Most Downloaded Files




Question: SurfStatsLive User Types
What types of users are there in SurfStatsLive?

Solution:

Server Administrators, Profile Administrators and SurfStatsLive Users (also called end-users).

Server Administrators administers the SurfStatsLive installation, assigns users the right to be Profile Administrators, reset profiles etc. The membership of a user to a group (specified in the Global Setup application) in the Windows user database determines whether a user is classified as a Server Administrator or not. The username and password of a Server Administrator can be validated by the Windows user database or by SurfStatsLive itself depending on the set up in the SurfStatsLive Global Setup Application.

Profile Administrators administers site profiles for one or more virtual servers. The Profile Administrator user type was introduced to handle the situation of sub-hosting. Sub-hosting is where a customer of a site host sub-sells parts of a virtual web server to other customers. Server Administrators can also administer profiles so in the case of a host that do not allow sub-hosting there is no need actually to have Profile Administrators. The username and password of a Profile Administrator is validated by SurfStatsLive itself.

SurfStatsLive Users or end-users have access to request and view reports. A SurfStatsLive User is a user that logs in with a site profile name and password to view reports. A SurfStatsLive user is synonymous with a site profile.

  

---

Question: Profile Group
What is a Profile Group?

Solution:

A Profile Group normally corresponds to a virtual web server or to a logical grouping of virtual web servers from different servers (for instance a collection of load balanced servers). These groups are stored by SurfStatsLive in its own database. Profile Administrators administers the site profiles in a profile group.

  

---

Question: Account Details Location
Where are Server Administrator user accounts  and SurfStatsLive user information stored?

Solution:

Server Administrator user accounts Depending on the Authentication setup in the Global Setup application they are either stored on the local Windows user database, on a Windows 2000 domain or by SurfStatsLive itself. Note that you cannot create user accounts for Server Administrators from within SurfStatsLive, except when you use SurfStatsLive authentication which only allows one username.

SurfStatsLive user information Site profiles (which is synonymous with SurfStatsLive users) are stored by SurfStatsLive in its own database.

  

---

Question: Bypass SurfStatsLive login screen
Can I bypass the SurfStatsLive login screen for instance from site hosting automation software?

Solution:

Yes, you can by posting a form to the page called verifyUser.asp on the website SurfStatsLive is running on. See below for more detail and sample code.

SurfStatsLive Automatic Login Sample

You can bypass the SurfStatsLive login screen by posting a form to a page called verifyUser.asp on the website that SurfStatsLive is running on. verifyUser.asp will do the necessary validation of the username, password and user type and log the user into SurfStatsLive and show the relevant user interface, based on the user type.

Three input fields must be present on the form performing the HTTP post: UserName, Password and selLoginType. selLoginType should be set to SurfStatsLiveUser, ProfileAdministrator or ServerAdministrator.

When clicking the text Log me in automatically, the following sample HTML code in a web page will post the necessary fields to verifyUser.asp and automatically log the user johnny, with password foo into the SurfStatsLive service running on a server called www.MySurfStatsLiveServer.com as a profile administrator.
........
........
<SCRIPT language=JavaScript>
function autoLogin()
{
document.myForm.submit();
}
</SCRIPT>
........
........
<a href="JavaScript:autoLogin()">Log me automatically in</a>
........
........
<form name=myForm method=post action="http://www.MySurfStatsLiveServer.com/login/verifyUser.asp">
<input type=hidden id=UserName name=UserName value=johnny></input>
<input type=hidden id=Password name=Password value=foo></input>
<input type=hidden id=selLoginType name=selLoginType value=ProfileAdministrator></input>
</form>
........
........

Please note: SurfStatsLive unfortunately does not support the administration of profile groups and profiles and server options from third-party web pages or script files.

  

---

Question: Web Browser Administration
Can all administration tasks be performed with a web browser?

Solution: Nearly all tasks, except for setting up the virtual web server to be used by SurfStatsLive, setting up the authentication provider and stopping and starting the SurfStatsLive Merging Service. Remote control software (like Terminal Services, PCAnywhere etc.) can be used for this task.

  

---

Question: Profile Group Import
I need to create a lot of Profile Groups, can I import them in some way?

Solution:

Yes, you can make use of the Administrative Script features in Server Administration. Also useful for SurfStatsLive upgrades.

The script file has to be run from within SurfStatsLive Server Administration if you created a script file. You cannot run the script file from a DOS command line or by using the Windows Script Host.

  

---

Question: SurfStatsLive and ZIP files
Can SurfStatsLive process compressed files?

Solution:

No it cannot. The problem is that unlike SurfStats Log Analyzer, SurfStatsLive keeps a record of each log file it processes and the last position in the log file. To process a zip file it would need to unzip the file first, check which files have changed and process the changed files, which gets tricky. You can set up SurfStatsLive to automatically move files after processing to a backup directory, where you can zip them.

  

---

Question: SurfStatsLive and Javascript
Do I require JavaScript in my web pages to use SurfStatsLive?

Solution: If the log files for the web server(s) is accessible on the same server or other servers on the network you don't. If the SurfStatsLive Server does not have access to the log files you can set up SurfStatsLive as Statistics Bureau to accept hits from browser clients generated by JavaScript.

