HTTP/1.1 404 Object Not Found Server: Microsoft-IIS/5.0 Date: Tue, 02 Jun 2009 14:11:38 GMT Cluster-Server: WEB1 P3P: CP="NOI ADMa OUR STP" X-Powered-By: ASP.NET Connection: close Content-Type: text/html

404 Object Not Found

Tech Tips

HTTP/1.1 404 Object Not Found Server: Microsoft-IIS/5.0 Date: Tue, 02 Jun 2009 14:11:38 GMT Cluster-Server: WEB1 P3P: CP="NOI ADMa OUR STP" X-Powered-By: ASP.NET Connection: close Content-Type: text/html

404 Object Not Found

 Buy Games

News
 Current / Submit
 Archive / Search
 POTD / Submit

Files
 Main Files

Community
 Hosted Sites
 Forums
 Chat
 Help Wanted
 Mailing Lists
 Get Hosted!
 Contact Us
 Advertise With Us
 Staff

Features
 Index
 Articles
 Mod of the Week
 Levels of the Week
 Model of the Week
 QuakeScopes
 QuakeCon 2005
 Dear Mynx
 PQ Poll
 Mailbag
 Rants N'Raves
 Tech Tips
 Week in Review
 Classic PQ


HTTP/1.1 404 Object Not Found Server: Microsoft-IIS/5.0 Date: Tue, 02 Jun 2009 14:11:38 GMT Cluster-Server: WEB1 P3P: CP="NOI ADMa OUR STP" X-Powered-By: ASP.NET Connection: close Content-Type: text/html

404 Object Not Found


    PlanetQuake | Features | Tech Tips | 10-07-2000
   

Tech Tips

How To Use Trace Route To Find Problems With Your Internet Connection

By John Gillett

PING PLOTTER TO THE RESCUE 

The tracert command-line utility is functional, but is lacking in several areas.  Enter stage right, a shareware program called Ping Plotter.  Ping Plotter has maintained a spot on my taskbar for quite some time now.  No easy feat on my system. 

Iíd like to thank Pete Ness for allowing me to use material from the Ping Plotter help files in this article.  I think it is the best program available for itís purpose, and I encourage anyone who is interested to visit the Ping Plotter Web site at www.pingplotter.com, where both a freeware version, and shareware versions are available for download. 

Basically, Ping Plotter is a trace route program, but it adds some graphs, plus some serious performance upgrades.  It uses the multi-threading capabilities of Windows to check performance on all hops in the route at the same time.  This has several advantages:  1)  It's a lot faster.  2) All hops are tested at the same time, rather than many seconds apart, so the comparison is better. 

Another advantage of Ping Plotter is that you can set it to a continuous mode, where it will test the same route over and over again, forever if you want.  That way you can watch the performance over a period of time without having to re-run the program over and over again.  In this vain, you could leave Ping Plotter running constantly; as it is compact, and while running only takes about two percent of your system resources.  This way, you could ping a nearby server, say, once every five minutes, and thus keep a log of your ISPís down time.  When it comes time to pay the bill, just pay for the percentage of the month you owe them for. 

Ping Plotter Functionality 

Here is a screenshot of Ping Plotterís main window taken from the help files, with numbered descriptions following.  This illustrates just some of the capabilities of Ping Plotter without requiring you to download and install it.  If you install it, the help files contain clickable figures with pop-up explanations for figures such as the one below, as well as other figures from the menus.  There isnít much that is not configurable to your needs, such as the size of packets to use for pinging, to the ability to send an email alert if pings reach a point you specify. 

1)  The "Address to Trace" edit box allows you to pick a new host to trace to. You can either type in an address or select one from the list below (double clicking will make it start tracing as well).

Make sure the address doesn't have a protocol specifier, or sub-directory. If you're trying to find out why http://www.pingplotter.com/orders.html isn't responding, just enter www.pingplotter.com.

Any time you successfully trace to a host, that host is added to the list below.

To delete a host, right-click it in the list and then select "Delete".

2) The "Route Change" pane is used to show the history of route changes.  Any time any hop in the route changes, Ping Plotter stores the old and new route data and adds the time of the change to this list box.  Double-clicking on any time will show the route as of that time.  This is the starting time for the change. 

3) The "# of times to trace" allows you to stop tracing after a certain number of traces.  If you're only interested in a set trace count, you can save some bandwidth usage by not allowing Ping Plotter to trace forever. 

