Shulou(Shulou.com)06/03 Report--

With the help of WebLog Expert, we can analyze the traffic records of the website, which is introduced in detail in this article.

WebLog Expert can analyze the traffic records of the website, and analyze the basic and important information such as Activity statistics, Access statistics, Information about visitors, Referrers, Information about errors and so on from the original traffic records, which can help us to understand and analyze the usage of the website more conveniently and quickly. This tool can analyze the logs of Apache and IIS servers and is very easy to use.

Next, from the installation of WebLog Expert to server log analysis, and then obtain the performance requirements to gradually unveil this tool.

The latest version of WebLog Expert software is available for download from http://www.weblogexpert.com/, and the installation process is very simple, so I won't repeat it here.

After the installation is complete, click [start]-[Program]-[WebLog Expert Lite]-[WebLog Expert Lite] to open the WebLog Expert tool and enter the WebLog Expert main interface, as shown in figure 1. Each row of records displayed in the list is an analysis object, and each record can be edited, deleted, and analyzed.

Figure 1

Now that we know the main interface of WebLog Expert, let's create a log analysis. Click the [New] button on the main interface, and the General dialog box (figure 2) will pop up and fill in Profile, Domain, and Index, respectively. Where Profile represents the name displayed in the main interface and analysis report; Domain is the domain name of the website; and Index is the file name of the front page of the analysis report.

Click the [next] button and the dialog box shown in figure 3 will be displayed. Enter the log file name for the analysis (you can use wildcards) in this dialog box, such as c:\ logs\ access.log or c:\ logs\ * .log.

、

Figure 2

Figure 3

Click "finish" and a new record will be added to the main interface shown in figure 1.

In the main interface, select the newly added record named "Weind Site" and click the [Analyze] button to start the analysis function. After a period of time, a more complete log analysis result file can be generated, as shown in figure 4.

Figure 4

The log analysis result file presents a lot of practical information, such as summary statistics, activity statistics, access statistics, visitor statistics and so on. This information can provide strong support for performance test cases and scenario design. Next, take the log file analysis results of the Apache server of the Software College of Hebei normal University as an example to explain the information in the analysis results.

The log file of my own unit's Apache server mainly records "access to the Bugfree defect management system". Figure 5 shows the summary statistics (General Statistics) of the system access log analysis report for the Bugfree system from December 11, 2008 to December 30, 2010.

Figure 5

General Statistics (summary statistics): make statistics on some basic performance indicators of the system, and provide some average and total values of the corresponding indicators. This information helps us to grasp the overall situation of the system in a certain period of time, such as: PV (page visits). As can be seen from figure 1.14, during the period from December 11, 2008 to December 30, 2010, a total of 5417 users visited the Bugfree system (Visitors), with 1670868 clicks (Hits) and 492092 page visits (Page Views). The total bandwidth uses 8.85GB.

Activity Statistics (activity Statistics): General Statistics provides the average and total values of some metrics during the operation of the whole system, but these averages and totals are often not enough for performance testing. We also need to look at some peak data, or prefer to see the value distribution of each indicator in different time periods. Activity Statistics can provide daily or even hourly metrics, such as daily user visits, daily clicks, hourly user visits, hourly clicks, and so on. The charts generated from these data reflect the distribution trends and peaks of related indicators. Figures 6 and 7 show the Activity Statistics information of the Bugfree defect management system of my work unit. It is easy to see the distribution trend and peak of each index from the chart, and easily analyze the peak number of a certain day or a certain period of time. This data is useful in the design of use cases and scenarios.

Figure 6

Figure 7

Access Statistics (access Statistics): Activity Statistics provides daily and hourly metrics data, focusing on the time distribution of metrics, while Access Statistics reflects the number of visits to businesses or pages. The Access Statistics menu supports statistical classification by pages, files, pictures, directories and entry pages in the system. Access Statistics can help us to identify the main business points of performance testing (performance testing is different from functional testing, which can not carry out performance testing for all functions of the whole system, which costs manpower and material resources and has no practical significance), so as to simulate the operation of a large number of users more truly and effectively. Figure 8 shows the Access Statistics information of the Software Academy's Bugfree defect management system (here there are too many diagrams, but only a partial demonstration). The "daily page visit trend" and "the most popular visit page" can be obtained from the chart, which is of great reference value to the use case and scenario design of performance testing.

Figure 8

Visitors (user Statistics): provides the index values of clicks, user visits and bandwidth classified and counted according to IP, as shown in figure 9, which can help us better analyze the type and bandwidth of visiting users.

Figure 9

The above describes the common charts for performance requirements analysis using WebLog Expert. In fact, there are many other charts in WebLog Expert, such as Referrers (submission Statistics): it can provide data on the distribution of indicators such as the hottest sites and hottest URL; and Browsers (browser Statistics): it can provide the allocation proportion of various browsers and operating systems used by users, so that readers can more truly simulate the use of users in performance tests. Errors (error statistics): can provide statistics of errors occurred during the operation of the system, through this information can help readers locate the problems existing in the system, and then effectively avoid the problems in the new version of the system by analyzing the causes of the problems.

In short, the WebLog Expert tool is easy to use and powerful, and can help readers analyze system logs effectively, and then better carry out performance requirements analysis. It is recommended that readers use this tool to conduct a performance requirements analysis in conjunction with a log file, so as to gain a deeper understanding of the meaning of each chart.

Note:

In addition to WebLog Expert tools, there are many good log analysis tools on the market, which can help us analyze and statistics different log files, and then help us to easily obtain performance testing requirements. Here is a summary of some commonly used software to share with you, interested readers can carry out research and analysis. 1) Analog: open source software, free; 2) AWStats: open source software, free; 3) Webalizer: open source software, free; 4) Summary: commercial software, 30-day trial version; 5) WebTrends: commercial software.

At this point, there will be a paragraph on a series of topics on "performance requirements analysis". I hope you can communicate more.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.