  

---

Question: Can SurfStatsLive hang my web server?
Can SurfStatsLive hang my web server?

Solution:

Nothing is impossible, but it is improbable that this application will hang your server. SurfStatsLive was designed to run on a web server and was rigorously tested by SurfStats and various web hosting companies. The following features improves reliability and performance:

  

  

Yes it can. When you log in as a SurfStatsLive user you can select the language from the Options tab.

Unfortunately it is not feasible for us to provide language translations for every language and we have therefore provided an language editor so you can add or modify the report phrases in yur required language.

  

I get and error when trying to create a new virtual web server

 "Error creating virtual web server: permission denied (70)".

You are most likely not logged in as a user with administrative rights.

You would also need to set the identity for the COM+ package on the "Identities" tab in the SurfStatsLive Global Setup Application. The account specified there is the account that must have the necessary rights. You may also want to give that account the right to log in as a batch job (Control Panel - Admin Tools - Local Security Policy).

  

Where can I find out what visitors, unique visitors, pages, hits, etc. mean?

Absolute URL - The full Internet address of a page or other World Wide Web resource. The absolute URL includes a protocol, such as "http," network location, and optional path and file name. For example, http://surfstats.com/ is an absolute URL. See also URL.

Active Server Page (ASP) - A document with a .ASP extension that contains embedded server-side scripting. ASP-compatible Web servers can execute these scripts. On the client side, an ASP is a standard HTML document that can be viewed on any platform using any Web browser.

Ad - A text, graphic or banner advertisement on a web page that takes the visitor to another site when clicked on.

Ad Click - A mouse click on an advertisement on a web site, which takes the visitor to another site.

Authentication - The verification of a user by matching a username and password in a multi-user or network environment. A user's name and password are compared against an authorized list, and, if the system detects a match, access is granted to the extent specified in the permission list for that user.

Bandwidth - The measure of the traffic (in kilobytes of data) transmitted from the web site.

Bookmark - Also referred to as an anchor. This is a named location on a Web page that can be the target of a hyperlink. A bookmark can be applied to a string of characters or exist on a page separately from any text. Bookmarks allow authors to link to a specific section of a target page. A bookmark is preceded by the hash sign (#) in an URL,.

Browser - The program used to locate and view HTML documents.

Cache – An area that stores text and image files from a web site already viewed. When returning to the web page again, the browser loads the files from the cache, thus saving time.

CGI (Common Gateway Interface) - A standard method of extending Web server functionality by executing programs or scripts on a Web server in response to Web browser requests. A common use of CGI is in form processing, where the Web browser sends form data to a CGI script on the server, the script integrates the data with a database, and sends back the results as a Web page. Use of CGI can make a Web page much more dynamic and add interactivity for the user.

Client (visitor or user) - A computer that accesses shared network resources provided by another computer on a local area network or the Internet. A unique IP address or URL identifies this visitor.

Client Session or visit - A client session or visit is started when a request for a file (of any type) was made from a unique IP Address. If no more file requests were made in the cut-off or idle time, the program assumes that the visit was terminated. If a file quest is then made after the cut-off time from the same IP Address, a new visit is counted. The program's default cut-off time is thirty minutes.

Client or visit tracking

There are three commonly used methods to track visitors.

1.      The most accurate way to track unique visitors is by forcing them to log on every time they visit your site. This way the authenticated name is used to track a client's activity on the site.

2.      Using cookies can improve the accuracy of tracking unique visitors. The cookie (saved on the client's pc) is used to differentiate between visitors that used the same IP Address.

3.      Tracking file requests from IP Addresses. Determining the uniqueness of a visitor by tracking their IP address is not always accurate because most ISP’s use dynamic IP Address allocation. The following illustrates this:

·        A visit can be incorrectly counted as a new visit. The same user can have a different IP Address if the session was terminated and he logged on again. The program will count this visit as a new visit.

·        A visit can be incorrectly continued. If the IP Address was allocated to another user, which logs on to your site within the cut-off time, he will be seen as the previous visitor.

·        Some ISPs provides the same client with different IP Addresses during a logon session. This can increase the number of unique visitors significantly.

Client error codes - Status Error Codes in the 4XX range indicating the type of errors that occurred during a file/page request. Also refer to the Status Codes.

Cookies - Files containing information about visits to a web site (e.g., user name and preferences). The user’s browser provides this information during the visit to a Web server. The server informs the browser to store some of this information (and perhaps other) on the client machine in a cookie file. When the visitor accesses the same web site again, the cookie is automatically presented to the server by the browser and the server can then configure it based on the information provided or react upon the information in the cookie.

Country Internet Codes - Top Level Domains or ISO Internet Codes are the suffix added to a domain name, e.g. myname.co.us. You can find the latest list at

http://www.iso.org/iso/en/prods-services/iso3166ma/02iso-3166-code-lists/list-en1.html

Database - A file containing records of information that are organized and presented to  serve a specific purpose, such as the facilitation of searching, sorting, and recombination of data. Databases can be published on the World Wide Web to let site visitors look up information in records or add new information to the database.

