TextPing User Guide V1 - PacketIQ - Intelligent Network & … TextPing User Guide v1.3.pdf ·...
43
TextPing ™ User Guide V1.3 © Copyright 2012 PacketIQ Inc. www.packetiq.com
Transcript of TextPing User Guide V1 - PacketIQ - Intelligent Network & … TextPing User Guide v1.3.pdf ·...
2
NOTICE PacketIQ Inc. provides this software program and this publication “as is” without warranty of any kind, either expressed or implied, including, but not limited to, the implied warranties of merchantability or fitness for any particular purpose. PacketIQ Inc. does not guarantee and is not responsible for the functionality, accuracy, or usefulness of this program beyond the features and intended purposes documented in this User Guide. PacketIQ Inc. will not be liable (i) to any person or company or entity for any incidental, consequential, or indirect damages (including damages for loss of business profits, business interruption, loss of business information, and the like) arising out the use of or inability to use this product even if PacketIQ Inc. or any authorized PacketIQ Inc. representative has been advised of the possibility of such damages, or (ii) for any claim by any other party.
Further, PacketIQ Inc. reserves the right to make changes or improvements to the product described in this guide and to this publication without obligation of PacketIQ Inc. to notify any person of such revision or changes, even if such changes reduce or eliminate a previous capability of the product.
Copyright © Copyright 2012 PacketIQ Inc. All rights reserved. No part of this publication may be reproduced, translated, or distributed without prior written permission of PacketIQ Inc.
Document Revision: 1.1
Product: PacketIQ TextPing
Software Version: 1.3.1.0
Version Date: December 20, 2012
About PacketIQ PacketIQ Inc. provides advanced network and application performance analysis, modeling, troubleshooting, and capacity management services. PacketIQ Inc. has developed a suite of innovative applications to enable delivery of superior analysis services, and markets these applications to clients and consulting professionals.
Contacting PacketIQ
Sales & Info [email protected]
www.packetiq.com
Technical Support [email protected]
3
Reporting Problems PacketIQ provides email support of this application on a best-effort basis.
Problems should be reported to [email protected]. Please include the following:
Version number and version date of the product
Environment the application is working in (Windows version – CPU – Memory)
Full description of the problem, including screenshots if applicable
If possible, copies or partial copies of a log file and/or profile and/or any supporting information with which the problem can be duplicated.
Suggesting Improvements Suggestions for adding additional features or functionality to this product can be sent to [email protected] for possible inclusion in subsequent versions.
Trademarks All trademarks, trade names, service marks and logos referenced or displayed herein belong to their respective companies.
Microsoft and Windows and Excel are registered trademarks or trademarks of Microsoft Corporation in the United States and/or other countries.
TextPing and PacketIQ and the PacketIQ logo are trademarks or registered trademarks of PacketIQ Inc.
Notice: Generation of Network Traffic by TextPing This application generates various amounts of network traffic (ICMP 'Ping' and/or DNS query packets) to provide packet trace file 'marking' during test captures and to perform an analysis of network latency and an approximate effective network throughput measurement. The user of this application is solely responsible for understanding the application control settings and their effects on generated traffic volumes.
PacketIQ is in no way responsible for disruptions of network operation due to the use of this application in any way by any persons regardless of presentation or interpretation of the contents of this document.
If there is any question as to the nature, volume, or circumstances of network traffic generation by this application the user should perform an independent analysis of this application in an isolated lab setting using the expected settings and user activities to acquire such knowledge.
4
Table of Contents
Table of Figures ..................................................................................................................................... 7
1 Introduction ..................................................................................................................................... 8
2 TextPing Download, Installation, and Removal ............................................................................... 9
2.1.1 Obtaining TextPing ........................................................................................................... 9
2.1.2 Installing TextPing ............................................................................................................ 9
2.1.3 Removing TextPing .......................................................................................................... 9
3 TextPing Quick Reference ............................................................................................................ 10
3.1 Using TextPing for Marking Packet Captures......................................................................... 12
4 TextPing Features & Operation ..................................................................................................... 14
4.1 Primary User Interface Controls ............................................................................................. 14
4.1.1.1 Target URL – Hostname – IP Address .......................................................................................... 14
4.1.1.2 Ping Interval (sec) (Continuous Pings) ......................................................................................... 14
4.1.1.3 Test – Start/Stop Function Button ............................................................................................... 15
4.1.1.4 Reset ............................................................................................................................................ 15
4.1.1.5 # of Pings/Tests............................................................................................................................ 15
4.1.1.6 Text Message to Send .................................................................................................................. 15
4.1.1.7 Send Button ................................................................................................................................. 15
4.1.1.8 Stopping a TextPing Function ...................................................................................................... 15
4.1.1.9 Send Text Message Function ....................................................................................................... 16
4.2 Menu Selections .................................................................................................................... 17
4.2.1.1 File – Load Profile ........................................................................................................................ 17
4.2.1.2 File – New Log File ....................................................................................................................... 17
4.2.1.3 File – Open Log File ...................................................................................................................... 17
4.2.1.4 File – Save Profile ......................................................................................................................... 17
4.2.1.5 File – Save Profile As… ................................................................................................................. 17
4.2.1.6 File - Exit ...................................................................................................................................... 18
4.2.1.7 Settings ........................................................................................................................................ 18
4.2.1.8 Instructions .................................................................................................................................. 18
4.2.1.9 Help – About TextPing ................................................................................................................. 18
4.3 TextPing Icons ....................................................................................................................... 19
4.3.1.1 Load Profile Icon .................................................................................................................. 19
4.3.1.2 New Log File Icon .................................................................................................................. 19
4.3.1.3 Open Log File Icon ............................................................................................................... 19
4.3.1.4 Save Profile Icon .................................................................................................................. 19
4.3.1.5 Settings Icon ........................................................................................................................ 19
4.3.1.6 Instructions Icon .................................................................................................................. 19
5
4.3.1.7 About TextPing Icon ............................................................................................................. 19
5 The TextPing Log File ................................................................................................................... 20
5.1 TextPing Log File Contents .................................................................................................... 20
5.1.1.1 TextPing Log Startup & Settings Info ........................................................................................... 20
5.1.1.2 Active Network Interfaces ........................................................................................................... 21
5.1.1.3 Target Info & Traceroute Results ................................................................................................ 22
5.1.1.4 Test Results Header and Measurement Data Formats ............................................................... 22
5.1.1.5 Changes to TextPing Settings ...................................................................................................... 24
5.1.1.6 Test Summary Information .......................................................................................................... 24
5.1.1.7 Test Summary Information – Latency Test Only ......................................................................... 24
5.2 Detailed Network Interface Information .................................................................................. 25
5.2.1 Detailed Network Interface Info ....................................................................................... 25
6 TextPing Profiles........................................................................................................................... 26
6.1 Creating and Editing a TextPing Profile ................................................................................. 26
6.1.1 Creating a Profile ............................................................................................................ 26
6.1.2 Editing a Profile .............................................................................................................. 26
7 Instructions Form .......................................................................................................................... 28
7.1 Example Instructions ............................................................................................................. 28
7.2 Editing and Saving Instructions Text ...................................................................................... 28
7.3 Printing a Profile .................................................................................................................... 29
8 TextPing Settings .......................................................................................................................... 30
8.1 Settings Form Controls .......................................................................................................... 30
8.1.1.1 Opening the TextPing Settings Form ........................................................................................... 30
8.1.1.2 Reset Defaults .............................................................................................................................. 30
8.1.1.3 Cancel .......................................................................................................................................... 30
8.1.1.4 Save.............................................................................................................................................. 30
8.1.1.5 Save to Profile .............................................................................................................................. 30
8.1.1.6 Close ............................................................................................................................................ 31
