Advanced Log Parser Charts Part 1 - Working With Configuration Scripts

本文涉及的产品
日志服务 SLS,月写入数据量 50GB 1个月
简介: http://blogs.msdn.com/b/carloc/archive/2008/08/07/charting-with-logparser.

http://blogs.msdn.com/b/carloc/archive/2008/08/07/charting-with-logparser.aspx

I recently had a situation where I wanted to customize the chart output from Log Parser, and after a bunch of research I eventually arrived at the conclusion that configuration scripts for create customized charts are probably the least-documented feature of Log Parser. After a lot of experimentation, (and a bit of frustration), I finally managed to achieve the results that I wanted. With that in mind, I thought that it would make a great blog series if I documented some of the settings that I used.

Log Parser and Chart Configuration Scripts

When you look in the Log Parser help file, it makes mention of using configuration scripts to customize charts, and it provides the following small JavaScript sample:

// Add a caption
chartSpace.HasChartSpaceTitle = true;
chartSpace.ChartSpaceTitle.Caption = "Generated by Log Parser 2.2";
chartSpace.ChartSpaceTitle.Font.Size = 6;
chartSpace.ChartSpaceTitle.Position = chartSpace.Constants.chTitlePositionBottom;

// Change the background color
chart.PlotArea.Interior.Color = chartSpace.Constants.chColorNone;

Unfortunately, this sample isn't very useful, although I found dozens of forum posts that quote this sample as a way to do things - but it's the only sample that most people cite. The Log Parser help file mentions looking at the MSDN ChartSpace Object Model documentation, but that documentation is only slightly more useful. These two references are what led me to my earlier conclusion that chart configuration scripts are not well-documented, and especially when you are trying to do something with Log Parser.

What I found to be particularly helpful was to use the Log Parser COM interface and write scripts by using Adersoft's VbsEdit and JsEdit. In case you haven't used either of those applications, they are great IDEs for writing scripts; they both give you a great debugging environment, and they have a great object browser that I used to discover what options were available to me. In the end, these two editors made it possible to create the chart configuration scripts that I will discuss in this blog series.

By the way, chart configuration scripts can be written in VBScript or JavaScript, but for this blog I will use VBScript for the Log Parser COM samples and JavaScript for the configuration script samples. I didn't have to do it that way, but it seemed like a good idea to help differentiate between the samples.

Using COM versus the Command-Line

For the samples in this blog series, I will use Log Parser's COM interface and VBScript to create my charts, but this is not necessary; everything that I am documenting can be done from the command-line version of Log parser, and I'll give you some quick examples to see the differences.

The following examples generate some simple area charts that plot the total number of hits by day, and both examples do exactly the same thing:

Command-Line:

logparser.exe "SELECT Date, COUNT(*) AS Hits INTO HitsByDay.gif FROM *.logGROUP BY Date ORDER BY Date" -i:W3C -fileType:GIF -groupSize:800x600-chartType:Area -categories:ON -values:ON -legend:OFF

COM Interface:
Option Explicit

Dim objLogQuery, strSQL
Dim objInputW3CFormat, objOutputChartFormat

Set objLogQuery = WScript.CreateObject("MSUtil.LogQuery")
Set objInputW3CFormat = WScript.CreateObject("MSUtil.LogQuery.W3CInputFormat")
Set objOutputChartFormat = WScript.CreateObject("MSUtil.LogQuery.ChartOutputFormat")

strSQL = "SELECT Date, COUNT(*) AS Hits " & _
    " INTO HitsByDay.gif " & _
    " FROM *.log " & _
    " GROUP BY Date " & _
    " ORDER BY Date"

objOutputChartFormat.groupSize = "800x600"
objOutputChartFormat.fileType = "GIF"
objOutputChartFormat.chartType = "Area"
objOutputChartFormat.categories = "ON"
objOutputChartFormat.values = "ON"
objOutputChartFormat.legend = "OFF"

objLogQuery.ExecuteBatch strSQL, objInputW3CFormat, objOutputChartFormat
Ugly Charts

Using some of the log files from one of my websites, the above samples created the following basic chart:

Taking a look at this chart makes it easy to see why you would want to customize your output; that light blue is pretty awful, and those values are pretty hard to read.