Directory - World Wide Web pages consist of varying files. These files are often set up in a directory structure, resembling "tree branches". Directories are usually structured to keep similar topics within a specifically named directory.

Domain Name - The text name corresponding to the numeric IP address of a computer or domain on the Internet.

Domain Name Lookup - Also called DNS lookup or IP Resolving. This is the process of converting a numeric IP address into a domain name or URL.

Domain Name Suffix - The last digits of a domain name can be used to identify the country or type of organization. Possible suffixes for the organization type includes:

.com = Commercial
.edu = Educational
.int = International
.gov = Government
.mil = Military
.net = Network
.org = Organization 
.xx = where the xx is a two digit country code, e.g. .uk for United Kingdom

Download Manager – A program that is used to download large files from a host computer. This type of program can usually resume the download process after an aborted file transfer thus reducing bandwidth requirements. There are various download managers available like GetRight, Gozilla, ReGet, BullitproofFTP, NetVampire, etc.

Entry pages - This is the first requested page from a client or how the client enters your site. If the client entered your site from a search engine, this would be the file name part of the URL submitted to the search engine. The server looks for the default page that can be any of the following in the order:

default.htm
index.phtml
index.pht
index.cgi
index.shtml
index.html
welcome.cgi
welcome.html
index.htm
index.php3

The first page that it finds will be displayed and if none of these files are found, it will display the list of files in the directory.

Exit pages - This is the last page your client has requested.

External hyperlink - A hyperlink pointing to a page or file that is outside of the current web, i.e. not on the current server.

Favicon.ico - An icon file that is placed on a visitor's links to a book-marked site (only in Internet Explorer 5 or later). Hits to this file can be an indication of how many visitors bookmarks your site. Also see http://www.favicon.com/ how to generate one.

Filters - A means of narrowing the scope of a report by specifying ranges or types of data to included or excluded.

Firewall - A method of protecting the files and programs on one network from users on another network. A firewall blocks unwanted access to a protected network, while giving the protected network access to networks outside of the firewall. A company will typically install a firewall to give users access to the Internet while protecting their internal information

Forms - An HTML page, which passes variables back to the server. These pages are used to gather information from users.

Folder - A named storage area on a computer containing files and other folders.

Frame - An area of a Web browser window defined by a frames page. A frame appears in a Web browser as one of a number of different areas in which pages can be displayed. A frame may be scrollable and resizable, and may have a border. You display a page in a frame by creating a hyperlink to the page and specifying the frame as part of the hyperlink.

Frames page - A page that divides a Web browser's window into different areas called frames that can independently display several Web pages.

FTP - File Transfer Protocol is a standard method of transferring files between computers over the Internet. You can do so with a FTP client program.

GIF - Graphics Interchange Format is a raster based image file format commonly used in HTML documents. The PNG format is slowly replacing this format as the GIF format is a proprietary format from Unisys.

Hit - A request for any object or file that is on a web site. This could be an html page, a file or a graphic on a page. A request for a page can generate a lot of hits depending on how many sub-elements of files the page consists of. This is an indicator of web site traffic but not an indicator of how pages were looked at. Also see Page and User.

Home Page - The main or default page in HTML (Hyper Text Markup Language) format of a Web site. This page usually provides visitors with an overview and links to the rest of the site. This page's address is referred to as URL.

HTML - Hyper Text Markup Language is used to write documents for the Internet.  It specifies hypertext links between related objects and documents.

HTTP - Hyper Text Transfer Protocol is a standard method of transferring data between a Web server and a Web client.

IP Address - The Internet Protocol address identifies a computer connected to the Internet.

Log File - A file created by a web or proxy server, which contains all of the access information regarding the activity on that server.

Log Source - A set of log files from a web server (for instance c:\winnt\system32\logfiles\w3svc15\*.log). A site profile can point to one or more log sources.

META tag - An HTML tag that must appear in the HEAD portion of the page. META tags supply information about a page but do not affect its appearance. Some search engines when adding a web site to the catalogue use these tags.

Organization - A commercial, academic, non-profit, government, or military entity that connects users to the Internet, identified by an entity's Internet domains.

Page - Any document, dynamic page, or form defined as a page file, e.g. .html pages..

Page (HTML) View or Request - The request for a file defined as a page file. The page is basically what you see after the transfer and can consist of many other files. Page requests do not include hits to images, component pages of a frame or other non-html files. The number of page Hits = the number of page views. Also see hit.

Platform - The operating system (i.e. Windows 95/98, Windows NT/2000, Linux, Unix, Macintosh, etc.) used by a visitor to the site.

Protocol - An established method of exchanging data over the Internet (e.g. FTP, HTTP, etc).

Proxy server - An Internet server that acts as a firewall, mediating traffic between a protected network and the Internet.

Referrer - The URL of an HTML page that refers to the site. Referrer data is useful to determine how users came to your site.

Relative URL - The Internet address of a page or other World Wide Web resource relative to the Internet address of the current page. A relative URL gives the path from the current page to the destination page or resource. A relative URL can include a protocol. For example, the relative URL Test/Sample.htm refers to the page Sample.htm in the folder Test, below the current folder.