8.2 User Settings ......................................................................................................................... 31
8.2.1.1 Enable Logging ............................................................................................................................. 31
8.2.1.2 Get Machine Info ......................................................................................................................... 31
8.2.1.3 Do Traceroute .............................................................................................................................. 32
8.2.1.4 Resolve Traceroute Hostnames ................................................................................................... 32
8.2.1.5 Enable Effective Throughput Testing .......................................................................................... 32
8.2.1.6 Test To Last Responding Node .................................................................................................... 32
8.2.1.7 Test to Closest Network Node to Target ..................................................................................... 33
8.2.1.8 Show Instructions on Profile Load ............................................................................................... 34
8.2.1.9 Default & Alternate Form Color .................................................................................................. 34
8.2.1.10 Number of Pings/Tests ................................................................................................................ 34
6
8.2.1.11 Ping Interval ................................................................................................................................. 34
8.3 Application Settings ............................................................................................................... 35
8.3.1.1 Start Traceroute At Hop #............................................................................................................ 35
8.3.1.2 Ping Timeout (ms) ........................................................................................................................ 35
8.3.1.3 Small Ping Size (Bytes) ................................................................................................................. 35
8.3.1.4 Large Ping Size (Bytes) ................................................................................................................. 36
8.3.1.5 Target – NTL Node Latency Diff (ms) ........................................................................................... 36
8.3.1.6 Don’t Fragment ............................................................................................................................ 36
8.3.1.7 Save Detailed WS Network Info to Log File ................................................................................. 37
8.3.1.8 Force Unique Text Messages ....................................................................................................... 37
8.4 Internal Settings ..................................................................................................................... 37
8.4.1.1 Un-Responsive Node Threshold .................................................................................................. 37
8.4.1.2 Bad Hop Max Threshold .............................................................................................................. 38
8.4.1.3 Large Ping Size Backoff Attempts ................................................................................................ 38
8.4.1.4 Max Time-to-Live ......................................................................................................................... 39
8.4.1.5 Inter-Ping Delay Time (ms) .......................................................................................................... 39
8.4.1.6 Thread Wait Timeout (ms)........................................................................................................... 39
9 Effective Throughput Measurements ............................................................................................ 40
9.1.1 Overview ........................................................................................................................ 40
9.1.2 Effective Throughput Formula ......................................................................................... 40
9.1.3 Practical Application of Effective Throughput Measurements .......................................... 40
9.1.3.1 Limitations of Effective Throughput Measurements .................................................................. 41
7
Table of Figures
Figure 1: TextPing Performing a Traceroute to a Target ........................................................................ 8
Figure 2: Wireshark Capture of TextPing Text Message ICMP Packets ............................................... 12
Figure 3: Wireshark Capture - Packet Summary .................................................................................. 13
Figure 4: Wireshark Capture – Data Payload Section with TextPing Messages ................................... 13
Figure 5: TextPing User Interface Controls .......................................................................................... 14
Figure 6: TextPing Function Button in Running and Stopping Mode .................................................... 16
Figure 7: TextPing Function Button in Sent and Ready to Send Modes ............................................... 16
Figure 8: TextPing Menus and Icons .................................................................................................... 17
Figure 9: About TextPing Form ............................................................................................................ 18
Figure 10: TextPing Menu and Icons ................................................................................................... 19
Figure 11: TextPing Log Startup, TextPing Settings, & Test Workstation Info ...................................... 20
Figure 12: TextPing Log Example - Active Network Interfaces ............................................................. 21
Figure 13: TextPing Log Example - Target and Traceroute Info ........................................................... 22
Figure 14: TextPing Log Test Results Header and Measurement Data Formats .................................. 23
Figure 15: TextPing Log Example - Settings Change ........................................................................... 24
Figure 16: TextPing Log Example - Test Summary Information ........................................................... 24
Figure 17: TextPing Log Example - Latency Only Test Summary Information ...................................... 24
Figure 18: TextPing Log Example - Expanded Network Interface Info ................................................. 25
Figure 19: TextPing Instructions Form Text ......................................................................................... 28
Figure 20: Instructions Form ‘Save to Profile’ Button after User Edits .................................................. 29
Figure 21: User Prompt to Create New Profile ..................................................................................... 29
Figure 22: TextPing Print Preview and Print Dialogs ............................................................................ 29
Figure 23: TextPing Settings Form....................................................................................................... 30
Figure 24: TracePing User Settings ..................................................................................................... 31
Figure 25: Test to Last Responding Node Indications .......................................................................... 33
Figure 26: TextPing Application Settings ............................................................................................. 35
Figure 27: TextPing Internal Settings ................................................................................................... 37
Figure 28: Non Responsive Node Error Message ................................................................................ 38
Figure 29: TextPing Log Example - Large Ping Backoff Algorithm Entries ........................................... 38
Figure 30: Effective Throughput vs RTT Ping Latency Graph .............................................................. 43
8
1 Introduction
Figure 1: TextPing Performing a Traceroute to a Target
TextPing™ is a Windows® application designed to enhance application test packet capture activities by a) recording CPU, Memory, OS, and network interface information from the test workstation, b) performing a traceroute to the destination host/server or network device, c) ‘marking’ the start of each application test activity in the packet capture file with a series of ICMP ‘Ping’ packets containing user-specified text messages for easy identification, and d) performing a latency and ‘Average Effective Throughput’ test on a continuous basis and/or each time a set of message packets are sent to provide information about network conditions at the time of the test.
TextPing was primarily designed to make it easier to do packet captures of a user exercising typical application functions for troubleshooting and performance analysis/modeling efforts while improving the ability to capture the natural flow of user interactions with the application, which can significantly increase the accuracy of network impact analysis and modeling. TextPing provides the ability to simply enter or select a text description of the next user activity and send a series of sequential Ping packets which can be easily located in one continuous packet capture, rather than having to stop the packet capture utility, name and save the file, and re-start the packet capture between each activity.
TextPing also gathers test workstation and traceroute information, and performs latency and Average Effective Throughput measurements during the application test activity to enhance test results.
To supplement the testing session a TextPing Profile can be pre-prepared which specifies any non-default TextPing user settings, the Target URL, Hostname, or IP addresses, a list of application test activities to perform, and (if desired) a set of Quick Reference Instructions accessible from the TextPing user interface to assist in guiding the user through the test scenario. All of the pertinent information from the test session is saved in a log file for later reference.
TextPing is a self-contained (.exe) application that can be saved to and executed from the test workstation Desktop (or preferred location) without the need for a formal installation or any other changes to the test environment (requires.NET Framework 3.5 or greater is pre-installed). After testing has been completed the TextPing.exe file, TextPing Log files, and any Profiles can simply be deleted.
9
2 TextPing Download, Installation, and Removal
2.1.1 Obtaining TextPing
The latest version of TextPing is available from the PacketIQ website:
http://www.packetiq.com/Tools/PacketIQ-TextPing.aspx
2.1.2 Installing TextPing
TextPing is a self-contained (.exe) application that can be saved to the test workstation Desktop or other location and executed without the need for a formal installation or any other changes to the workstation environment (with the possible exception that .NET Framework 3.5 be pre-installed).
You can optionally save TextPing to a specific directory (instead of the Desktop) and create a shortcut icon by right-clicking the TextPing.exe file and, while holding down the right-click button, dragging the shortcut icon to the Desktop or Quick Launch bar.
2.1.3 Removing TextPing
Since no formal installation was performed, TextPing can be removed from the workstation by simply deleting the TextPing.exe file.
10
3 TextPing Quick Reference
Note: A .pdf of these Quick Reference instructions is available at http://www.packetiq.com/Tools/TextPing.aspx
1. Double-click the TextPing icon to start the application.
2. If you were provided with a TextPing Profile file, Click 'File - Load Profile' or click the
icon and select / open the file.
Note: Additional instructions may have been provided in the Profile – click ‘Instructions’ in the TextPing Menu bar or
click the icon.
Note: You should also have a pre-prepared Test Script to guide the application testing.
3. Select or enter a Target URL, Hostname,
or IP address and click ‘Test’.
(If your Profile only included one Target it will appear - just click 'Test')
4. You will be prompted to create a TextPing Log file - accept or modify the suggested filename and click 'Save'.
5. TextPing will indicate ‘Running…’
Information about the test workstation will be collected, the Target will be resolved, and a Traceroute to the Target address will be run - this may take a few minutes.
When the traceroute completes, the button will change to ‘Start/Stop’.
6. Set the Ping Interval (sec) control to the desired number of Pings per test, and then click the ‘Start/Stop’ button to start the Continuous Pings test function. The button will turn Green and indicate ‘Running’ while performing a test, then turn Blue and indicate ‘Waiting…’ between tests. Latency and Effective Throughput measurement results will be displayed (and logged) for each test.
