a device is configured.
63
See traffic summary information by device
•
Select a device
– packets are units by which
802.11 traffic is transported.
– broadcasts are packets sent
to all station on a network
(SSID).
– (packet) retries are sent
when a receiving station fails
to acknowledge a packet
64
See device detail overview
•
•
Highlight a device and click Details
See detailed configuration information and options for
additional troubleshooting tools and tests
65
What are the problems with wireless
networks
•
•
•
•
•
•
•
Don’t know what network types are active
– feature: Network Discovery
Don’t know who is using the WLAN
– feature: Device Discovery
Need to verify device configurations
– feature: Device Discovery Details
Not getting the expected performance (“too slow”)
– feature: Channel statistics
Security concerns (e.g. rogue clients)
Can’t connect or login
WLAN isn’t documented
66
Troubleshooting wireless LAN
performance and connectivity issues
•
•
•
A common measure of wireless network performance is speed, typically
expressed in Mbps
Speed is impacted by the transmission technology and local RF
conditions
Enterprise WLAN technologies:
– 802.11b networks operate in the 2.4 GHz Industrial, Scientific and
Medical (ISM) band of the RF spectrum, sub-divided into 14
channels 22 MHz wide, with indoor range of ~300 feet and max
throughput of 11mbps.
– 802.11g networks operate in the 2.4 GHz band of the RF
spectrum, sub-divided into 14 channels 22MHz wide, with indoor
range of ~300 feet, and max throughput of 54mbps.
– 802.11a networks operating in the 5 GHz band of the RF
spectrum (less contention than the 2.4 GHz band), is sub-divided
into 12 non-overlapping channels 20 MHz wide, with indoor
range of 60-80 feet, and max throughput of 54mpbs.
67
Analyze performance and connectivity
issues by examining channel statistics
•
A channel is defined as the common frequency that is used
to direct communication between an 802.11 radio card and
an access point.
– 802.11b/g
• 14 channels; 11 in US
• Channels 1, 6, and 11 non-overlapping
• Uses 22MHz wide channels
– 802.11a
• 30 channels
• Uses 20MHz wide channels
• NO overlapping channels
68
Channels
•
•
Start wireless LAN performance analysis by looking at channel
usage
Click on Channels – Details then select the “Total Utilization %”
channel metric
69
Channels
•
Next look at coverage
– indication of signal quality
•
Click on Channels – Details then select the “Signal vs
Noise dBm” metric
70
Noise
Non 802.11 energy (non-compliant signal strength)
• Get noise readings from radio card
• Indicates noise floor level
• Used to evaluate channel quality
•
– Signal-to-Noise ratio
•
•
•
Noise level impacts connectivity and performance
Shows distribution of noise across channels
Channel Quality represented by signal-to-noise ratio (rule
of thumb?)
71
Channels
•
•
Next look at ability to successfully transmit packets
Click on Channels – Details then select the “Retry %” channel metric
72
Channels
•
•
If you want to drill deeper on a
specific channel, click on
channel of interest then Details
– focused on a channel, get
more accurate information
Same channel metrics as
available via pull down box
73
Channels
•
If you want to know more about the devices on a channel,
click on channel of interest then Devices
74
Analyze wireless network health by
examining channel utilization
•
A channel utilization measures the % of bandwidth
consumed by packets transmitted by wireless devices
(Access Points, mobile clients, etc.) on a single frequency
75
Channel utilization
Click on Channels Utilization then Details for utilization by
frame (traffic) type and protocol
•
•
Frame types
–
•
unicast is a packet for transporting information
directly from one point (node) to another.
–
–
–
–
–
multicast is a packet that contains a group
address that delivers the same packet to more
than one destination.
broadcast is a packet where the destination
address refers to all nodes on the network.
retries are typically sent when a receiving
station fails to acknowledge a packet.
FCS errors are indications that corrupted
packets have been received.
Crosstalk is an indication that packets sent on
one channel have been received on another
channel due to channel overlap.
Protocols
–
–
–
–
–
–
data is a frame that is used to send data;
contains the payload.
management is a frame used to enable
stations to establish and maintain
communications.
control is a frame that assist in the delivery of
data frames between stations.
retries are typically sent when a receiving
station fails to acknowledge a packet.
FCS errors are indications that corrupted
packets have been received.
Crosstalk is an indication that packets sent on
one channel have been received on another
channel due to channel overlap.