Robots - Also called search spiders. These are the programs used by search engines to catalog web sites.

Robots.txt - A file that instructs the search engine's robot which directories to catalog or to skip. If you do not have this file, it will show up as a 404 error (file not found) when a search robot visits your site. You can use an empty file.

Root web - The default, top-level web provided by a Web server. To access the root web, you supply the URL of the server without specifying a page name or sub-web.

Script - A type of computer code that can be directly executed by a program that understands the language in which the script is written. Scripts do not need to be compiled to be run.

Status Codes - A code entered in every hit or log line in the log file that indicates the status of a file/page request.

Information codes 1XX - This code indicates a provisional response, consisting only of the Status-Line and optional headers, and is terminated by an empty line. It is usually only for experimental purposes.

100 Continue
101 Switching Protocols

Successful 2XX - This status code indicates that the client's request was successfully received, understood, and accepted. Possible Successful Status Codes are:

200 OK
201 Created
202 Accepted
203 Non-Authoritative Information
204 No Content
205 Reset Content
206 Partial Content

Redirection 3xx - This code indicates that further action needs to be taken by the user agent in order to complete the request.

300 Multiple Choices
301 Moved Permanently
302 Moved Temporarily
303 See Other
304 Not Modified
305 Use Proxy
306 Unused
307 Temporary Redirect

Client Error 4XX – These are errors that occurred due to an invalid or failed request by the client or visitor's browser. Possible client error codes are:

400 Bad Request
401 Unauthorized
402 Payment Required
403 Forbidden
404 Not Found. This is the most common error on the reports, and means that someone has requested an object that is not within your site. This could be because of:

Bad or outdated links to pages or files within your site.
Bad or outdated links to pages or files from outside your site
Someone mistyped or misspelled the URL to get to an object on your site.

405 Method Not Allowed
406 None Acceptable
407 Proxy Authentication Required
408 Request Timeout
409 Conflict
410 Gone
411 Length Required
412 Precondition Failed
413 Request Entity Too Large
414 Request-URI Too Long
415 Unsupported Media Type
416 Requested Range Not possible
417 Expectation Failed

Server Error 5XX - These are errors that occurred due to an invalid or failed task on the server. Possible server error codes are:

500 Internal Server Error
501 Not Implemented
502 Bad Gateway
503 Service Unavailable
504 Gateway Timeout
505 HTTP Version not supported

Internal Errors 6XX - These are errors that occurred in the link checking program itself (these are unofficial error codes used to report problems).

600 Malformed URI
601 Connection Timed Out
602 Unknown Error
603 Could Not Parse Reply
604 Protocol Not Supported

SSL or Secure Sockets Layer - A proposed open standard developed by Netscape Communications for establishing a secure communications channel to prevent the interception of critical information, such as credit card numbers. The primary purpose of SSL is to enable secure electronic financial transactions on the World Wide Web, although it is designed to work with other Internet services as well. This technology, which uses public key encryption, is incorporated into many Web browser and Web servers. Refer to http://www.rsa.com for more on public key encryption and the subject of cryptography.

Server - Also called host. This is a computer that offers services on a network. On the World Wide Web, a server is a computer that runs the Web server software that responds to HTTP protocol requests.

Server Error Codes - Status Error Codes in the 5XX range indicating the type of errors that occurred during a file/page request. See Status Codes 

Shared Client Session - A Session, which is, shared among 2 or more views i.e. a client who has visited more than 1 Page. Other Log File Analyzers shows this as separate clients, which do not tally with the total clients. Some show it as "Not Applicable"

Spiders or robots - An automated program, which searches the Internet and add data to search engines. On a web site the robots.txt file specifies which directories the spider should catalog.

Time Zone - This is the time relative to Greenwich Mean Time ( 0:00 ), e.g. GMT - 6 for Central America. Most World Wide Web servers record hits or requests in local time where the server is located. Microsoft IIS servers use GMT by default.

Unknown Referrers - When the referrer field in a log file is not logged, the program reports the hit as an unknown referrer. The referrer field can be empty when the visitor as a spider, the visitor entered the URL directly in the browser or selected the link from favorites. Some browsers can also leave the field empty. Some log analysers ignore this field. SurfStats Log Analyzer V8 and SurfStatsLive V3 and later records only the original referrer. If a visitor used Google to find your site, it will be shown as the referrer but other links on your web site, e.g. from index.html to buy.html, will not show index.html as referrer.

URL - Uniform Resource Locator is a means of identifying an exact location on the Internet. An URL consists of four parts:

Protocol Type (HTTP),
Machine Name (domain),
Directory Path (/), and
File Name (default.htm),

e.g. http://www.surfstats.com/surfstats/default.htm

User (client or visitor) - Anyone who visits the site at least once. A cookie or an IP address uniquely identifies a user. If there are no cookies available, the IP Address is used to determine a unique visitor. A unique visit is the number of unique visits in a specific time period from a unique IP Address or cookie used.

User (client) Session - A series of consecutive requests from a user to an Internet site.  A user session is terminated when a user does not make another request for more than 30 minutes.

