从Amazon S3备份 [英] Backup from Amazon S3

问题描述

我刚刚完成设置女将S3作为CDN为我们的网站。从现在开始，我们将举办万吨画面直接在云中。很酷！ - 但留下我一个问题，关于备份。早些时候，我们备份了一切通过上传图片，每天到Amazon一次，但我真的不希望备份到同一个地方，我主机上的文件。

I have just finished setting up Amazone S3 as a CDN for our website. From now on we will host tons of picture directly in the cloud. It is cool! - but leave me with a problem in regards to backup. Earlier we backup up everything by uploading the pictures to amazon once a day, but I really don't want to backup to the same place I host the files.

如何从女将S3在最有效的方式做我的备份？

我已经考虑：

• 拉出所有的S3文件到外部FTP服务器（这是昂贵的 存储和I将备份相同的文件一次又一次）。
• 在拉修改/添加S3文件到FTP服务器（我将需要监控 哪些文件被修改/添加，而这将需要工作半天 备份）
• 拉动文件到其他云服务（与缺点如上）

• Pull all the S3 files to an external ftp server (it is expensive in storage and I will backup the same files again and again).
• Pull modified/added S3 files to to an ftp server (i will need to monitor what files are changed/added, and it will require a half day of work to backup)
• Pull the files to an alternative cloud service (same cons as above)

任何解决方案，我现在不呢？如 myrepono.com 一个解决方案女将S3将是很酷。

Any solution I do not now of? A solution like myrepono.com for Amazone s3 would be cool.

干杯，彼得·

推荐答案

一个并不是由Amazon S3目前解决的主要事情是一个历史或回收站为删除或修改文件的任何概念。这样做的原因是没有解决亚马逊的脆弱性服务的失败，而是由别人来解决恶意的或意外删除的文件的访问凭据，并没有任何恢复选项。

One of the main things not addressed by Amazon S3 at the moment is any concept of a history or recycle bin for files that are deleted or modified. The reason for this is not to address Amazon's vulnerability to failure of the service, but to address malicious or accidental removal of files by someone with the access credentials, and not having any recovery option.

此单点故障仍是一个weakspot为AWS服务，甚至对那些具有快照功能（RDS，EC2但不S3）中，它仍是一个问题，因为与访问恶意用户可以用相同的删除快照单一入口点。

This single point of failure still is a weakspot for AWS services, and even on ones with snapshot capability (RDS, EC2, but not S3), it still is a problem because a malicious user with access could remove snapshots with the same single entry point.

您有下列情形之间的区别： 1）哎呀，基础设施失败了，我失去了我的档案！ （不太可能与S3） 2）哎呀，我（或客户）意外删除的目的，该文件，但我想回去！ 3）哎呀，我的系统API的脚本应该只删除一个文件，但由于编程错误，它循环通过所有的人，并删除了所有的人！ 4）哎呀，有人则获得了我的AWS帐号，删除了我的档案！

You have to differentiate between the following scenarios: 1) Oops, the infrastructure failed and I lost my file! (very unlikely with S3) 2) Oops, I (or a client) accidentally deleted that file on purpose but I want it back! 3) Oops, That API script on my system was supposed to only remove one file, but due to a programming error it looped through all of them and deleted all of them! 4) Oops, someone got access to my AWS account and deleted my files!

1是非常低的担心基于S3是如何工作的。 2你可以通过系统利用回收桶故意删除程序周围。但现在3和4留给你真正的和实质性的数据丢失，并且没有一个解决方案，在S3上本身能够解决这个问题。

1 is very low worry based on how S3 works. 2 you can program around by making systems utilize a recycling bucket for intentional deletions. But right now 3 and 4 leave you with real and substantive data loss, and don't have a solution on S3 itself that can solve it.

所以，你要么希望这最后的两件事情没有发生，或者你留下S3保持定期备份在其他地方，这在目前是非常麻烦的。

So you either have to hope these last two things don't happen, or you are left keeping periodic backups of S3 elsewhere, which at the moment is really cumbersome.

我认为最好的事情是亚马逊自动添加某种缺失保留作为一个功能，这意味着任何删除任何类型的可以48小时或东西被逆转这是一去不复返了。

I think the best thing would for Amazon to add some sort of deletion retention automatically as a feature, meaning that any delete of any kind could be reversed for 48 hours or something before it was gone for good.

这篇关于从Amazon S3备份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！