76
Channel utilization
•
•
The distribution of frames and protocols is useful to
understand network health
Click Packet Rates on Channel Utilization, Details
77
What are the problems with wireless
networks
•
•
•
•
•
•
•
Don’t know what network types are active
– feature: Network Discovery
Don’t know who is using the WLAN
– feature: Device Discovery
Need to verify device configurations
– feature: Device Discovery Details
Not getting the expected performance (“too slow”)
– feature: Channel statistics
Security concerns (e.g. rogue clients)
– feature: Security Scan
Can’t connect or login
WLAN isn’t documented
78
Identifying security vulnerabilities
•
Security threats are identified two ways
– unprotected devices are not using encryption to manage
connections
•
unprotected devices can be a path into your network
– unauthorized devices are not approved by the IT manager to
be on the network
• EtherScope assumes all devices are unauthorized until made otherwise
79
Unprotected devices
•
Click Security Scan and the Unprotected hyperlink
80
Locating unprotected devices
•
•
Highlight the device of interest then click Details
Click the Locate hyperlink
81
Locating methods
•
2
Convergence method
–
–
–
–
–
•
1
Use common omni-directional antenna
Divide search area into quadrants
Measure signal strength in each corner
Move to corner with highest power
Repeat until found
5
4
9
6, 10
12
11
8
3, 7
Vector method
– Use uni-directional antenna
– Divide search area into quadrants
– From center, measure signal strength every
90°
– Move to quadrant with highest power
– Repeat until found
1
3
2
82
Unauthorized devices
•
Click Security Scan and the Unauthorized hyperlink
– same Location feature to find devices
insert screen shot
of security scan
unauthorized
devices
83
Authorizing devices
•
Click Instrument Settings – Authorization to change the
current level of authorization
– select devices individually or by level
– change to authorized, unauthorized or neighbor
• remember to Apply!
84
What are the problems with wireless
networks
•
•
•
•
•
•
•
Don’t know what network types are active
– feature: Network Discovery
Don’t know who is using the WLAN
– feature: Device Discovery
Need to verify device configurations
– feature: Device Discovery Details
Not getting the expected performance (“too slow”)
– feature: Channel statistics
Security concerns (e.g. rogue clients)
– feature: Security Scan
Can’t connect or login
– feature: Login Diagnosis
WLAN isn’t documented
85
Diagnosing client login problems
•
If a client is unable to login, you can observe the login process to see
where the process fails
– Click Device Discovery, Details, Mobile Clients then highlight the
client (with ES physically near)
– Click Details, Login Diagnosis and Start
– Attempt a client login
Insert screen shot and emphasize
key points
86
What are the problems with wireless
networks
•
•
•
•
•
•
•
Don’t know what network types are active
– feature: Network Discovery
Don’t know who is using the WLAN
– feature: Device Discovery
Need to verify device configurations
– feature: Device Discovery Details
Not getting the expected performance (“too slow”)
– feature: Channel statistics
Security concerns (e.g. rogue clients)
– feature: Security Scan
Can’t connect or login
– feature: Login Diagnosis
WLAN isn’t documented
– feature: Reports
87
Additional EtherScope WLAN Features
Other wireless LAN features
Problem log
• Survey site changes
• Monitoring key devices
• Top talkers
• Using network maintenance tools
• Configuring your EtherScope
•
89
Detecting problems
•
•
EtherScope automatically identifies
potential problems and lists them by
severity
Problem identification is thresholdbased
– click Wireless Instrument
Settings – Wireless Problems
90
Survey site changes
•
•
Select a location from which to
periodically survey the WLAN
– click Site Survey, Details, edit
Location then Save
– survey data includes APs,
channels, error and retry rates,
utilization, signal and noise
measurements
Go back to that location at a future
date and recall the Previous Survey
– Differences will be highlighted
in red
91
Monitoring key devices
•
•
Select any discovered device with an IP address
Click Key Devices, Details, then select device(s) from the pull down menu
insert screen shot
of key devices,
details
92
Learn who is hogging network bandwidth
•
•
Click on Channels Top Talkers –
Details
Isolate where the problem is by
filtering on channel, network
93
What’s a network manager to do?
•
What does a network manager do if EtherScope uncovers
a performance or connectivity problem?