7. Start the packet capture application (Wireshark or similar) and ensure packets are being captured.
8. Select or enter a Text Message that describes the application test activity you are about to perform and click 'Send'.
Note: Try to ‘Send’ between Continuous Ping tests to avoid logging conflicts.
One or more ‘Ping’ packets containing the selected / entered text message (as controlled by the ‘# of Pings/Tests’ setting) will be sent to the Target. The TextPing Function button will turn Green and indicate ‘Running…’ until all the Ping packets are sent, then then Blue and indicate ‘Sent’.
9. When you observe that the Text Message packets have been sent (button turns Blue), perform the applicable application test transaction / activity per the Test Script.
Repeat Steps 8 and 9 for each application transaction / activity in the Test Script.
10. When you are finished testing the application, stop the packet capture application and save the packet trace file with a filename that describes the application, version, and the date and time the test occurred.
11. Click the Continuous Pings button to stop the interval testing. The button may turn Yellow and indicate ‘Stopping…’ for several seconds – wait until the button indicates ‘Start/Stop’. Then exit the TextPing application (File – Exit or just click the top-right ‘X’).
12. If applicable, send a) the packet trace file, and b) the TextPing Log file to your analyst/consultant; if changes were made to the TextPing Instructions and/or to a custom Profile, send a copy of the modified Profile .txt file as well.
12
3.1 Using TextPing for Marking Packet Captures
As was stated in the Introduction, TextPing was designed to make it easier to do packet captures of a user exercising typical application functions for troubleshooting and performance analysis/modeling efforts while improving the ability to capture the natural flow of user interactions with the application, which can significantly increase the accuracy of network impact analysis and modeling.
TextPing provides the ability to simply enter or select a text description of the next user activity and send a series of sequential Ping packets which can be easily located within a packet capture; this makes it practical to do one continuous packet capture rather than having to stop the packet capture utility, name and save the file, and re-start the packet capture between each application test activity.
Figure 2 is an illustration of a Wireshark capture containing TextPing Text Message (ICMP) packets; the default Wireshark coloring rules give these a Violet background:
Figure 2: Wireshark Capture of TextPing Text Message ICMP Packets
After locating the start of a specific user activity as marked by the applicable TextPing Text Message packets, and noting the end of that activity as marked by the start of TextPing packets for the next activity, the analyst can – if desired – save the range of packets for a given test activity as a separate trace file for an activity-specific analysis.
Note: The TextPing log file records the starting time of a Text Message test in millisecond resolution to make it easier to find the start of a specific series of Text Message packets within a packet trace file by referencing the Absolute DateTime timestamps from the analyzer.
Note: The time each Ping packet was transmitted is also contained in the message payload of each packet, in millisecond resolution. Note that the timestamps may differ from the test workstation clock by ~1 ms or more, and that the timestamp will not be correct in ICMP ‘reply’ packets because they will have been an echo of the ‘request’ packet by the remote host and incur the round trip network delay.
13
Figure 3 illustrates how the ICMP packets generated by TextPing are generally grouped together in a trace file.
The differing packet sizes (Pkt Size column) from the alternate Small Ping / Large Ping Requests/Responses to support Average Effective Throughput measurements can be seen, as can the relative differences in Delta Times for Small Ping responses vs Large Ping responses (~3 ms in this example).
Figure 3: Wireshark Capture - Packet Summary
Figure 4 illustrates how the user specified Text Message is contained within the ICMP Ping packets data payload.
The timestamp and Text Message in this example is “10/19/2012 18:32:33.561 PacketIQ TextPing Message: 1. Login to myApplication…” followed by a repeating pattern of “1. Login to myApplication…” text sufficient to generate the standard Large Ping total payload size of 1432 bytes (1474 Bytes – 42 Bytes for packet headers = 1432 Bytes payload).
Figure 4: Wireshark Capture – Data Payload Section with TextPing Messages
As can be seen in the illustrations above, locating and identifying TextPing message packets within a packet trace file is relatively easy and supports continuous-capture application testing.
14
4 TextPing Features & Operation
This section offers detailed information about the TextPing user controls.
The reader may wish to review the ‘TextPing Quick Reference’ in Section 3 to gain familiarity for basic operating activities and typical work flow before reading this section.
Figure 5: TextPing User Interface Controls
4.1 Primary User Interface Controls
4.1.1.1 Target URL – Hostname – IP Address
This field allows the user to enter a valid URL, Hostname, or IP address of the host/server or network device you want to send the Ping packets containing Text Messages to.
If more than one Target has been previously entered or provided in a Profile the desired target can be selected by clicking the drop-down arrow and making a selection from the list.
4.1.1.2 Ping Interval (sec) (Continuous Pings)
This controls the interval between Effective Throughput tests (or ‘Ping’ Latency measurements if Effective Throughput testing is disabled) in seconds.
The number of Ping packets (latency only) or sets of Small and Large Pings for Effective Throughput measurements that are sent for each test interval is controlled by the ‘# of Pings/Tests’ control (discussed below).
Note: Be aware that short Ping Interval settings coupled with high Number of Pings/Tests settings can generate a significant amount of network traffic.
15
4.1.1.3 Test – Start/Stop Function Button
This button is used to initiate a ‘Test’ of a new URL – Hostname – IP Address target (resolve hostname and perform a traceroute) at the beginning of a test session (or after changing the Target), after which it controls the Start/Stop function of Continuous Pings testing.
Note: A traceroute can take from several seconds up to several minutes to complete, depending on the length of the path being recorded and how many ‘hops’ respond in a timely fashion (or fail to respond). Unless disabled, a traceroute should be allowed to complete before attempting to send Text Messages to support other TextPing measurement functions.
Note: Traceroute, Average Effective Throughput, and latency measurement functions are run in a separate ‘thread’ so that the TextPing user interface does not ‘lock up’ while these functions are active.
4.1.1.4 Reset
This button clears all of the Target and Text Message entries that may have been residing in the drop-down fields and restores TextPing to a normal operating condition in case of unreliable operation. If this fails to resolve an errant condition, exit and re-start TextPing.
4.1.1.5 # of Pings/Tests
This controls how many Effective Throughput tests are to be conducted, or if Effective Throughput measurements are disabled, how many ICMP ‘Ping’ packets get sent to the Target for each test iteration when Continuous Pings are running, and/or each time the user enters or selects a Text Message and clicks ‘Send’.
Note: Be aware that short Ping Interval settings coupled with high Number of Pings/Tests settings can generate a significant amount of network traffic.
4.1.1.6 Text Message to Send
Text entered (or selected from a list) in this field is sent as a text ‘payload’ in ICMP ‘Ping’ packets to the Target when the user clicks the ‘Send’ button.
4.1.1.7 Send Button
The TextPing Function button’s background color, label, and specific function will vary depending upon what state the current user interaction with TextPing is in – usually ‘Send’ or ‘Sent’ – but is primarily used to ‘Send’ a Text Message for marking packet trace files.
4.1.1.8 Stopping a TextPing Function
Clicking one of the two Function buttons while information gathering or test functions are taking place will result in the Function button turning Yellow and indicating a ‘Stopping…’ condition; the traceroute or measurement activity will be aborted, and when the currently running thread completes the test it was running the button will return to a normal background color and indicate ‘Test’, ‘Start/Stop’, or ‘Send’, depending on the button and the position in the test session. At this point the information gathering (traceroute,
16
Continuous Pings, or Text Message Send function can be repeated by clicking the appropriate Function button again.
Figure 6: TextPing Function Button in Running and Stopping Mode
Note: Pressing ‘Enter’ will invoke the Function button (as if it were clicked); which button is activated depends on which entry field (Target or Text Message) was last entered.
4.1.1.9 Send Text Message Function
When the information gathering and traceroute completes, the Text Message Function button will indicate ‘Send’ and a message under the Text Message field will appear; the specific message contents will vary depending on whether the user entered a message, selected the new message from a list, etc.
Clicking ‘Send’ will cause TextPing to send the specified number of Ping packets containing the Text Message text and perform latency and (if enabled) Average Effective Throughput measurements. While Ping packets are being sent the button will turn Green and indicate ‘Running…’ (and can be stopped by clicking the Function button again); when all the packets have been sent the Function button will contain a Blue background and indicate ‘Sent’. When the user enters or selects the next message the Function button returns to a normal background color and indicates it is ready to ‘Send’ again. This process can be repeated for all of the desired application test activities.
