简体繁体 English

Graphite的示例磁盘空间，CPU和/或内存监视配置

[英]Example disk space, CPU and / or memory monitoring configuration for Graphite

原文 2018-09-20 05:43:21 6 1 graphite/ graphite-carbon

We are looking for a simple monitoring tool for basic stuff like disk space, CPU, folder sizes, memory usage. 我们正在寻找一种简单的监视工具来监视基本内容，例如磁盘空间，CPU，文件夹大小，内存使用情况。

Graphite looks promising. 石墨看起来很有前途。 For a demo I want to create some example data to put in Graphite for one or more of such metrics. 对于演示，我想创建一些示例数据以将一个或多个此类指标放入Graphite中。

What is best way to approach this. 什么是解决此问题的最佳方法。 I have Graphite running in a Docker. 我有在Docker中运行的Graphite。 How do I configure Graphite and send some test data to it? 如何配置Graphite并向其中发送一些测试数据？ For example for 例如

daily disk space metrics 每日磁盘空间指标
daily folder sizes 每日文件夹大小
hourly CPU 每小时CPU
hourly memory 每小时的记忆

1 个解决方案

Graphite is not a data collector. 石墨不是数据收集器。 From graphiteapp.org : 来自石墨app.org ：

What Graphite is and is not. 什么是石墨，什么不是。

Graphite does two things: 石墨有两件事：

Store numeric time-series data 存储数字时间序列数据

Render graphs of this data on demand 按需呈现此数据的图形

Graphite is not a collection agent, but it offers the simplest path for getting your measurements into a time-series database. Graphite不是收集代理，但是它提供了将测量结果放入时间序列数据库的最简单途径。 Feeding your metrics into Graphite couldn't be any easier. 将指标输入Graphite再简单不过了。

You will need a collector, Collectd and Telegraf seem to be popular choices at the moment but there are many others, see list of collectors . 您将需要一个收集器，“收集器”和“ Telegraf”目前似乎是最受欢迎的选择，但还有许多其他选择，请参见收集器列表。 Disclosure: I contributed to both projects so might be biased. 披露：我对这两个项目都做出了贡献，因此可能会有偏见。

Your intervals are very long, Graphite is usually used with way smaller intervals, 10s to 1m. 您的间隔非常长，通常以较小的间隔（10s至1m）使用Graphite。 I don't see why it won't work with intervals of hours though. 我不明白为什么它不能每隔几个小时工作一次。 Make sure to configure your storage-schemas.conf accordingly, the default setting is 1m. 确保相应地配置storage-schemas.conf，默认设置为1m。