If for example a visitor with IP Address 1.2.3.4, visits the site, logs out and another visitor logs in an hour later with the same IP Address, there would be two user sessions but one unique visitor. 

  

Log file analysing is not an accurate science and you should not waste too much time on specific figures. You will also find that no two log analyzers report the same figures.

Report accuracy

There are many factors that influence the accuracy of reports from analyzing log files. If the generated reports look inconsistent, check that:

You have specified the path and log file names correctly.
The log files are parsed and entered in the database correctly. You can do this by specifying a small log file and comparing the reports to the actual entries in your log file.
You have specified the log files correctly.
You do not have many error lines in your log files that are being discarded by the program.
You have specified filters correctly. Note that filters are not case sensitive.
You have specified the period to report on correctly.
You have used the correct method to populate the database. We recommend that the first option (overwrite the database) be used to populate the database. Changing filters when using the Add to database option would mean that the database could contain inconsistent entries.

Unknown referrers

The report will sometimes show up some fields as "Unknown" if the log line does not contain all the referrer information. The fields will show as “-“  in the log line. This may happen when:

The user typed the URL of the web site directly into the Address field of the browser, i.e. the URL was not referred from another site.
Software was used that blocks the sending of browser information to the server.
The browser software does not send referrer information, e.g. some spiders and e-mail siphoning software. 

Assumptions

Log Analyzers have to make certain assumptions when analyzing log files. These include how to track unique visitors (using IP Addresses or Cookies), the cut-off time to count a visitor as a new visitor and how errors in log files are handled during the parsing process.

A user session includes a series of consecutive requests from a user to an Internet site.  A user session is terminated when a user does not make another request for more than 30 minutes. This is referred to as the cut-off time. This means a user may log in many times per day and if the requests are always less than 30 minutes in-between, the program will only report it as one user session.

The period being analysed also plays a role. If you determine the number of unique users every day and add it up to get a week’s unique users, this would probably not be the same as the number of unique users determined per week. The reason for this is that a unique user might not be counted as a unique user early the next day when using the IP Address as unique identifier and the cut-off time is less than 30 minutes. A daily count would show 7 but a weekly count would show 1 when cookies are used and the same user visits every day.

Dynamic IP Allocation

Due to the shortage of IP Addresses, most ISP’s dynamically allocate IP Addresses to their users from their limited range of IP Addresses. This means that a visitor with a certain IP Address might not be the same visitor with the same IP Address a little later.

Mapping individual IP addresses or requests onto users or even using cookies are not entirely accurate -- the exact number of individual users can only be determined by requiring them to log on individually to your site.

Caching

One problem in providing accurate results is the caching performed by your PC and ISP. Caching reduces the number of pages requested at the host server, as popular pages are stored locally either as temporary files on your hard drive or in the ISP's page cache. Since log files only record files requested from the server, the number registered in the log file can be incorrectly low.

Visitor’s Disk Cache

If your web site has been visited, the downloaded files are on the visitor’s hard disk. Requests for files on your web site would be first served from the visitor’s cache and if not available (or old), would only then be served from your web site’s server. This would result in fewer hits in your web site’s log files.

There are JavaScript code available on the Internet that you can put in your web pages to force a reload or refresh of your web page. You can search for  “force reload JavaScript” on a Search Engine to find a suitable script.

Caching Servers

Many ISP's are running cache servers to reduce web traffic and speed up the transfer of commonly accessed web sites by their users. There are also several national and international caching servers that many ISP's are using in order to speed up transfers. This means for instance that if one AOL user visits a site, then any of the other AOL users who visit that site after that would load the site from the AOL cache, and not from the visited site.  The larger ISP’s like AOL, Earthlink, MSN, Prodigy, MSN, Earthlink, etc., all use caching servers and hits from these users could be much more than reflected in the log file.

Geographical location

Visitors’ locations cannot always be accurately determined, as an IP address does not necessarily specify the geographical location of the user.  A large number of IP addresses can also remain unresolved.

Hits from “non-visitors”

Spiders or robots from search engines that catalogue your site can exaggerate your site statistics, as they are not true visitors. Eliminating them from your overall scores can provide more realistic figures.

Firewall or proxy server

The incoming web traffic that passes through a firewall or proxy server to reach the web server is in most cases identified by a single IP address of the firewall or proxy server.  This is called Address Translation. The log file will in a case like this only contain one unique IP Address even if multiple users from behind the same firewall or proxy server visited the web site. Analyzers will these visits as visits from the same user.

Errors in log files

Errors in log files are usually the first item to look for when you are getting inconsistent reports or reports that are not making sense. Errors in a log line may cause that the log analyzer discard the line’s information. If your log files are not very large, you can edit the log file with an editor like Wordpad or Notepad.  Look for strange characters or misplaced text strings and remove or correct it.

You should correct the cause of the problem (e.g. an overloaded server) if your log file contains many errors. Your reports will only be as accurate as your source data.

Page view and user sessions

Some clients have reported that page views and user sessions do not tie up. Please keep the following in mind when interpreting report results:

·        Files not defined as a page type and graphic files also trigger a user session.

·        The user session cut-off time.

·        Dynamic IP Address location.

·        The effect of caching.

