Frequently Asked Questions



This FAQ is related to the following IP software testing tools:

  LanTraffic V2 IP Traffic Generator Product info...
  IP Traffic - Test & Measure IP Traffic Generator and QoS Measurement Tool Product info...
  NetDisturb Impairment Emulator Product info...

The Questions

A.   Traffic Generators


1 What is the difference between IP Traffic - Test & Measure and LanTraffic V2 software as they seem to be quite similar?  
2 What is the minimum packet interval supported by the current versions of IP Traffic - Test & Measure and LanTraffic V2?
3 What is the throughput difference between IP Traffic - Test & Measure and LanTraffic V2?
4 I am looking for IP Traffic Generator that can set a ToS marking (i.e. layer 3 Quality of service marking).
Can your product do this?

A1.   About LanTraffic V2

1 I installed LanTraffic V2 on my PC to generate any IP traffic to another PC in the network, this lab is to saturate the 2Mbps bandwidth assigned to these PC’s. I can’t get it working. I receive some “connection drop” messages, Could you please help?
2 I have installed LanTraffic V2 on two laptops connected to a switch.  I setup one connection (one laptop is the sender and the other one is the receiver). The physical link is 100mbps.  When I run the test in unitary mode, LanTraffic V2 reports a throughput of 570kbps.  I would expect to see 90-100mbps. Why is this?


A2.   About IP Traffic - Test & Measure

1 I want to use IP Traffic - Test & Measure to generate IP traffic from PC connected to media converter (copper Ethernet to fiber Ethernet) and then we put loop back in media converter so the IP packet will go back to the source PC (with your product installed). In other word, there is no target MAC address/IP address defined. Can I do that with your product?
2 Can IP Traffic – Test & Measure work with only one PC? Or is it necessary to use two PCs?  
3 I’m using a wireless network. With IP Traffic – Test & Measure, when I specify the payload to be less than 6 bytes, it consumes 100% CPU usage. Do you know what is going wrong?
4 Is it possible to modify the parameters file of IP Traffic – Test & Measure for the IP generator for one connection and apply it to all connections (similar to the IP address and protocol selection). It is a very tedious job to change it for everything one by one.
5 Is it possible to configure and start multiple IP Traffic – Test & Measure remotes at the same time? In our testing the IP Answering will be running on embedded computers at remote locations and will not have a keyboard or monitor connected to them. If we want to change a configuration, we will need to access them from a Central Server running the IP Generator of IP Traffic – Test & Measure. 
6 Can I configure the remote IP Traffic – Test & Measure (in answering mode) to start automatically after reset on Windows? 
7 I am using IP Traffic – Test & Measure on a network which has a high network delay (from 40ms to 60 ms). With default configuration, I can’t get a throughput higher than 18 Mbits per second with one TCP connection. How can I configure IP Traffic – Test & Measure to get the best throughput? 


B.   About NetDisturb

1 Can I run NetDisturb on a Windows Vista system and when will there be a Windows Vista version available? 
2 I have 2 identical NICs on my PC equipped with NetDisturb, both plugged in with working Ethernet cables. When I run the server, it seems OK. When I run the client after that, and click on OK, I get the error message saying: ”Connection with NetDisturb Server can't be established.”   
3 Description of the NetDisturb loss-of-connection-problem: when choosing a delay/jitter law that concerns a router/simulation with limited bandwidth, the emulated connection gets interrupted after a random amount of time, likely 2 to 15 minutes. The NIC that connects to the LAN ceases to output packets. We have intelligent switches that shutdown a port when more than 100 broadcasts are received on a port. We are using PCI-Raiser card. 
4 After we have connected one device on each of the ports on our NetDisturb machine, that traffic is not passing between them, even when the default 'other IP flows' filter is activated. We are in the belief that this should pass all traffic with no impairments. It may be worth noting that we are using VLAN trunk ports to connect these devices, would this have an effect on the behaviour? 
5 I just installed NetDisturb version 4.6 and, I get a message “Embedded Lan Card can’t be used by NetDisturb since it has the TCP/IP stack activated. All NICs that you want to use with NetDisturb must have all protocols stacks unchecked”. How do I do that on Windows XP?  