Specifying Configuration Scripts

If you remember the incredibly basic configuration script from earlier, you only need to add one parameter to each example in order to specify the configuration script:

Command-Line:

logparser.exe "SELECT Date, COUNT(*) AS Hits INTO HitsByDay.gif FROM *.logGROUP BY Date ORDER BY Date" -i:W3C -fileType:GIF -groupSize:800x600-chartType:Area -categories:ON -values:ON -legend:OFF -config:HitsByDay.js

COM Interface:
Option Explicit

Dim objLogQuery, strSQL
Dim objInputW3CFormat, objOutputChartFormat

Set objLogQuery = WScript.CreateObject("MSUtil.LogQuery")
Set objInputW3CFormat = WScript.CreateObject("MSUtil.LogQuery.W3CInputFormat")
Set objOutputChartFormat = WScript.CreateObject("MSUtil.LogQuery.ChartOutputFormat")

strSQL = "SELECT Date, COUNT(*) AS Hits " & _
    " INTO HitsByDay.gif " & _
    " FROM *.log " & _
    " GROUP BY Date " & _
    " ORDER BY Date"

objOutputChartFormat.groupSize = "800x600"
objOutputChartFormat.fileType = "GIF"
objOutputChartFormat.chartType = "Area"
objOutputChartFormat.categories = "ON"
objOutputChartFormat.values = "ON"
objOutputChartFormat.legend = "OFF"
objOutputChartFormat.config = "HitsByDay.js"

objLogQuery.ExecuteBatch strSQL, objInputW3CFormat, objOutputChartFormat
Simple Output

Taking a look at the resulting chart, you can see why I mentioned earlier that the configuration script wasn't very useful; all it does is add a centered title to the bottom of the chart:

Yup - that's a pretty useless sample configuration script for chart customization.

Next...

In my subsequent posts, I'll show how to make this chart (and several other types of charts) look a lot better.

相关实践学习
日志服务之使用Nginx模式采集日志
本文介绍如何通过日志服务控制台创建Nginx模式的Logtail配置快速采集Nginx日志并进行多维度分析。
目录
相关文章
|
监控
用Log Parser Studio分析IIS日志
发现一个强大的图形化IIS日志分析工具——Log Parser Studio,下面分享一个实际操作案例。 1. 安装Log Parser Studio a) 需要先安装Log Parser,下载地址:http://www.microsoft.com/en-us/download/details.aspx?displaylang=en&id=24659 b) 安装Log Parser Studio,下载之后解压即可。
1889 0
|
Windows SQL
windows管理员利器之用Log Parser Studio分析IIS日志(附逐浪CMS官方命令集)
原文:windows管理员利器之用Log Parser Studio分析IIS日志(附逐浪CMS官方命令集) Log Parser Studio是一个强大的IIS图形分析工具,值得推荐。
2317 0
Microsoft Log Parser Toolkit
Microsoft Log Parser Toolkit
689 0
|
20天前
|
XML 安全 Java
【日志框架整合】Slf4j、Log4j、Log4j2、Logback配置模板
本文介绍了Java日志框架的基本概念和使用方法,重点讨论了SLF4J、Log4j、Logback和Log4j2之间的关系及其性能对比。SLF4J作为一个日志抽象层,允许开发者使用统一的日志接口,而Log4j、Logback和Log4j2则是具体的日志实现框架。Log4j2在性能上优于Logback,推荐在新项目中使用。文章还详细说明了如何在Spring Boot项目中配置Log4j2和Logback,以及如何使用Lombok简化日志记录。最后,提供了一些日志配置的最佳实践,包括滚动日志、统一日志格式和提高日志性能的方法。
157 30
【日志框架整合】Slf4j、Log4j、Log4j2、Logback配置模板
|
1月前
|
XML JSON Java
Logback 与 log4j2 性能对比:谁才是日志框架的性能王者?
【10月更文挑战第5天】在Java开发中,日志框架是不可或缺的工具,它们帮助我们记录系统运行时的信息、警告和错误,对于开发人员来说至关重要。在众多日志框架中,Logback和log4j2以其卓越的性能和丰富的功能脱颖而出,成为开发者们的首选。本文将深入探讨Logback与log4j2在性能方面的对比,通过详细的分析和实例,帮助大家理解两者之间的性能差异,以便在实际项目中做出更明智的选择。
246 3
|
3月前
|
Kubernetes Ubuntu Windows
【Azure K8S | AKS】分享从AKS集群的Node中查看日志的方法(/var/log)
【Azure K8S | AKS】分享从AKS集群的Node中查看日志的方法(/var/log)
136 3
|
1月前
|
存储 缓存 关系型数据库
MySQL事务日志-Redo Log工作原理分析
事务的隔离性和原子性分别通过锁和事务日志实现,而持久性则依赖于事务日志中的`Redo Log`。在MySQL中,`Redo Log`确保已提交事务的数据能持久保存,即使系统崩溃也能通过重做日志恢复数据。其工作原理是记录数据在内存中的更改,待事务提交时写入磁盘。此外,`Redo Log`采用简单的物理日志格式和高效的顺序IO,确保快速提交。通过不同的落盘策略,可在性能和安全性之间做出权衡。
1646 14
|
1月前
|
Python
log日志学习
【10月更文挑战第9天】 python处理log打印模块log的使用和介绍
35 0
|
1月前
|
数据可视化
Tensorboard可视化学习笔记(一):如何可视化通过网页查看log日志
关于如何使用TensorBoard进行数据可视化的教程,包括TensorBoard的安装、配置环境变量、将数据写入TensorBoard、启动TensorBoard以及如何通过网页查看日志文件。
214 0
下一篇
无影云桌面