谷歌应用引擎请求日志细分 [英] google app engine request log breakdown

问题描述

以下是管理控制台中的一个python appengine应用程序的示例行，其中的日志部分已过滤为仅请求：

  01-22 07:19 PM 55.483 / zone / 5 200 82ms 161cpu_ms 115api_cpu_ms 0kb Mozilla / 5.0（Windows; U; W ... 
  

有什么专栏的解释吗？

这是我很好奇的：

第3栏：55.483 - 我认为这是日期时间中的秒数，但在'PM'之后很奇怪。

起初，我认为是时候为响应提供服务了 - 但是在这里82ms小于161cpu_ms和115api_cpu_ms，那么这次是什么时候？第9栏：0kb - 我假设这是响应的大小，对于这个请求，我们的响应小于1kb，所以我假设它们正在四舍五入。

解决方案

<第6列：是服务响应的时间.CPU时间较长的原因是因为许多操作s是并行的，你在做什么？ CPU是用于生成请求的总CPU。

第9列：可能会减少舍入和更多整数截断。

Here is an example line from one of python appengine apps in the admin console Logs section filtered to 'Request Only':

01-22 07:19PM 55.483 /zone/5 200 82ms 161cpu_ms 115api_cpu_ms 0kb Mozilla/5.0 (Windows; U; W...

Is there an explanation somewhere of what the columns are?

Here's what I'm curious about:

Column 3: 55.483 - I assume this is seconds in the datetime, but it is strange to be after the 'PM'

Column 6: At first I thought it was the time to serve the response - but here 82ms is less than 161cpu_ms and 115api_cpu_ms, so what is this time?

Column 9: 0kb - I assume this is the size of the response. Our response is less than 1kb for this request, so I assume they are rounding down.

解决方案

Column 6: is the time to serve the response. The reason CPU time is greater is because many operations are parallelized. What are you doing in this request? The CPU is the total CPU being used to generate the request.

Column 9: It's probably less rounding down and more integer truncation.

这篇关于谷歌应用引擎请求日志细分的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！