4) The Trace Interval is the amount of time Ping Plotter will wait between each sample set.  If you're doing a long term monitoring project, you may want to set this to 1 minute (or more).  If you're just doing a quick set, you might want to set this to something lower (5 seconds or 10 second).  If the up/down arrow doesn't have the amount of time you want, just type in the time interval you want (e.g. 3.5 seconds). 

5) Itís somewhat important to understand how "Samples to Include" works. 

When running a trace, Ping Plotter can look at just the most current samples.  This is great to watch "trending" (where the response changes over time).  If you include ALL samples (type 0 in this field), then after a bunch of samples, new samples don't affect the graph very much.  Setting this to something like 10 allows you to see how the response times are right now.  All numbers in the trace (upper) graph are affected by this.  When zooming in on the timeline graph, itís important to not have this set to "ALL". 

6) When the word "Querying" is displayed here, Ping Plotter is waiting on the internet for something - either a trace packet to return or a name lookup to return.  When nothing is displayed here, Ping Plotter is just waiting for you to hit "Trace" or "Resume" - or it's waiting for the specified time interval to elapse so it can get its next sample. 

7) The PL% Column indicates the percentage of packets lost (number lost divided by the samples to include that have been sent out)

If you're only including the last 10 samples, then only the number of packets lost in the last 10 samples are shown here. If you want to find out packet loss over the entire session, change the "Samples to Include" number to 0 (which includes ALL samples). The number of packets lost can be shown by enabling the ERR column by right-clicking the graph.  

8) Any hop number surrounded by brackets (like [10]) means that that hop is being monitored for an alert (see alerts for details on setting up an alert). Multiple alerts can be configured for the same IP, and alerts don't work unless some IP in your current route is being monitored (ie: has brackets around it).  

9) Any hop that is underlined means that it's being traced on a time-line graph (in the bottom portion). The underline makes it easy to de-select a hop from being monitored.  

10) The Avg column shows the average response time of the last N samples (where N is the samples to include).  Any timeouts are not included in this sample. 

11) This column shows the individual sample time of the most recent sample included in the set.  If a number is displayed as ERR, that means that the packet was lost - a packet was sent out but was never returned. 

12) The blue X on the graph is the graph of the current (or most recent) sample taken. 

13) The number shown here indicates the graph scale. 

If a dynamic scale is being used, then this number is the maximum response time of any of the included sample set.  This number can change (and WILL change) as new samples are received.

If a fixed scale is being used, this number will always equal that scale.  You can change to a fixed scale by going to Advanced Options. 

14) The black line on the graph indicates the range of times received.  The leftmost side of the line is the minimum response time - the rightmost being the maximum response time.

This is great for showing clear network bottlenecks. If the best response time of a hop is higher than the worst of the previous hop, you're pretty sure there's some problem between these two hops.  

15) The red line on the graph shows the average response times for the route. Lost packets are not included in this average. 

16) This column shows the individual sample time of the most recent sample included in the set.  If a number is displayed as ERR, that means that the packet was lost - a packet was sent out but was never returned. 

17) The round trip line is optional (it can be removed from the "View" Menu).  This shows exactly the same as the last server in the chain, but it's sometimes nice to have this number re-announced for ease of reading.  This line shows the time it takes to get from your computer to the destination server and back again. 

18) The time under the graph shows what time these samples were taken.  If you have more samples than can be displayed on the graph, you can either change the scale (by right-clicking the graph) or scroll the graph (by clicking on the graph and then dragging left or right).

19) A red line in the time-line graph indicates a timeout in this time period. You move the focus of the upper graph to this time by double-clicking it - to find out where the timeout first occurred (in the upper graph).

A lost packet always makes a pixel show up as red. If there is more than one sample in the period - and at least one sample succeeded, then the red line will indicate the average of the successful packets in this time period. If there were no successful packets in the period, then the red line will go from top to the bottom of the scale. 

 




[Main Page] [Features] [Files] [Forums] [Contact] [Hosting Info]

HTTP/1.1 404 Object Not Found Server: Microsoft-IIS/5.0 Date: Tue, 02 Jun 2009 14:11:38 GMT Cluster-Server: WEB1 P3P: CP="NOI ADMa OUR STP" X-Powered-By: ASP.NET Connection: close Content-Type: text/html

404 Object Not Found