C.   About the license protection scheme (Site Code / Site Key) for trial or permanent licenses

1 We powered up the machine where we installed the ZTI Software on and the Site Code changed, the machine wasn’t connected to our LAN at the time. It’s the same machine; can you explain why the Site Code might change? I had uninstalled your software and, when I reinstalled it, the site code I had has been denied. Can you explain why?
2 I plugged off my PC. Now I got a message with a title: ”Network License” and an explanation : “The program uses a network and the maximum number of simultaneous users has been exceeded ...”. How can I remove the user that seems to use this software?  



The Answers

A.   Traffic Generators

1 What is the difference between IP Traffic - Test & Measure and LanTraffic V2 software as they seem to be quite similar?
  They are not!
IP Traffic - Test & Measure is a traffic generator and should be used when you want to calculate the packet loss or the transit delay. It includes features in the internal generator to replay the captured traffic. The Traffic Observer of IP Traffic - Test & Measure gives you information on the IP level throughput. This IP Level analysis is only available with IP Traffic - Test & Measure software.
LanTraffic V2 should be used when you want a better traffic throughput and don’t need the IP level statistics.
   
2 What is the minimum packet interval supported by the current versions of IP Traffic - Test & Measure and LanTraffic V2?
  - IP Traffic - Test & Measure and LanTraffic V2 allows you to get an inter-packet delay less than 1 millisecond. By specifying 0 as inter packet value, the internal packet generator send the packets as faster as possible. Linked to the performances of your target computer (physical architecture, CPU power, ram size...) you may able to send tens of thousands of packets per second.

- IP Traffic - Test & Measure and LanTraffic V2 can calculate for you the best inter-packet delay when you specify the number of packets per second to send or when you fix a throughput limitation. 

For example, if you specify 3,000 packets per second, LanTraffic V2 will calculate the number of packets per millisecond to send. Then at the end of each transmitted packet, it will check if and how long it should wait before sending other packets. LanTraffic V2 has a built in adjustment control and a millisecond accuracy calculation: You may send more than 3 packets per millisecond. In this case, LanTraffic V2 is going to wait a little bit to balance the amount of packets sent in advance. For instance when your system is able to send 10 packets per millisecond, LanTraffic V2 is going to wait 2 milliseconds before sending other packets.
   
3 What is the throughput difference between IP Traffic - Test & Measure and LanTraffic V2?
  The IP Traffic - Test & Measure and LanTraffic V2 software are both IP Traffic Generator. They are able to generate a high traffic throughput.
The main difference is that IP Traffic - Test & Measure may also capture the traffic while generating and calculate the statistics displayed in the Traffic Observer tab that LanTraffic V2 doesn’t calculate: when activated, these operations will impact the traffic throughput of IP Traffic - Test & Measure.   To get the same high level of traffic throughput with both products, you should disable the capture and the Traffic Observer statistics from IP Traffic - Test & Measure by selecting the “Configuration/Disable the “Traffic observer” tab to increase performance” option.
   
4 I am looking for IP Traffic Generator that can set a ToS marking (i.e. layer 3 Quality of service marking).
Can your product do this?
  Yes, both IP Traffic - Test & Measure and LanTraffic V2 allow you to set the quality of service marking. With IPv4, you define the DSCP value that fills the IP header field named Type Of Service (TOS).
With IPv6, the DSCP value is not available.