Figure 7: TextPing Function Button in Sent and Ready to Send Modes
17
Note: By default, TextPing does not allow the same Text Message to be sent a second consecutive time; the text must be altered or a different message selected. This is to prevent the user from mistakenly sending the same message again during Test Script-driven testing. This feature can be disabled if desired in Settings – Application Settings: disable the ‘Force Unique Text Messages’ option and click ‘Save’.
4.2 Menu Selections
TextPing Menu options are reasonably intuitive; a description of each is provided below.
Figure 8: TextPing Menus and Icons
4.2.1.1 File – Load Profile
Allows loading a pre-prepared Profile which can optionally contain specific TextPing settings, one or more Target URLs, Hostnames, or IP addresses, multiple Text Messages representing application test activities, and Instructions to guide the user through the test activities.
4.2.1.2 File – New Log File
Allows the user to create a new log file. This can be performed before a test scenario, or a new log file can be created between testing of multiple applications.
Note: When a new log file is created the information gathering function (resolve the target, get test workstation information, and a traceroute) will be initiated upon the next user activity so that the new log file contains this crucial information.
4.2.1.3 File – Open Log File
When TextPing has just been launched a previously generated log file can be selected instead of creating a new one.
The information gathering function (resolve the target, get test workstation information, and perform a traceroute) will be initiated upon the next user activity so that the log file contains current environment information.
4.2.1.4 File – Save Profile
If a Profile was not initially loaded, selecting this option will prompt the user to select or edit a suggested location and filename for a new Profile. If a Profile was originally loaded, the current TextPing settings, Target(s), Text Message(s), and any changes to the Instructions text will be saved to the currently-loaded Profile (the Profile will be over-written/replaced).
4.2.1.5 File – Save Profile As…
Selecting File – Save Profile will prompt the user to select or edit a suggested location and filename for a new Profile.
Profile files have a .txt extension to allow easy editing by an ANSI text editor such as NotePad.
18
4.2.1.6 File - Exit
Exits TextPing and saves the Test Summary information to the log file (if applicable).
4.2.1.7 Settings
Opens the TextPing Settings form. See Section 8 - TextPing Settings for more information.
4.2.1.8 Instructions
Clicking the Instructions menu item opens the Instructions form. See Section 7 - Instructions Form for more information.
4.2.1.9 Help – About TextPing
Clicking Help – About TextPing displays the ‘About TextPing’ form. The Version and Version Date of the currently running copy of TextPing can be obtained from this form.
Figure 9: About TextPing Form
Note: Clicking the ‘PacketIQ website’ link from the ‘About TextPing’ form will open a browser window to the TextPing webpage so the user can check for updated information and versions of TextPing and/or obtain a current version of this User Guide or a TextPing Quick Reference Guide.
19
4.3 TextPing Icons
Figure 10: TextPing Menu and Icons
4.3.1.1 Load Profile Icon
Allows loading a pre-prepared TextPing Profile which can optionally contain specific TextPing settings, one or more Target URLs, Hostnames, or IP addresses, multiple Text Messages representing application test activities, and Instructions to guide the user through the test activities.
4.3.1.2 New Log File Icon
Allows the user to create a new log file. This can be performed before a test scenario, or a new log file can be created between testing of multiple applications.
Note: When a new log file is created the information gathering function (resolve the target, get test workstation information, and a traceroute) will be initiated upon the next user activity so that the new log file contains this crucial information.
4.3.1.3 Open Log File Icon
The Open Log File icon provides an easy method of opening / viewing the current log file.
Alternatively, when TextPing has just been launched a previously generated log file can be selected instead of creating a new one; the information gathering function (resolve the Target, get test workstation information, and perform a traceroute) will be initiated upon the next user activity so that the log file contains current environment information. This new information and any additional Text Msgs / tests will be appended to the selected log file.
4.3.1.4 Save Profile Icon
If a Profile was not initially loaded, selecting this option will prompt the user to select or edit a suggested location and filename for a new Profile. If a Profile was originally loaded, the current TextPing settings, Target(s), and Text Message(s) will be save to the currently-loaded Profile (the Profile will be over-written).
4.3.1.5 Settings Icon
Opens the TextPing Settings form. See Section 8 - TextPing Settings for more information.
4.3.1.6 Instructions Icon
Opens the Instructions form. See Section 7 - Instructions Form for more information.
4.3.1.7 About TextPing Icon
Displays the ‘About TextPing’ form. See Section 4.2.1.9 Help – About TextPing for more information.
20
5 The TextPing Log File
The TextPing log file contains a wide range of information about the application and test workstation environment, as well as latency and Average Effective Throughput measurement results for each set of user Text Messages sent. The specifics of each section will be covered in the following sections.
5.1 TextPing Log File Contents
5.1.1.1 TextPing Log Startup & Settings Info
The TextPing Version, file location, the DateTime the log file was created, and the test workstation Time Zone are recorded at the top of the log file, followed by the current TextPing User, Application, and Internal settings, and test workstation System Info.
Figure 11: TextPing Log Startup, TextPing Settings, & Test Workstation Info
21
5.1.1.2 Active Network Interfaces
This section lists the interface types and configuration information on just the active network interfaces.
Note: Additional info on active as well as inactive interfaces can be logged by enabling the ‘Save Detailed WS Network Info to Log File’ option in Application Settings.
Figure 12: TextPing Log Example - Active Network Interfaces
22
5.1.1.3 Target Info & Traceroute Results
During the TextPing information gathering function (after clicking ‘Test’) the specified Target URL, Hostname, or IP Address is resolved, and the resolved IP address is provided to a DNS query to provide the DNS information in case it’s relevant, as the DNS resolved hostname may be different than the initial hostname provided, as exemplified below.
Figure 13: TextPing Log Example - Target and Traceroute Info
Note: The traceroute information saved to the log file is provided in Windows format, and a ‘tracert <IP Address> command line entry included, to allow a simple copy/paste interaction with the PacketIQ TraceRoute Resolver. See the Tools section of www.packetiq.com for more information on the TraceRoute Resolver.
5.1.1.4 Test Results Header and Measurement Data Formats
Individual latency and Effective Throughput measurement values are logged in Comma Separated Value (.csv) format to support importing the data into other analysis tools such as Excel®. The measurement entries follow the same format regardless of whether they were originated by the Continuous Ping or Text Message Send functions.
Following the completion of the traceroute for a new target, TextPing creates a TEST RESULTS line entry in the log file, under which is logged a set of headers (also in CSV format) for the test results data. The headers include:
Timestamp,TargetIPAddr,Lat1MS,Lat2MS,EffTPKbps
23
Timestamp is the DateTime (with millisecond resolution) that the first ‘Ping’ was sent for that test.
TargetIPAddr is the IP address of the target node the Ping packets were sent to.
Lat1MS is the Small Ping packet latency.
Lat2MS is the Large Ping packet latency (if Effective Throughput measurements are enabled).
EffTPKbps is the calculated Effective Throughput value in Kbps for that measurement.
Note: If Latency only tests are run, the Lat2MS and EffTPKbps entrys will be blank.
The figure below illustrates test data from an initial Text Message activity (5 measurements), along with the Text Message Summary data, followed by two sets of 5 measurements in 15 second intervals from Continuous Pings.
Figure 14: TextPing Log Test Results Header and Measurement Data Formats
Note: If the Small Ping latency is larger than the Large Ping latency for a given Effective Throughput test, the ‘EffTPKbps’ entry will be labelled with ‘N/A’ This occurs when the short-term network loading or responding device utilization happen to be higher at the instance the Small Ping was received than for the Large Ping.
24
5.1.1.5 Changes to TextPing Settings
If any settings are changed mid-test, TextPing logs the DateTime stamp and the changed setting(s) in the log file at the point where the changes are made.
Figure 15: TextPing Log Example - Settings Change
5.1.1.6 Test Summary Information
When the user stops a Continuous Pings test, or the TextPing application is exited (the user clicks File – Exit or the top-right-hand ‘X’ on the user interface), or a new log file is created, TextPing will log the summarized latency and Average Effective Throughput test results for all of the Text Message transmittals performed so far. The DateTime the TextPing application or log was exited is also indicated if appropriate.