One visit can include many hits.  A report on downloaded files may show a few hits but no user session. This is because the visit is counted at the start page where a visitor enters the site. Only a request for the entry page will count as the start of a user session.

  

Tips & Tricks

General tips
Insist on referrer information in log files or find a host that provides log files with this information. You cannot get meaningful information of your web site's activity without referrer information.

If you have multiple sites, set up a site profile for each site and also a profile that combines all of them.

Use a third party submission service like World Submit to submit your site to search engines. When using a third party services your e-mail address is not supplied to search engines and you would not get the usual, mostly unwanted, e-mail replies.

Do not over analyze. Use the Parento principle (also know as the 20 - 80 percent principle) when analyzing a web site. This means that you should first focus on the few clients that generate most of the traffic to your web site, the few referrers that refer most traffic to your web site, the few countries that most of your clients come from, the few incorrect links that causes most 404 errors, the search engines that generates most of the traffic, etc.

Use keywords to improve popularity
Use the Top Search Keywords and Top Search Phrases reports under the Referrer heading to see which keywords or phrases visitors use to find your site. Gather statistics for a couple of months before you decide on your search phrases.  Build focused pages around queries and ensure that you have pages within your web site with good, solid content for these terms to please the crawlers. Do not pick keywords that are too popular or broad. Once you have set everything up, add the keywords or phrases in the Meta tags and submit to search engines. Use the Page Ranking Tool supplied in SurfStats Log Analyzer to check your ranking at the most populare Search Engines.

Look for visitors that never click past your homepage
If the number of single access pages is unreasonably high or visitors seldom click past your homepage, this can indicate a weak page that does not inspire visitors to further explorer your web site. An image that takes a long time to load on a main entry page and that is used, as an entry page is a definite No-No. When using a Flash entry page, also create a link to bypass the Flash entry page. The result will surprise you when you compare hits between the Flash and Bypass pages..

Look for visitors that bypass your payoff page
If a substantial number of visitors exit your site without visiting your payoff page, the pages linking to your payoff page could be weak and may need some atention. Make sure you have clear links and a reason for visitors to visit your payoff page.

Look for incomplete hits by noting the bytes transferred
Incomplete hits can indicate server capacity and connectivity problems or that a page is too large. See if there are date/time patterns, which also indicate capacity problems.

Look for Quality, not just Quantity
Track hits from ads and reciprocal links by tracing their clicks through to your payoff page. It doesn't pay to run ads that generate hits but seldom click-through to your payoff page. Also see the section on advertising analysis.

Look for visitors repeatedly entering your site on pages other than your homepage
This could indicate the other pages have important search engine keywords in their text that should be included in your homepage text, or that you should add their Meta tags to your homepage Meta tags.

Look for patterns of search engine keywords that result in hits on your payoff page
This helps target the search engines, understand your visitors' needs, and select the HTML text that best motivates your visitors.

Look for referrers that generate a lot of traffic to your site
Visit the site that generate a lot of traffic to your site and see why they linked to your site. This can help you to market your site more effectively.  No referrers from search engines indicate that you are either not listed or not listed effectively on search engines. You can check your site's popularity on the web with services that provide link popularity checking like www.LinkPopularity.com  They provide this service for free and offer tips on how to improve your link popularity.

Look for search robot activity
Add a robots.txt to the root directory of your web site. Search robots can generate a lot of hits, especially to the robots.txt file. Exclude (filter out) accesses to the robots.txt file by search engine robots as this can influence or distort traffic significantly. If you get no requests for this file it means that you should submit your site to search engines. No hits to the robots.txt  file indicates that your are not listed on search engines that send out robots to catalog your site.

Look where clients exit your web site
This can be the last page they accessed or they can exit it by clicking on an interesting link on your page (also refer to the section how to create hits to external links in your log files in Advertising Analysis). 

Look for web site problems
Web site problems that completely break a web page are pretty easy to spot (like very few hits or error codes in your log file), but others may be more difficult to find. If your site gets a large number of unique visitors, but the total number of page views is small, the problem may be due to:

Poor site navigation
Uninteresting site content
Browsers that cannot view your site’s pages properly

Use the detail daily error report and the other error reports to monitor, find and reduce errors on your web pages. Start with the most errors of incorrect links and fix it.

Do visitors find your site interesting?
See if clients bookmark it.  Add a "favicon.ico" file to the root directory of your web site. Also add the "favicon.ico" file to every directory of your site.  When a visitor,  using Internet Explorer 5+,  bookmarks your site, a request is made for the file "favicon.ico"  You can then filter on the file "favicon.ico" to see the number of hits. By knowing the number of hits from the total browser hits and those just Internet Explorer 5+, you can get a good ballpark figure of the total number of bookmarks to your site. By generating a report on hits to the "favicon.ico" file over time, you can see if you are getting new visitors and if the number is improving.  (Repeat visitors have probably book-marked you page already and will not bookmark it again). By placing  a "favicon.ico" file in every directory, you can also see which directories are bookmarked.

See if there are many single access pages in relation to the number of hits per page. The alarm bells should ring if the number of single accesses to default.htm (your home page) is about the same as the total number of accesses to this page.