A1.   About LanTraffic V2

1 I installed LanTraffic V2 on my PC to generate any IP traffic to another PC in the network, this lab is to saturate the 2Mbps bandwidth assigned to these PC’s. I can’t get it working. I receive some “connection drop” messages, Could you please help?
  There are some reasons of connection failure with LanTraffic V2. Assuming there is no error in the IP addresses i.e. PC#1 and PC#2 uses the same subnet or a Router/Nat translates addresses, the most common reasons are the following:  
1) The 'Receiver - Traffic + Statistics' tab on PC#2 is not started when the 'Sender - Traffic + Statistics' on PC#1 connections starts. Start the Receiver on PC#2 before trying to establish connection with the PC#1.
2) PC#1 can't reach PC#2. To ensure that the connectivity is correct, try to Ping PC#2 from PC#1 to check the link.
3) PC#2 doesn't wait on the same port on PC#1’s configuration. We suggest you to start the test with the default configuration on both PCs. To apply  the default configuration, use the menu File/New on both sides; start the 'Receiver - Traffic + Statistics' on  PC#2; then enter the IP address of  PC#2 in one connection in the 'Sender - Parameters' tab of the PC#1. PC#1 is now ready to establish connections with PC#2.
4) PC#1 or PC#2 or both are running an antivirus application which prevents from using the configured port.  We suggest using the ‘Getting Started’ help when you launch the software to check your configuration. 
   
2 I have installed LanTraffic V2 on two laptops connected to a switch.  I setup one connection (one laptop is the sender and the other one is the receiver). The physical link is 100mbps.  When I run the test in unitary mode, LanTraffic V2 reports a throughput of 570kbps.  I would expect to see 90-100mbps. Why is this?
  You get this throughput because it is related to the default traffic generator parameters.
Indeed by default, the data size used to fill the TCP or UDP payload is 1460 bytes and the inter packet delay (latency between each packet sent) is 20 milliseconds (0 millisecond with new software versions).
With a null value as inter-packet delay (0 millisecond), LanTraffic V2 generates traffic as fast as possible. Be careful that all LanTraffic V2 statistics are based on data used to fill payload for TCP or UDP packets (those statistics calculations don't integrate the TCP or UDP headers nor the IP and Ethernet headers).
The throughput value used to define Ethernet network is a physical throughput (Ethernet throughput).  
   

A2.   About IP Traffic - Test & Measure

1 I want to use IP Traffic - Test & Measure to generate IP traffic from PC connected to media converter (copper Ethernet to fiber Ethernet) and then we put loop back in media converter so the IP packet will go back to the source PC (with your product installed). In other word, there is no target MAC address/IP address defined. Can I do that with your product?
  IP Traffic - Test & Measure is an application running on the top of the TCP/IP Windows stack.
IP Traffic - Test & Measure is able to use different networks: Satellite, ISDN, ATM, Wireless, FTTx...
So it can be used on any network that supports IP protocol.
   
2 Can IP Traffic – Test & Measure work with only one PC? Or is it necessary to use two PCs?  
  The current version of IP Traffic - Test & Measure needs two PCs because the traffic should go out from the NIC, and it is not the case if you use only one PC. We are going to detail 2 major configurations:
- If there are 2 NICs in the PC, you could expect to send the traffic from NIC1 to NIC2, and vice-versa. It is possible to do that but Windows doesn't send the traffic to the NIC: it loops back the traffic internally. We cannot change this Windows procedure.
- It is the same procedure if you use your own IP address with only one NIC: Windows loops internally not using the NIC. In both cases, Windows reacts as if you were using the 127.0.0.1 IP address.
   
3 I’m using a wireless network. With IP Traffic – Test & Measure, when I specify the payload to be less than 6 bytes, it consumes 100% CPU usage. Do you know what is going wrong?
  Yes, the 100% CPU usage is 'normal': because the packet size is small, the overhead is very large and the PC makes a lot of interface exchanges to send the data.
We don't know your Wifi architecture but the Wifi processor may also consume PC CPU to achieve the job. On the other hand, the number of packets handled by IP Traffic – Test & Measure, used to generate statistics for the Traffic Observer tab is also quite large, and consumes CPU too. 
   