– minimize congestion and/or interference by load-balancing
channels
• APs and clients
– improve coverage by adding APs
– alter the RF environment
• remove interferers
• re-position APs, office equipment, walls
94
Network maintenance tools
•
EtherScope contains many tools to help in configuring and
maintaining the WLAN
95
Tools – Connectivity Testing
•
•
Ping
– Easy method for verifying IP-level
connectivity between EtherScope and
another device
– Sends packets to a device to see if it sends
a return or an 'echo.' If all the packets
'timeout' the device may be disconnected
from the network or at least unreachable
– Launched from within EtherScope
application; results displayed on
EtherScope Console
– Can run multiple sessions
Trace Route
– Used to determine the IP path a packet
uses to reach a device. It shows the
number of hops and the IP addresses of
devices along the path to a destination host
– Launched from within EtherScope
application; results displayed on
EtherScope Console
– Can run multiple sessions
96
Tools – Infrastructure Configuration
•
Web browser
– Uses Konqueror as the browser; more limited than multi-purpose Internet Explorer
or NetScape
– Used to remotely access an infrastructure device web interface for checking and
changing configuration
– Used to access web sites for knowledge base information
•
Telnet/Secure Telnet
– Used to remotely access an infrastructure device command line interface for
checking and changing configuration
– Opens a telnet session
– Secure Telnet offers more security before access a device console via an additional
login
•
Terminal
– Allows EtherScope to be used as an ASCII terminal (terminal emulation)
– Connect to infrastructure device and EtherScope via a serial cable; opens a
terminal window
– Used to initially configure a switch (e.g., set IP settings, passwords, enable SNMP)
97
Tools – Infrastructure maintenance
•
FTP/TFTP
– File Transfer Protocol (FTP) is a common means used to move files
between devices
– Used to transfer configuration and user files to and from
EtherScope
– Trivial File Transfer Protocol is a simple version of FTP used by
most infrastructure venders to “put” configuration files and new
operating system version onto switches and routers
– Files can be “put” on EtherScope and then used to update a switch
or router configuration or version of code
– User configuration files can be stored on a compact flash card
98
Configurations (Wireless Instrument
Settings)
•
TCP/IP Connection Log provides an unique view into the WLAN
connection process.
insert screen shot
of connection log
99
Configurations (Wireless Instrument
Settings)
•
•
Wireless Security is where
you set the SSID that
EtherScope will use to
establish link
Depending upon the
security authentication type
you may have to enter
additional keys, user
names, passwords or
certificates
100
Configurations (Wireless Instrument
Settings)
•
•
Click Radio to set the Country,
Active Bands, Transmit Settings
and Signal measurement units
You can also apply signal
strength corrections
– using corrections you can
have the EtherScope WLAN
card emulate the RF
characteristics of other
WLAN cards
101
Tools for network maintenance
Only available in LAN mode
• Trace Switch Route
• Traffic Generator
• SSH Telnet
• CDP Port Reporter
102
Tools
•
Trace Switch Route (TSR)
– Used for troubleshooting connectivity
– Determines the MAC data path between
EtherScope and another device including the
starting and ending device and any switches
along the path
– SNMP community strings must be configured for
TSR use
– Switches must be discovered (or user added)
for TSR to work properly
•
Traffic Generator
– Optional application that is used to generate
synthetic traffic
– Used to test the performance of a network by
creating various network loads (high utilization,
simulate large number of users)
– Can enable a password to protect the
inexperienced user from creating serious
network problems
103
Tools
•
CDP Port Reporter
– Utility that provides quick reporting
of switch name, address, port
identification
– Provides above information from
point where EtherScope is
connected to the LAN
– Cisco Discovery Protocol is used as
the basis discovering; does not
require SNMP to be enabled
– Designed to work in homogeneous
environments
– Launched from within EtherScope
application; runs and reports results
from EtherScope console
104
Channel metric definitions
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Signal Strength - the average signal strength of all detected frames. Error, crosstalk and probe request
frames are not included. Display of dBm or percent values configured globally via Radio Settings.
Noise - the average channel noise value when any frame is detected (including error, crosstalk and probe
request frames).
Signal vs Noise - Signal strength value displayed on top along with the noise value on the bottom.
Signal to Noise Ratio - a measure of average signal strength relative to average noise. The ratio is always
displayed in dBm.
Good Packet Rate - The number of good frames detected. Error, crosstalk or retry frames are not included.
Error Packet Rate - The number of frames not received intact. Crosstalk or retry frames are not included.
Retry Packet Rate - The number of retransmitted frames detected. Error and crosstalk frames are not
checked for retry status.
Crosstalk Packet Rate - The number of frames detected on a channel that were actually transmitted on a
different channel. A frame must be received error free to be identified as crosstalk.
Good Octet Rate - The number of good frame octets detected.
Error Octet Rate - The number of frame octets not received intact.
Retry Octet Rate - The number of retransmitted frame octets detected.