See the number of accesses to your home page (usually default.htm) in relation to accesses to all other pages. This is a good indication of how many visitors click past your home page.

See if visitors recommend your site to their friends and associates. At  http://www.recommend-it.com/  you can get a script that visitors can click on to recommend your site and also informs you when it has been recommended.

See if they link to your site. You can find out by typing the following in the search window of most search engines:

Link: www.yoursite.com where yoursite.com is you web page address.

See the time visitors spend looking at your pages. 

See how many return visitors you get.

See the number of page views per visitor.

Find out more about visitors
If you have received e-mail from a visitor, you can look at the suffix of his e-mail address and then look-up the country from the ISO country code list. This list is available in the online help in SurfStats (Help-Contents-Appendix-List of country and ISO internet codes).

You can also do a Domain Name System (DNS) lookup of a visitor's IP Address, also referred to as IP Address Resolution.  A DNS lookup queries DNS servers for the IP Address of a domain name. Note that some IP addresses cannot be resolved due to a variety of reasons like some ISP's allocating dynamic IP addresses to their clients, the host is not available, etc.  The resulting URL may contain the domain name and the country of origin of your visitor.

You can also find out more about your visitors by clicking on the IP Address hyperlink in a report and browsing to your visitor’s web site.

You can query IP Address databases through a whois lookup at the three Regional Internet Registries (RIRs) worldwide, which collectively provide IP registration services to all regions around the globe.  

ARIN (At http://www.arin.net/whois/index.html) -  North America, South America, the Caribbean and sub-Saharan Africa.
RIPE (At http://www.ripe.net/db/whois.html) - Europe, Middle East, parts of Africa
APNIC (At http://whois.apnic.net/apnic-bin/whois.pl) - Asia Pacific

or at

AllWhois at http://www.allwhois.com/
Network Solutions  at http://www.networksolutions.com/cgi-bin/whois/whois
Department of Defense at http://nic.mil/cgi-bin/whois
GeekTools at http://www.geektools.com/
UXN Spam Combat at http://combat.uxn.com/

Look for server attacks
Server attacks are worm or virus attacks as well as unauthorized users trying to gain access to protected directories.

A common server attack is a request to the “cmd.exe” file caused by the Nimda or the Code Red virus present on a Windows machine. If the machine is on a local network, you can find out which machine it is on and inform the network administrator. If this file was requested from the Internet, you can do a DNS lookup through the program or do a Whois lookup (using www.allwhois.com or  www.arin.net ) to find out who this comes from. If you can find out where the request came from, you can inform the originator of  the virus on their server.

These requests usually cause many 404 (File Not Found) errors in your log files. If you find successful accesses to the file cmd.exe on your Windows Server, your server is probably infected.

Clean up your log files before processing
You could speed up processing of your log files if you remove unwanted entries (e.g. requests for images files, 404 errors, junk entries, etc.) from your log files first. You can do this with the free SurfStats FileFilter available at:

http://www.surfstats.com/download.asp

  

Extended or Referrer information is most probably not logged. To get all the reports available in SurfStatsLive and SurfStats Log Analyzer, you need to have referrer information in your log files. Contact your web host or server administrator to set extended looging on.

The following is a very basic description on how to set up extended logging to get referrer information in log files on Apache and Internet Information Server (IIS) servers:

Apache

Please refer to httpd.apache.org for documentation on the Apache web server.

We suggest that you use the Combined Log File Format and also log referrer information. You can configure Apache to generate combined log files by editing the "httpd.conf" file in the \Apache\conf\ directory and change entries as follows:

Specify the log file format with the LogFormat directive, example: LogFormat "%h %l %u %t \"%r\" %>s %b \"%{Referer}i\" \"%{User-Agent}i\""

Specify that you want a combined log file stored and the location, for example:  CustomLog logs/combined.log combined

IIS

We suggest you use the W3C Extended Log File Format as it is the only IIS5/6 format that logs Browser/Platform and Referrer on IIS (Internet Information Server). SurfStats also supports the other available IIS types. To set it up on IIS5, (IIS6 and IIS4 should be very similar):

Click on Start - Programs - Administrative Tools - Internet Service Manager - select your server and web site- right click on it and select Properties - Choose Active log format (W3C Extended Log File Format) - Click on Properties - Extended Properties and set the required fields.

At least the following fields should be logged:

Date  
Time
Client IP address (c-ip)
User Name (cs-username)
Server IP address (s-ip)
Server Port (s-port)
Method (cs-method)
URl Stem (cs-uri-stem)
Url Query (cs-uri-query)
Protocol Status (sc-status)

Bytes sent (sc-bytes)*
Bytes received (cs-bytes)*
User Agent (cs(User-Agent))* - for browser information
Cookie (cs(Cookie))* - for cookie information
Referer (cs(Referer))* - for referrer information

We recommend logging of all extended properties. Stop and start the web server after editing the properties. Create an entry in the log file by accessing your web site with a browser and check if it is correctly logged in the log file.

* By default is NOT logged by IIS.

Please note that IIS reserves and locks a 64 KB block in the current log file, which it is logging to. The current 64 KB block in use cannot be added to the database. Some clients reported an error “Database empty” and expected this information to be added the database.