4 Is it possible to modify the parameters file of IP Traffic – Test & Measure for the IP generator for one connection and apply it to all connections (similar to the IP address and protocol selection). It is a very tedious job to change it for everything one by one.
  A copy/paste mechanism is available to copy parameters from one connection to another one.
To do it, click on the “Connection #XX” label that you want to copy. Then the “Connection #XX” label is highlighted in dark blue. Click on the target connection label where you want to copy the parameters specified for the “Connection #XX”.
Redo this operation each time you want to copy parameters from one connection to another one.  
   
5 Is it possible to configure and start multiple IP Traffic – Test & Measure remotes at the same time? In our testing the IP Answering will be running on embedded computers at remote locations and will not have a keyboard or monitor connected to them. If we want to change a configuration, we will need to access them from a Central Server running the IP Generator of IP Traffic – Test & Measure
  Yes, we propose to use command line remote control software - such as rexec or rsh - to start the 'Automation Tool for IP Traffic' installed on each remote PC.
A command line service is installed on each remote. This command line service receives the command for your Central Server to start Automation Tool for IP Traffic with a '-Start' parameter to launch a scenario and start IP Traffic - Test & Measure or a '-Stop' parameter to stop 'IP Traffic - Test & Measure'.
On the Central Server part, you could use a .bat file to start/stop each remote. There is a need of an external rsh, because neither rexec or rsh are available from Microsoft for Windows XP environment - they are only available on Windows 2000 Server versions.   
   
6 Can I configure the remote IP Traffic – Test & Measure (in answering mode) to start automatically after reset on Windows? 
  From the version 2.3, IP Traffic - Test & Measure is able start automatically when Windows is started, via the standard Start/All Programs/Startup Windows mechanism.
When starting, IP Traffic - Test & Measure loads automatically the latest context file. The point is the 'Start All Local Processes' or the 'Start Receiving Traffic' are not started automatically with the application: there is a need of a User to push these Start buttons.
We extend the “Automation Tool for IP Traffic’ to offer such feature – to load a scenario and to start it. This allows starting the remote IP Traffic - Test & Measure, the IP Generator and/or IP Answering, as soon as Windows has been started in a User Logon context.    
   
7 I am using IP Traffic – Test & Measure on a network which has a high network delay (from 40ms to 60 ms). With default configuration, I can’t get a throughput higher than 18 Mbits per second with one TCP connection. How can I configure IP Traffic – Test & Measure to get the best throughput? 
  The throughput of TCP connections is handled by some Winsock parameters that can be configured using the ‘Configuration/Microsoft TCP Stack Parameters” menu. The TCP Windows size may be configured per software.
IP Traffic - Test & Measure allows extending the TCP Windows size by configuring the SO_RCVBUF and SO_SNDBUF. With IP Traffic - Test & Measure release 2.5, the TCP window size may be sized up to 1 Mbytes while previous versions were limited to 65 Kbytes.     


B.   About NetDisturb

1 Can I run NetDisturb on a Windows Vista system and when will there be a Windows Vista version available? 
  The current version of NetDisturb doesn't support Windows Vista: We are actually testing NetDisturb v4.7 in our labs and Windows Vista is included in the list of tests to be done.
The version supporting Windows Vista is going to be available soon.
   
2 I have 2 identical NICs on my PC equipped with NetDisturb, both plugged in with working Ethernet cables. When I run the server, it seems OK. When I run the client after that, and click on OK, I get the error message saying: ”Connection with NetDisturb Server can't be established.”   
  If the NetDisturb Client can't access to the NetDisturb Server, it may be due to a restriction on the RPC authorization of your PC.
For example, it is often the case when you are using Windows XP SP2 with a previous version of NetDisturb v4.6.  
To solve this problem, you have to change the relevant key in the registry such as: [HKEY_LOCAL_MACHINE\SOFTWARE\Policies\Microsoft\Windows NT\RPC] "RestrictRemoteClients"=dword:00000000.  