Crosstalk Octet Rate - The number of frame octets detected on a channel that were actually transmitted on a
different channel.
Total Utilization % - The percentage of the channel's bandwidth that is being utilized by all frame
transmissions.
Retry % - The percentage of all detected frames on the channel that are retransmission frames.
Crosstalk % - The percentage of all detected frames on the channel that were actually transmitted on a
different channel.
Note: in Channels mode, these values are computed and updated for each channel scan period. In Channel
mode, these values are computed and updated once a second.
105
RFC 2544 Testing
Service Level Agreement (SLA)
•
•
A contract between a network service provider and a
customer that specifies, usually in measurable terms,
services the network service provider will furnish
A service level agreement would typically contain the
following information:
– A description of the nature of service to be provided
– The expected performance level of the service, specifically its
reliability and responsiveness
– The procedure for reporting problems with the service
– The time-frame for response and problem resolution
– The process for monitoring and reporting the service level
– he consequences for the service provider not meeting its
obligations
– Escape clauses and constraints
107
SLA and expected performance
•
•
Performance metrics quantify end-user visible perceptions of service
performance
Typical performance metrics specified within a SLA
– Throughput
• The maximum rate at which data can be transported from source to destination with zero
lost frames
– Latency
• The total time it takes for a frame to travel from source to destination. The context for a
relevant latency test is while the link is under load. The network needs to be able to deliver
specified latency at specified utilization. RFC 2544 clearly states you have to do this test
under the full throughput load
– Frame loss
 Test determines the percentage of frames that left the source but never reached the
destination. RFC 2544 recommends starting Frame Loss Rate tests at 100%, then stepping
down to the next rate to test for lost frames. A successful test is defined by two consecutive
tests without frame loss. Some frame loss rate tests waste time because they aren’t
designed to terminate the test once this loss rate has been located.
108
RFC 2544 testing
•
•
•
•
•
IETF RFC 2544 “discusses and defines a number of tests that may be
used to describe the performance characteristics of a network
interconnecting device.”
The RFC 2544 guidelines provide a standard testing methodology
RFC 2544 is optimized for testing single routing devices in a laboratory
environment
Not all the tests are applicable for providers installing or
troubleshooting Ethernet-based IP on active WAN links
3 of the 6 tests specified by RFC 2544 are really necessary to
characterize performance in an Internet environment
– Throughput
– Latency
– Frame Loss Rate
109
RFC 2544 tests
•
•
While the RFC provides testing guidelines, implementations differ by
solution provider
Advantages of EtherScope’s implementation of RFC 2544
– Control over test configurations for predictability
• User-defined parameters allow for tradeoffs between accuracy and test duration
• Adaptive algorithms speed time to derive results
– Custom test suites for test automation
• Fully characterize a single link by running multiple tests with varying parameters
• Test multiple links from a single destination using multiple EtherScope remotes
• Construct a test suite, click Start and all tests run automatically
– Accuracy for precise measurements
• Latency measured with microsecond accuracy
• Useful when characterizing single device performance
– Reports to document result
110
RFC 2544 step by step
Define which test(s) to run
2. Add device(s) (aka ES remotes)
3. Configure tests
4. Run tests (as one complete suite)
5. Save results (graphs too)
1.
111
RFC 2544 step by step
1. Define which test(s) to run
112
RFC 2544 step by step
2. Add device(s) (aka ES remotes)
•
•
•
Can differ by test
Same device multiple times with different configurations
Multiple devices to test several links
EtherScope Main
EtherScope R1
EtherScope R2
113
RFC 2544 step by step
3. Configure tests
–
–
Adjust parameters to tradeoff accuracy for quicker test completion
Adjust parameters (like CoS priority and IP TOS parameter) to
observe performance differences
114
RFC 2544 step by step
4. Run tests (as one complete suite)
115
RFC 2544 step by step
5. Save results (graphs too)
116
RFC 2544 versus ITO
•
•
•
EtherScope v3 features both RFC 2544 tests and ITO tests
RFC 2544 tests – use to fully characterize link performance
– Throughput [maximum rate at which none of the transmitted frames
are dropped]
– Latency [total time it takes for a frame to travel from source to
destination]
– Frame loss [percentage for frames lost throughout the range of test
rates, starting at max rate and stepping down until rate with no lost
frames is reached]
ITO tests – use to quickly verify the throughput of a link
– Throughput [measure frame loss at a user defined rate]
– Traffic generator [generate network traffic, use to create different
loads to test performance]
117