Figure 16: TextPing Log Example - Test Summary Information
5.1.1.7 Test Summary Information – Latency Test Only
If ‘Enable Effective Throughput Testing’ has been disabled in TextPing Settings the format of the logged information includes the Minimum, Average, and Maximum latency values for all of the tests performed.
Figure 17: TextPing Log Example - Latency Only Test Summary Information
25
5.2 Detailed Network Interface Information
If ‘Save Detailed WS Network Info to Log File’ option is enabled, TextPing logs additional information about test workstation interfaces.
5.2.1 Detailed Network Interface Info
The contents of detailed network interface information depend on the type of interfaces being recorded; an example of expanded information for a Bluetooth and Wireless interface are shown below.
Figure 18: TextPing Log Example - Expanded Network Interface Info
26
6 TextPing Profiles
A ‘Profile’ file can be pre-prepared (or created from a previously completed test session) and loaded into TextPing at the start of a test session to provide one or more of the following:
Configuration of TextPing settings to meet the test scenario requirements.
A Target URL, Hostname, or IP address (or a list of Targets to choose from).
A list of Test Messages that can be sequentially selected and ‘Sent’ during the test.
A customized set of Quick Reference Instructions that opens for viewing automatically when loading the profile, and/or that the user can view (and/or print) by clicking the ‘Instructions’ menu option or Instructions icon.
6.1 Creating and Editing a TextPing Profile
6.1.1 Creating a Profile
The easiest method of creating an initial Profile that can be customized to specific testing scenarios is to generate the file from TextPing itself.
1. Start TextPing.
2. Click ‘Settings’ and configure the desired User, Application, and Internal settings.
3. Open the Instructions form and make / edit any entries (if desired).
4. Click ‘Save to Profile’.
5. Accept or edit the suggested filename and location and click ‘Save’, then ‘Close’.
6.1.2 Editing a Profile
6. Open a newly created (or previously created) TextPing Profile file with a text editor.
7. Add or edit any non-default TextPing User, Application, and/or Internal Settings; for example (these settings are not suggested, but only used here for illustration):
# TextPing Settings
# User Settings ResolveTracerouteHostnames=False
# Application Settings ForceUniqueTextMsgs=False UseUTCTimestamps=True
# Internal Settings InterPingTimeMs=10
8. Add or edit the desired Target entries between the indicated ‘START’ and ‘END’ tags; for example, if your application URL was myApplication.myCorporation.com:
# Target URLs - Hostnames - IP Addresses # Edit the entries between the 'START' and 'END' tags.
START TARGET URLS – HOSTNAMES – IP ADDRESSES myApplication.myCorporation.com <= Add your Target(s) here END TARGET URLS – HOSTNAMES – IP ADDRESSES
Additional Targets can be added within the START and END tags, one per line.
9. Add or edit the desired Text Messages you want the user to Send during the application testing, one Text Message per line, between the ‘START’ and ‘END tags:
27
# Test Script Items - Text Messages # Edit the entries between the 'START' and 'END' tags. START TEST SCRIPT ITEMS - TEXT MESSAGES 1. Login to 'myApplication' 2. Select 'Employee Records' 3. Type Boss's last name in the 'Search' field and click 'Search' 4. Select Boss's name and click 'Edit Employee' 5. Click 'Terminate Employee', then click 'Confirm Termination' 6. Click 'Notify Employee of Status Changes' 7. Select Notification Options 'Email', 'Facebook', & 'Twitter', then click 'Send Notification' 8. Click 'Exit', then 'Log Out' END TEST SCRIPT ITEMS - TEXT MESSAGES
10. Finally, add or edit any Instructions to be presentable to the user by clicking the ‘Instructions’ menu option or icon, again by added line items to between the ‘START’ and ‘END’ tags:
# Quick Reference Instructions # Edit the entries between the 'START' and 'END' tags. START INSTRUCTIONS TEXT TextPing Quick Reference Instructions ============================================================================ 1. Double-click the TextPing icon to start the application. 2. If you were provided with a TextPing Profile file, Click 'File - Load Profile' (or click the first icon (Load Profile)) and select / open the file (TextPing Profile < app name> .txt or similar). Note: You should also have a pre-prepared Test Script to guide the application testing. 3. Select or enter a Target URL, Hostname, or IP Address and click 'Test'. ( If your Profile only included one Target it will appear - just click 'Test' ) 4. You will be prompted to create a TextPing Log file - accept or modify the suggested filename and click 'Save'. 5. TextPing will indicate ‘Running…’ … (omitted for brevity) 10. When you are finished testing the application, stop the packet capture application and save the packet trace file with a filename that describes the application, version, and the date and time the test occurred. 11. Click the Continuous Pings button to stop the interval testing. The button may turn Yellow and indicate ‘Stopping…’ for several seconds – wait until the button indicates ‘Start/Stop’. Then exit the TextPing application (File – Exit or just click the top-right 'X'). 12. If applicable, send a) the packet trace file, and b) the TextPing Log file to your analyst/consultant; if changes were made to the TextPing Instructions and/or to a custom Profile, send a copy of the modified Profile .txt file as well. END INSTRUCTIONS TEXT
11. Alternatively, a new/custom Profile can be generated by copy/pasting the desired sections from an existing profile and modifying the specific entries as appropriate for the application.
Note: A TextPing Profile file should only be edited by an ANSI text editor such as NotePad to ensure that no extraneous formatting characters are included which could result in unreliable operation. If it is necessary to edit this file with other applications care should be taken to save the file with text (.txt) formatting only, and a .txt extension.
28
7 Instructions Form
The Instructions form text offers the user quick-reference guidance with using TextPing to support application test packet captures, or can be modified (within a Profile) to allow taking notes during the application or other testing scenarios.
7.1 Example Instructions
An example of using the TextPing Instructions form is illustrated in Figure 19 below:
Figure 19: TextPing Instructions Form Text
7.2 Editing and Saving Instructions Text
The text in the Instructions form can be edited by the user to modify or supplement the default instructions (or the instructions provided in a custom Profile) and the changes saved to the existing or a new Profile.
If modifications are made to the instructions, the ‘Save to Profile’ button will turn Yellow:
29
Figure 20: Instructions Form ‘Save to Profile’ Button after User Edits
If a custom profile has been provided and the ‘Save to Profile’ button is clicked after the user has edited the text, the new text will be saved to the Profile (the existing Profile will be over-written). If a Profile was not previously opened, the user will be prompted to create a new Profile:
Figure 21: User Prompt to Create New Profile
7.3 Printing a Profile
The text contents of the Instructions form can be viewed before printing by clicking ‘Print Profile, and/or printed using the standard print dialog by clicking the ‘Print’ button.
Figure 22: TextPing Print Preview and Print Dialogs
30
8 TextPing Settings
Figure 23: TextPing Settings Form
8.1 Settings Form Controls
8.1.1.1 Opening the TextPing Settings Form
TextPing Settings can be viewed and changes made to the default settings by clicking the
‘Settings’ Menu selection or the Settings icon:
Note: Changing any of the TextPing settings will turn cause the ‘Save’ and ‘Save to Profile’ buttons to turn Yellow to remind the user to save the settings before closing the form.
Note: If the User changes one or more settings then clicks ‘Save’ or ‘Save to Profile’, the Close button will turn Green to indicate a successful ‘Save’ and the next action.
8.1.1.2 Reset Defaults
All TextPing settings can be returned to their default values by clicking the ‘Reset Defaults’ button. You must click the ‘Save’ or ‘Save to Profile’ buttons for the settings to take effect.
8.1.1.3 Cancel
Cancels any settings that were changed and closes the Settings form.
8.1.1.4 Save
Clicking ‘Save’ causes any changes to the settings to take effect in the TextPing application. Clicking this button does not close the Settings form.
8.1.1.5 Save to Profile
Clicking ‘Save to Profile’ saves the new settings to the TextPing application and the currently loaded Profile file. If a Profile was not previously loaded, the user will be prompted to create one. Clicking this button does not close the Settings form.
Note: Only settings whose values differ from TextPing defaults will be saved in a Profile.
31
Note: If ‘Save to Profile’ is clicked, and the user is prompted to create a new Profile but clicks ‘Cancel’, the changed settings will still take effect in the TextPing application.
8.1.1.6 Close
Clicking this button closes the form. If the user has made any setting changes and clicks ‘Close’ without first clicking ‘Save’ or ‘Save to Profile’, the changes will be discarded.