Also note that the default setting for ASP buffering is ON in IIS5/6. This causes the sc-bytes field to be logged as 0. You should switch ASP buffering OFF to log this field. If you do not do this, you will not be able to track bandwidth in reports on ASP files.

  

  

Note that you cannot have more than 20 records displayed on graphs.

You can set the number of entries or records (table and sub-table entries) to report on. There are two places to set this:

1. Edit Site - Contents or
2. Using the Control Panel - Report Options,

  

The database's index was problably damaged during installation. The best would be to uninstall the program, delete the directory

\Program Files\SurfStats8000

and re-install the program.

  

Filtering out robots is quite complex. SurfStats classifies anything that looks for robots.txt as a robot. This works in most cases but is not always effective as subsequent visits from robots may not request this file. Reports in the "Traffic" category display visits from humans and robots. Reports in the "Visitors" category displays visits from human visitors only. Reports in the "Spiders" category displays visits from spiders only. In some of these reports you can reclassify visitors by clicking and editing the "Type" field displayed in the report's tables.

Note that you must have a robots.txt file (even if it is empty) before SurfStats will make the distinction between robots and visitors.

  

  

The page lookup utility has got stuck in repeated loops.
I entered my site URL as http://www.mysite.com/mysub/

What is wrong? 

Specify the URL as http://www.mysite.com/

The program takes the URL and adds the path from the log file to the URL. If the path was specified as http://www.mysite.com/mysub/, the path will be added to give http://www.mysite.com/mysub/mysub/ and will continue on doing this.

  

See the "Upgrading to a new version" section in SurfStatsLive installation guide at http://www.surfstats.com/sslive_install/index.html

This document is also in the readme file in the latest version available from www.surfstats.com/download.asp

  

The easiest would be to have a look at the instructional video about this topic at http://www.surfstats.com/instructionalvideos.asp. The last two videos there. If you set it up it is advisable to first generate reports so that the page data is filled in then set it up from conversion tracking as shown in the videos, then regenerate reports.

If you have additional queries please do not hesitate to contact us.

[--- UPDATED: 4/7/2005 5:16:55 AM ---]

  

SurfStatsLive V3.0 and prior included the SurfCmd utility. A better solution (for V3.0), which makes use of a Windows Scripting Host (WSH) file, is available at http://www.surfstats.com/demo/SSLive3000AutoAdmin.zip. Unzip the contents of the zip file to any folder on the server running SurfStatsLive and read the readme file included in the zip.

Versions of SurfStatsLive after V3.0 will include the WSH file in the installation. 

  

We have done a huge amount of investigation into this and so far have been unable to reproduce it at our offices. The only common thing so far with users encountering this problem was that they use 64-bit CPUs. We'll check if we can reproduce it on a computer with a 64-bit CPU.

  

  

Short Answer...

Please try the following:

1. Go to Settings - Control Panel - System - Advanced tab - click Settings button (under Performance) - Data Execution Prevention tab.

2. Select the 2nd option (Turn on DEP for all programs...) and in the exception list add

a) DLLHOST.EXE (located in c:\windows\system32) for SurfStatsLive

or

b) SurfStats.exe (located in c:\Program Files\SurfStats<Version>) for SurfStats Website Traffic Analyzer.

Please Note: If you are running SurfStats Website Traffic Analyzer V8.4 or later add surffox8400.exe to the DEP exception list. If you are running V3.4 or later of SurfStatsLive please add surffoxlive3400.exe to the DEP exception list. For previous version follow the instructions above.

Long Answer...

1. Verify the Administrator account is running for Windows XP. For Windows 2003 log in with the Administrator account. 
2. Go to Start->Control Panel and choose the System applet.
   
   [XP Home users may need to choose the "Classic View" link to be able to see the System applet.]
3. Click on the Advanced tab.
4. Select the Settings button under Startup and Recovery.
5. Locate the line that specify what the default operating system is, which is similar to the following:
   "Microsoft Windows XP Professional" /NoExecute=OptIn .........  or
   "Windows Server 2003, Web" /NoExecute=OptIn.......
   (depending on whether you use Win XP or Win 2003).
6. Locate and change the boot.ini, by clicking on the Edit button, and then change /NoExecute=OptIn to /NoExecute=AlwaysOff. 
7. Save the file and reboot. 

An alternate way to solve the DEP problem on Win XP/Win 2003 is to go to Settings - Control Panel - System - Advanced tab - click Settings button (under Performance) - Data Execution Prevention tab. Select the 2nd option (Turn on DEP for all programs...) and in the exception list add DLLHOST.EXE (located in c:\windows\system32) if you are running SurfStatsLive or SurfStats.exe (located in c:\Program Files\SurfStats<Version>) if you are running SurfStats Log Analyzer. 

  

The SurfStats IP Visitor Database can be downloaded from www.surfstats.com/demo/sdbase.zip

Note that this zipped file is about 40 MB in size. We suggest that you use a download manager like Getright that can do resumed downloads to get it.
Unzip the downloaded file, read the readme file for installation instructions and run setup.exe to install. Then activate the database with the activation code.