You can download from ZTI web site the .REG file that changes this registry entry by using:
http://www.zti-telecom.com/download/NetDisturb-EnableRPC.reg

Then, copy that file on a temporary location and double-click on it. Windows will ask you if you agree to change the registry: select Yes. Then reboot your PC who’s going to restart NetDisturb.

Note: since NetDisturb V4.6, the client dialogs with the server using the SOAP mechanism in order to prevent from encountering security restriction problems. You may also upgrade to version 4.6.

For further information please email us at: support@zti-telecom.com 
   
3 Description of the NetDisturb loss-of-connection-problem: when choosing a delay/jitter law that concerns a router/simulation with limited bandwidth, the emulated connection gets interrupted after a random amount of time, likely 2 to 15 minutes. The NIC that connects to the LAN ceases to output packets. We have intelligent switches that shutdown a port when more than 100 broadcasts are received on a port. We are using PCI-Raiser card. 
  It is not a matter of NetDisturb nor of your NIC (we recommend 3COM, Intel or Broadcom NIC but others can be used too). You may try to replace the PC to make disappear the problem, because it should be due to the PCI-Raiser (for example some HP are using such technology so 1 PCI-slot is used for 2 LAN-cards). This is of the main reason of the symptom.
   
4 After we have connected one device on each of the ports on our NetDisturb machine, that traffic is not passing between them, even when the default 'other IP flows' filter is activated. We are in the belief that this should pass all traffic with no impairments. It may be worth noting that we are using VLAN trunk ports to connect these devices, would this have an effect on the behaviour? 
  Regarding the VLAN configuration with NetDisturb, there are some NICs that should be configured to support the VLAN forwarding: there should be a parameter in the NIC driver configuration to allow the VLAN going through. There are examples of the configuration parameters for Broadcom and Intel NICs driver that you can download at http://www.zti-telecom.com/download/VLAN-Capture-NIC-configuration.pdf.
   
5 I just installed NetDisturb version 4.6 and, I get a message “Embedded Lan Card can’t be used by NetDisturb since it has the TCP/IP stack activated. All NICs that you want to use with NetDisturb must have all protocols stacks unchecked”. How do I do that on Windows XP?  
  Because NetDisturb is an Ethernet Bridge, there should be no local protocol stack able to handle the Ethernet frames received. When active, these protocols may conflict with the NetDisturb impairment. The simplest way to deactivate these protocols is to access the Properties for each NIC that NetDisturb is going to use and to unselect all protocols (called items) from the list.
More details can be found in the NetDisturb User Guide version 4.6 § 3.7.1 page 30.
 
C.   About the license protection scheme (Site Code / Site Key) for trial or permanent licenses

1 We powered up the machine where we installed the ZTI Software on and the Site Code changed, the machine wasn’t connected to our LAN at the time. It’s the same machine; can you explain why the Site Code might change?
I had uninstalled your software and, when I reinstalled it, the site code I had has been denied. Can you explain why?
  Every time the software is installed on a machine the Site Code changes. This is part of the license protection security scheme.
To get the Site Key for your license, please do not hesitate to email at contact@zti-telecom.com.  
Please consider to use the USB Software Protection Key to avoid the request for a new Site Key each time you have to reinstall the software or in case of crash disk.
   
2 I plugged off my PC. Now I got a message with a title: ”Network License” and an explanation : “The program uses a network and the maximum number of simultaneous users has been exceeded ...”. How can I remove the user that seems to use this software ?    
  There is a file with TB2 extension in the installation directory. This file contains the current number of users for the software. When you delete the .TB2 file, you remove all the users of the software and you set the license available for a new user of the software.

 

Copyright © 1989-2008 ZTI. All Rights Reserved. :: Email Us | :: Terms of Use | :: Privacy Statement  
 ZTI | 1 boulevard d'Armor | BP 20254 | 22302 Lannion | France | Tel +33 2 9648 4343 | Fax +33 2 9648 1485