8.2 User Settings
The TextPing User Settings control the primary application functions and user interface settings that may be changed by the user.
Figure 24: TracePing User Settings
8.2.1.1 Enable Logging
TextPing was designed to collect and store information about the test workstation, the desired Target IP address and hostname, Traceroute results, the content of test messages, and the results of latency and average effective throughput measurements in a log file.
In cases where this is unsupportable or undesired TextPing can be operated without the logging function by disabling the Enable Logging setting.
Default Setting: Enabled [Recommended]
Note: Disabling the logging function will automatically disable other functions which rely upon the logging capability for practical functionality, including ‘Get Machine Info’ and ‘Save Detailed WS Network Info to Log File’.
Note: Even if Logging is disabled, a traceroute will still be performed (unless disabled) as information from this activity is needed to support the ‘Test to Last Responding Node’ and ‘Test to Closest Network Node to Target’ functions/options.
8.2.1.2 Get Machine Info
By default, the ‘Get Machine Info’ function attempts to gather and log the following basic information about the test workstation:
ComputerName
UserName
32
System Operating System, including the OS Architecture (32/64 bit) and any Service Packs
Processor Type
Total System Memory
Active Network Interfaces
If this function causes unreliable or fatal TextPing operation it can be disabled.
Default Setting: Enabled
8.2.1.3 Do Traceroute
TextPing typically performs a Traceroute to the Target node during the ‘Test’ function. This Traceroute information is logged for later reference, and is also utilized to support other TextPing functionality such as ‘Test to Last responding Node’ and ‘Test to Closest Network Node to Target’.
If Traceroute information is not desired, this function can be disabled.
Default Setting: Enabled
Note: Disabling ‘Do Traceroute’ also disables functions which rely on Traceroute information including ‘Test to Last Responding Node’, and ‘Test to Closest Network Node to Target’.
8.2.1.4 Resolve Traceroute Hostnames
By default, TextPing will perform a DNS query to resolve the IP address of each node in the Traceroute results. If this information is not desired, this function can be disabled to speed up Traceroute completion, particularly when a number of the nodes cannot be resolved.
Default Setting: Enabled
8.2.1.5 Enable Effective Throughput Testing
During the ‘Send’ function, TextPing first sends a small payload Ping packet, then sends a large payload Ping packet; this Small Ping / Large Ping scenario is repeated the number of times required to match the value specified by the ‘# of Pings’ setting on the TextPing main form or User Settings.
The measured Round Trip Time (RTT) latency difference between the Small and Large Ping return times is utilized to calculate an approximate Average Effective Throughput value – this value is displayed and logged. If the Average Effective Throughput measurements fail to function properly or are undesired this option can be disabled.
Default Setting: Enabled
Note: See Section 9 - Effective Throughput for more information.
8.2.1.6 Test To Last Responding Node
In some cases a Ping response cannot be obtained from the Target node/host/server because either a firewall is blocking ICMP (Ping) requests or the target itself has been configured to not respond. When the traceroute is conducted at the beginning of the test, TextPing keeps track of the last node in the traceroute path that responds to Ping requests; if the target cannot be reached TextPing will conduct latency and Average Effective Throughput tests to the last responding node instead. This contingency can be detected by observing that TextPing reports the Last Responding Node (which will be different than the
33
Resolved Target Address) at the end of the traceroute, and when Text Messages are sent TextPing reports that it is ‘Testing to last responding node:’ plus the node IP address.
Figure 25: Test to Last Responding Node Indications
This option attempts to address the need to assess network conditions in the presence of a firewall or non-responding target node without resorting to sending TCP or UDP packets to a specified port on the target by performing the ICMP testing to the last responding node; if the application is being tested from a remote user location to a host/server located in a major data center, the last responding node is often relatively close (latency-wise) to the Target node, so the latency measurements will accurately reflect the network path delay (since the firewall to server latency will typically be < 1 ms) and the average network throughput (since the most bandwidth constricting element will be the WAN link(s) between user and data center. This offers a best-effort measurement of network conditions.
Default Setting: Enabled
Note: Be aware that some network routers may respond properly to an ICMP TTL Expired event during a traceroute, but still not respond to ICMP request packets – in these cases tests to the ‘Last Responding Node’ will fail. It may be necessary to review the traceroute results (in the TextPing log) to identify a node closest to the target that will respond properly.
8.2.1.7 Test to Closest Network Node to Target
In many application testing scenarios the Target may be the host/server that is actually hosting the application. Since this device may be quite busy servicing numerous concurrent users, and conjecturing that responding to ICMP ‘Ping’ requests may not be a very high priority to an application server, performing latency and Average Effective Throughput testing to a network node (router or Layer 3 switch) close to the application server may offer higher Ping response reliability and measurement accuracy.
While performing the traceroute to the Target node at the onset of testing, TextPing keeps track of the next-to-last responding node to the Target. If minimum Ping latencies and/or Average Effective Throughput measurements seem to vary significantly (or fail frequently) across several tests it may be prudent to enable this option to see if these measurements improve.
Default Setting: Disabled
34
8.2.1.8 Show Instructions on Profile Load
If this featured is enabled, an Instructions form will appear automatically when a Profile is opened to offer the user/tester guidance on using TextPing and performing the desired testing. The instructions can be customized and pre-prepared in the Profile, and/or the user can edit / add comments and save the modified text back into the Profile by clicking File – Save Profile or clicking the Save Profile icon.
Note: See Section 6 - TextPing Profiles for details on configuring Profiles.
The option to show the Instructions window after loading a Profile can also be enabled by including the entry ‘ShowInstructionsOnProfileLoad=True’ in the Profile. This can also be accomplished by enabling this option and saving the existing Profile or creating a new one.
Default Setting: Disabled
8.2.1.9 Default & Alternate Form Color
This setting allows selection of an ‘Alternate Form Color’ which may be more pleasing to a member of the feminine gender that may have been coerced into performing application testing. Default Setting: Default
8.2.1.10 Number of Pings/Tests
This setting controls the number of Ping packets sent during each Continuous Pings test or when a Text Message Send function is activated.
This setting must be balanced between sending enough Pings to allow an analyst/consultant to easily find the Ping packets containing the Text Message and to increase the accuracy of Average Effective Throughput measurements, without overly burdening the network and the target device that must respond to each Ping.
Range: 1 to 32 Default Setting: 5 Maximum Recommended Setting: 10
Note: The ‘Test Start DateTime’ entry in the TextPing log file for each Text Message reflects the time to 1 millisecond resolution to aid an analyst in locating the start of the Ping packets in a packet trace file by the packet timestamps. The relevant accuracy of this DateTime stamp assumes that TextPing is running on the same test workstation as the packet capture application (such as Wireshark).
8.2.1.11 Ping Interval
The Ping Interval control setting controls the interval – in seconds - between successive latency and Effective Throughput tests when Continuous Pings are running.
Range: 1 to 3600 (1 second to 1 hour) Default Setting: 15 seconds
Note: Be aware that short Ping Interval settings coupled with high Number of Pings/Tests settings can generate a significant amount of network traffic.
35
8.3 Application Settings
The TextPing Application Settings control the application settings that may be changed by the user to affect Traceroute, Average Effective Throughput, and Logging options.
Figure 26: TextPing Application Settings
8.3.1.1 Start Traceroute At Hop #
This setting can be incremented to skip the first several nodes of a traceroute to avoid known non-response or other issues close to the test workstation.
Range: 1 to 128 Default Setting: 1
8.3.1.2 Ping Timeout (ms)
This setting controls how long TextPing will wait (in milliseconds) for a response to an ICMP Ping request. Setting this value too low may cause missed responses from busy nodes over long network paths; setting this value high will cause traceroutes and Pings to non-responding nodes to consume more time before recording a timeout.
Range: 10 to 120,000 (.1 to 120 seconds) Default Setting: 1,000
8.3.1.3 Small Ping Size (Bytes)
This setting controls the data payload (Text Message) size (in Bytes) in the small Pings utilized for Average Effective Throughput measurements, or normal latency measurements if Average Effective Throughput testing is disabled. The format of the text within the small Ping packet payload text can take several forms, depending on the size of the user Text Message and this control setting.
