Google stopped sales and support of Urchin on March 31, 2012. Visit Web Metrics Review to discover the changes coming to the Webmtrix Service.
Purdue Marketing and Media has offered to assist user in setting up Google Analytics and/or Google Custom Search
Visit webmetrix.itap.purdue.edu to view the current web log analysis service.
Frequently Asked Questions
Under the Sheets
Provides web log analysis using Urchin version 6 for web server domains hosted by IT@P SO.
This description will leave many things out that are not relevant to finding and understanding web log file analisys.
A web page is typically made up of many components. The web page may be generated in the web browser (as in HTML format) or on the web server and sent as a single file (as in MHTML format). For this exaple, the browser requests URL www.purdue.edu/cio/employement.
- The internet identifies machines by numbers. Thus, if the requested domain name is www.purdue.edu and is not on the Browsers cache, Browser contacts Domain Name Server to retrieve IP address for www.purdue.edu. Therefore, IP addresses are slightly more reliable metrics than domain names. For www.purdue.edu returns the IP address of 22.214.171.124.
- The browser sends a request to 126.96.36.199 for domain www.purdue.edu for the uri /cio/employment/
- The request arrives athe the F5 (188.8.131.52). The F5 looks at the domain name (www.purdue.edu) and URI (/cio/employment) to decide how to direct the request. For this example, the request is sent to an IIS/ColdFusion web server. This is why you will see multiple profiles for www.purdue.edu, some with extentions like ASP-CFMX. Your web developer can help you decide which profile(s) to select.
- The request arrives at the IIS web server. The IIS web server then;
- IIS goes to the www.purdue.edu domain instance
- IIS starts to gather information for the log file *
- IIS goes to the cio directory and retrieves the default or index file
- IIS looks at the file extension and checks the web server’s MIME type table to see how to process the file, a few examples;
- For HTML styl pages, the requested file is returned. (EX: .html)
- For MHTML style pages, the web page is built on the server and returned. (EX: .cfm)
- For SHTML pages, part of the page is built on the server returned. (EX: .shtml)
- NOTE: Your web developer can tell you how your pages are designed.
- The IIS log file record is created with the following information; (NOTE: Each log record is called a hit. There will be 1 to many hits to create a web page.)
- Date & time request was received
- Client IP Address (typically the browser workstation)
- User name (if authenticated/login during the current session, the login user name or blank)
- Service Name (name used by IIS to distinguish from other web services running on the server)
- Server IP address (web server IP address, for cluster load balancing metrics)
- Server Port (typically 80, SSL encrypted = 443, other = specific to application)
- Method (type of request, typically GET)
- URI stem (the part of the URL after the domain name, for this example = /cio/employment)
- URI query (used by forms and applications to pass information)
- rotocol Status (code for successful, file not found, etc.)
- Protocol Substatus (code to provide more detail on Protocol Status)
- Win 32 status (Operating system code, sometimes helpful in diagnosing problems)
- Bytes sent (Size of HTML response sent to browser)
- Bytes received (Size of request from Browser)
- Time Taken (length of time on web server between request received and response sent)
- Protocol Version (HTTP version of request received)
- Host (domain name, for this site = www.purdue.edu)
- User Agent (web browser or service name that sent the request)
- Referer (if a web page link was clicked to send this request, URL of that web page)
- The web browser receives the HTML and starts to display the page. For HTML & SHTML style pages, the web repeating the above steps until all pieces of to build the page are retrieved, creating multiple requests in the log file.
In this example, the /cio/employment pages link to the Taleo servers for individual job descriptions. Those log records are at Taleo and are not available for our reporting.
Daily, the web server rotates the log files and webmetrix pulls a copy of the previous days log files to the webmetrix machine for processing.