A setting of zero (0) causes a small Ping text payload to be constructed that includes a timestamp and the text ‘PacketIQ TextPing Message: ‘ plus the user Text Message plus a trailer consisting of four periods ‘....‘ For example, if the user Text Message is ‘1. Login to MyApplication’, the resultant Ping payload text could be ‘10/18/2012 11:21:15.379 PacketIQ
TextPing Message: 1. Login to MyApplication….’.
If this setting is greater than zero (0) but smaller than a value that would support the combined message structure for the (0) setting then the packet payload text will consist of only the user Text Message plus enough trailing periods to match the payload size setting, or as much of the Text Message as needed to match the setting.
If this setting is greater than the size of the zero (0) setting text payload construct the zero (0) scenario text message (“10/18/2012 11:21:15.379 PacketIQ TextPing Message: “ + Text
36
Message + “….”) is constructed, followed by a repeated pattern of just the user Text Message and four periods sufficient to match the small Ping payload size setting or a subset thereof.
Range: 0 or 1 to 511 Default Setting: 0 [0 = controlled by Text Message Size]
8.3.1.4 Large Ping Size (Bytes)
This setting controls the data payload (Text Message) size (in Bytes) in the large Pings utilized for Average Effective Throughput measurements. The Ping packet payload text consists of the same construct as used in Small Ping payloads (“10/18/2012 11:21:15.379
PacketIQ TextPing Message: “ + Text Message + “….”) followed by a repeating pattern of just the user Text Message and four periods sufficient to match the large Ping payload size setting.
Range: 256 – 1472 Bytes Default Setting: 1432
Note: The maximum Ethernet packet size for typical 10/100/1000 Mbps Local Area Networks is limited to 1518 bytes consisting of 42 bytes for Ethernet(14) + IP(20) + ICMP(8) headers + 1472 bytes ICMP (Text Message) payload + 4 bytes CRC = 1518, thus the maximum recommended Large Ping Size setting is 1472 bytes.
8.3.1.5 Target – NTL Node Latency Diff (ms)
This setting controls a latency difference threshold between the Target node and the next-to-last node recorded by TextPing during the traceroute (based on the average of the three Traceroute latency measurements), and supports / works in conjunction with the “Test to Closest Network Node to Target” control in the User Settings.
This setting is normally set to its maximum value of 100 milliseconds so that its effect is negated and the ‘Test to Closest Node…’ function is basically controlled by whether that User Settings control is enabled.
However, if this setting is set to a lower value – for example, 1 ms – and the ‘Test to Closest Node…’ function is enabled, TextPing will determine if it should test to the ‘Closest’ node or the Target node depending on whether the Closest node is within the threshold set by this value.
For example, if the ‘Test to Closest Node…’ option is enabled, the Target – NTL Node Latency Diff value is set to 1 ms, and the closet (next to last) node to the Target has an average Traceroute latency that is within 1 ms of the Target, then testing will take place to the Closest node. Otherwise, TextPing will ignore this setting and test to the Target exclusively.
Range: 0.0 – 100.0 (milliseconds) Default Setting: 100.0
8.3.1.6 Don’t Fragment
Every network has a MTU (Maximum Transmission Unit) which is the size of the largest packet that can be transmitted; packets larger than the allowable MTU must be divided into smaller packets or fragments to enable them to traverse the network.
The ‘Don’t Fragment’ option (if enabled) disallows this fragmentation which may cause TextPing to receive a response to small Ping packets but not the large ones because the large packets will not be able to traverse the network. This failure to respond to large packets only (when ‘Don’t’ Fragment’ is enabled) is an indicator to the analyst that there may be a packet size limited segment of the network in the path between the test workstation and the target node.
37
Note: Network segment which cause large packets to be fragmented (usually to 512 bytes) may adversely affect application response time performance due to the additional time it takes to fragment / transmit / re-assemble multiple large packets. These scenarios, if discovered, should be investigated.
Default Setting: Enabled [Don’t allow fragmentation]
8.3.1.7 Save Detailed WS Network Info to Log File
By default TextPing saves summary information on just the network interfaces in the test workstation which are active / operational to the TextPing log file. Enabling this option causes TextPing to record a more detailed set of data on all of the interfaces whether they are operational or not.
Default Setting: Disabled
8.3.1.8 Force Unique Text Messages
By default, TextPing does not allow the same Text Message to be sent a second consecutive time; the text must be altered or a different message selected. This is to prevent the user from mistakenly sending the same message again during Test Script-driven testing. This feature can be disabled if desired by disabling this option.
8.4 Internal Settings
Figure 27: TextPing Internal Settings
8.4.1.1 Un-Responsive Node Threshold
This setting affects the number of times a non-responsive node will be tested during Average Effective Throughput or latency testing. If the target node doesn’t respond to three attempts (if using the default setting), the test will be aborted and the error message depicted below will be presented to the user.
Assuming the target node was believed to be reachable, the user should check all other testing presumptions or select another node that will respond.
38
Figure 28: Non Responsive Node Error Message
Range: 1 - 128 Default Setting: 3
8.4.1.2 Bad Hop Max Threshold
This setting controls the number of times TextPing will attempt to get a response from a non-responsive node during a traceroute. For example, with the default setting of ‘3’, if a particular node along the traceroute path (controlled by the Time-to-Live value) doesn’t respond to three attempts to get a response from that node the node will be skipped, logged with a ‘Request Timed Out’ message for that hop, and the traceroute will continue with the next hop.
Range: 1 – 10 Default Setting: 3
8.4.1.3 Large Ping Size Backoff Attempts
In some cases a network node will respond to smaller Ping packets but not to the larger Ping packets used to conduct Average Effective Throughput measurements. When a response has been received to the small Ping packet but not the large, TextPing will reduce the size of the large Ping packets by 25% and try again. This setting controls the number of large Ping packet size reductions that are attempted before aborting the test.
Figure 29: TextPing Log Example - Large Ping Backoff Algorithm Entries
The number of attempts TextPing will make to get a response to a large Ping before executing the back-off algorithm is the Un-Responsive Node Threshold setting (Settings – Internal Settings) minus 1. With the default setting of 3, TextPing will attempt to get a response to a large Ping 2 times before executing the back-off sequence.
This situation may also be caused by the presence of a segment in the network path between the test workstation and the target node that cannot pass larger Ping packets do to a restricted MTU (Maximum Transmission Unit) size without fragmenting the large packets into smaller sizes (512 Bytes, for example). Since the default TextPing setting is to set the Don’t Fragment value in the IP header, these packets cannot be passed and will be
39
dropped. If this large packet failure situation is detected by the user (by comparing the large Ping packet sizes in the log file with the Large Ping Size (Bytes) setting in the Application Settings) it may be prudent to disable the ‘Don’t Fragment’ setting and see if this resolves the problem. Success with the ‘Don’t Fragment’ option disabled suggests that the limiting segment needs to be identified and possibly rectified to avoid the effects this situation may have on application performance.
Note: If the back-off algorithm is successful, the largest MTU that can be accommodated by the network path will be a value somewhere between the successful Large Ping packet size and the last failing Large Ping packet size (check the TextPing log file for these values).
Range: 1 – 10 Default Setting: 3
8.4.1.4 Max Time-to-Live
This setting controls the Time-To-Live (TTL) value configured in the ICMP ‘Ping’ packets sent by TextPing during Traceroute and Send Message functions.
When a router receives an ICMP Ping packet who’s Time-to-Live has expired, an ICMP response packet is returned to the sender indicating ‘TTL Expired’ and the packet is not forwarded to the next node in the path. The TextPing TTL setting therefore affects the greatest number of network ‘hops’ a Traceroute can record, and the greatest number of hops between the test workstation and the Target node.
Range: 1 to 128 Default Setting: 64
8.4.1.5 Inter-Ping Delay Time (ms)
This setting controls the amount of time (in milliseconds) TextPing waits between transmitting successive Ping packets during traceroutes, and between Ping packets sent during Average Effective Throughput and/or latency tests containing user Text Messages.
Increasing the inter-ping delay time will reduce the network loading effects of these activities, but can significantly increase test completion times and reduce Average Effective Throughput measurement accuracy.
Range: 0 – 60,000 Default Setting: 0
8.4.1.6 Thread Wait Timeout (ms)
Unused and not applicable in this version.
40
9 Effective Throughput Measurements
9.1.1 Overview
During normal operation, when a Text Message is selected or entered and the ‘Send’ button is clicked, TextPing sends a series of ICMP ‘Ping’ packets containing the user Text Message. During the ‘Send’ function, TextPing sends / receives a Small payload Ping packet, then send / receives a Large payload Ping packet; this Small Ping / Large Ping scenario is repeated the number of times specified by the ‘# of Pings’ setting on the user interface or the Settings form.
The latency difference between the Small and Large Ping packet Round Trip Times (caused by the difference in how long it takes to send/receive a Large packet vs a Small one across the network) is utilized to calculate an approximate Effective Throughput value (one value per Small/Large Ping test); the results are logged and displayed in the user interface.
9.1.2 Effective Throughput Formula
PacketIQ derived and utilizes this formula to calculate Average Effective Throughput values:
Effective Throughput (Kbps) = (Ping Payload Size Difference (Bytes) * 8) / (Ping Latency Difference (milliseconds) / 2)
Example:
Small Ping payload size: 32 Bytes Large Ping payload size: 1432 Bytes
If a set of Small and Large Ping tests are performed and the average RTT times were:
Small Ping RTT: 66 ms Large Ping RTT: 75 ms
The Average Effective Throughput would be calculated as follows:
Eff TP (Kbps) = (Large Ping Payload Bytes – Small Ping Payload Bytes) * 8 (Large Ping RTT – Small Ping RTT) / 2
= ((1432 – 32) * 8) / ((75 – 66) / 2)
= (1400 * 8) / (9 / 2)
= (11,200 / 4.5)
= 2,488.89 Kbps = 2.49 Mbps
By enabling the ‘Save Detailed Test Results to Log File’ setting, and applying the above formula to each set of measured latencies (small & large Ping RTTs) and Ping payload sizes, the user can calculate and verify the reported Average Effective Throughput values.
9.1.3 Practical Application of Effective Throughput Measurements
Effective Throughput measurements are a practical way to quickly and easily gain insight into the general bandwidth availability across the entire network path between the test workstation and the a target host/server or network device. This information can verify the expected bandwidth availability and/or confirm link sizing, especially with small or busy links.
41
If one or more links in the network path between the test workstation and the target host/server or network device is heavily loaded, this condition (and its effects on data throughput and application performance) will be revealed by a relatively low Effective Throughput value (compared to the stated link capacity). This is useful as a quick ‘sanity’ check of link bandwidth capacity and availability, especially in cases where bandwidth usage reports cannot be readily obtained for all the links in the path.
In some cases Effective Throughput measurements may suggest a greater link capacity (higher Effective Throughput) that what was reported by network management personnel or bandwidth reports, despite the fact that the link(s) are also carrying some level of user traffic; in this case it is likely that the link’s ‘Committed Data Rate’ (CDR) was reported, which can be less than the physical link capacity - the actual physical link speed and capability could be greater (sometimes, significantly so) than what was reported – which would allow ‘quicker’ transport of user packets and better than expected application end user response times. Effective Throughput measurements can offer insights in these scenarios that may not have otherwise been apparent.
Finally, in cases where bandwidth usage/availability cannot be measured, Effective Throughput measurements can suffice for calculating approximate effects of network transport times for application troubleshooting and/or network impact analysis/modeling.
The Effective Throughput measurements conducted by TextPing when users send Text Messages during application test captures is intended to help verify the expected link capacity, and to be utilized when processing the packet trace files to calculate the percentage of overall response times that should be attributed to network transport; in addition, the Small Ping minimum RTT latency is utilized to calculate the effects of App Turns delay.
9.1.3.1 Limitations of Effective Throughput Measurements
Effective Throughput measurements, while offering a practical and relatively accurate portrayal of general network conditions, do have a number of limitations that should be understood and taken into consideration.
1. Ping Round Trip Time (RTT) measurements can vary significantly depending on the link usage levels in either or both directions, and/or how busy the responding host/server/network device is at the instant the Ping packet traverses the network links or received at the responding end. This issue is also affected by the possibility that the responding device may not put a very high priority on responding to ICMP Ping packets when it is busy with more legitimate traffic, resulting in somewhat erratic response times. It is therefore not unusual for a Small Ping latency value to be higher than the corresponding Large Ping latency value for a given test (which makes an Effective Throughput test impossible and invalid); however, if this occurs frequently it is a strong indicator of an extremely ‘busy’ link or responding device.
2. Effective Throughput measurements are conducted by measuring the Round Trip Time RTT of Ping packets that travel in both directions on network links which may be symmetrical but will usually not have symmetrical loading; this uneven loading will affect the network transport times of the Ping packets by differing amounts in each direction, and these non-symmetrical effects cannot measured or predicted. The end result is that an Effective Throughput measurement will likely be ‘high’ for one side of a given link and ‘low’ for the other if compared to actual availability measurements.
3. Given the nature of how Effective Throughput measurements are taken and the effects of latency measurement resolution (whether the latency is measured in
42
millisecond or microsecond increments), the accuracy of these measurements gets higher as the effective throughput gets lower. In other words, when there is only 1 ms difference between a Small Ping latency and Large Ping latency, the calculated effective throughput value will be high (21,280 Kbps, for example, for a Large-Small payload size difference of 1,330 Bytes), but the accuracy and resolution of that Eff TP value will be less. In this case, .1 ms of additional RTT latency (1.1 ms) results in a calculated Eff TP of 19,345 Kbps – a fairly large difference.
Conversely, if there is a large difference between the Large and Small RTTs, say 5.0 ms for a likely scenario, the calculated Eff TP (for the same payload size difference as above) is 4,256 Kbps. An increase of .1 ms (5.1 ms) results in 4,172.5 Kbps – a relatively small difference (83.5 Kbps).
This issue is compounded if the RTT measurements are limited to 1 ms resolution (such as is the case with the standard Windows ‘ping’ command), which offers a less accurate Eff TP calculation. Again using the above payload sizes as an example, a RTT change from 5 to 6 milliseconds results in a calculated Eff TP difference of 4,256 - 3,546.7 = 709.3 Kbps.
TextPing RTT measurements are accurate to .1 usec (.0001 millisecond), which is much greater resolution than really needed to accurately measure network latencies.
For network and application performance considerations, the loss of resolution and accuracy at higher effective throughputs is not a major concern because most Client/Server applications only really need 2-5 Mbps of available bandwidth to perform well – and this lower range is where Effective Throughput measurements are a reasonably good indicator of true bandwidth availability (throughput).
4. The accuracy of Effective Throughput measurements is also affected by the difference in Large vs Small payload sizes; the large the difference, the more accurate the Eff TP measurement. For this reason the smallest practical payload size should be utilized for the Small Ping, and the largest practical payload size used for the Large Ping – without exceeding the typical 1500 Byte Maximum Transmission Unit (MTU) characteristics of most modern LAN/WAN network facilities.
The manner in which the resolution and accuracy of calculated Effective Throughput measurements will vary with differing payload sizes and RTT latency differences is illustrated in Figure 30 below – moving to the right along the x-axis, as the difference in RTT times increase, the Effective Throughput decreases – but the difference in Eff TP with changes in RTT latency become much smaller.
43
Figure 30: Effective Throughput vs RTT Ping Latency Graph
5. For all the reasons given above, the ‘average’ of a sequential number Effective Throughput tests will be more accurate that the results of a single measurement, which is one of the reasons a ‘set’ of Pings (10 or more) is utilized by TextPing. The other reason is that numerous sequential Ping packets are much easier to spot within a packet trace file than a single pair of packets.
END
© Copyright 2012, PacketIQ Inc. All rights reserved.
0.0
10,000.0
20,000.0
30,000.0
40,000.0
50,000.0
60,000.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
2.0
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
2.9
3.0
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
4.0
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
5.0
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9
6.0
6.1
6.2
6.3
6.4
6.5
6.6
6.7
6.8
6.9
7.0
7.1
7.2
7.3
7.4
7.5
7.6
7.7
Effe
ctiv
e T
hro
ugh
pu
t (K
bp
s)
Latency Difference (Large - Small Ping RTT) (ms)
Effective Throughput (Kbps) vs RTT Ping Latency Difference (ms) 1000 Byte Payload Diff
1332 Byte Payload Diff
480 Byte Payload Diff
