灾难恢复

Website and online network disasters have a variety of causes, from human error to device failure, to fire, floods and earthquakes. And whilst there is little way of predicting when they are going to happen, it’s Sod’s law that it will be when you least expect it and when it’s most inconvenient. So, to help, we’ve put together a list of tips to help you avoid disasters and to help you recover should the worst happen.

网站和在线网络灾难有多种原因，从人为错误到设备故障，再到火灾，洪水和地震。 尽管无法预测何时将要发生，但根据Sod的定律，当您最不期望时和最不方便时便会如此。 因此，为提供帮助，我们汇总了一些提示，以帮助您避免灾难并在最坏的情况下恢复。

Before we provide the solutions, it is helpful to understand exactly what we mean by disaster. When it comes to websites or online networks, a disaster is when you go offline. If you are not able to get back on again, then the situation can be catastrophic. In essence, there are only two things that can prevent you getting back online – either you have no alternative hardware available or your data and software have been lost.

在提供解决方案之前，准确了解灾难的含义是有帮助的。 当涉及到网站或在线网络时，灾难就在于您离线。 如果您无法再次恢复正常，则情况可能是灾难性的。 从本质上讲，只有两件事可以阻止您重新上网-要么没有可用的替代硬件，要么数据和软件丢失。

如何避免灾难 (How to avoid disaster)

When it comes to hardware failure, avoiding disaster means that you need to have alternative hardware and software applications in place that can kick in as soon as possible after the operating server goes down. It’s also important that you carefully monitor your system.

当涉及到硬件故障时，避免灾难意味着您需要拥有备用的硬件和软件应用程序，这些应用程序可以在运行服务器停机后尽快启动。 仔细监视系统也很重要。

服务器监控的重要性 (The importance of server monitoring)

Whilst you cannot tell when human error and environmental damage will happen, there are ways to monitor the health of your system. For example, eUKhost’s server monitoring will constantly monitor your server, network devices’ uptime, performance and available resources to ensure your server and network devices always remain online and in top condition. These monitoring systems are designed to check server load, CPU and RAM usage, disk IO performance, running services, partition usage, RAID health and much more, thereby helping to prevent disasters happening.

尽管您无法确定何时会发生人为错误和环境破坏，但仍有一些方法可以监视系统的运行状况。 例如， eUKhost的服务器监视功能将不断监视您的服务器，网络设备的正常运行时间，性能和可用资源，以确保您的服务器和网络设备始终保持联机并处于最佳状态。 这些监控系统旨在检查服务器负载，CPU和RAM使用情况，磁盘IO性能，正在运行的服务，分区使用情况，RAID运行状况等，从而有助于防止灾难的发生。

传统灾难恢复方法 (Traditional disaster recovery methods)

Using traditional disaster recovery methods, private network users will need to have a backup server primed and ready to go at all times. If you have a private network, this involves purchasing and maintaining other servers and ensuring that the software you need to run your system is in place. Ideally, these servers should be in a geographically different location from your main IT centre so that if the cause of you going offline is a flood or fire, the backup server will not be affected.

使用传统的灾难恢复方法，专用网络用户将需要准备好备用服务器并随时准备就绪。 如果您有专用网络，则需要购买和维护其他服务器，并确保安装了运行系统所需的软件。 理想情况下，这些服务器应与主要IT中心位于不同的地理位置，这样，如果您下线的原因是洪水或火灾，备份服务器将不会受到影响。

为了确保有完整的备份，您将需要以下内容： (To ensure you have a thorough backup in place, you will need the following:)

• secure premises to house the backup servers together with power and cooling安全的场所，以容纳备用服务器以及电源和散热装置
• enough capacity to scale the environment足以扩展环境的容量
• onsite IT support to maintain the infrastructure现场IT支持以维护基础架构
• an ISP contract that provides the backup site with enough bandwidth提供备份站点足够带宽的ISP合同
• firewalls, routers, switches and load balancers防火墙，路由器，交换机和负载平衡器
• capacity to run all your mission-critical services运行所有关键任务服务的能力

The downside to this is that it means you need to invest heavily in resources just in case of a failure. For the vast majority of time, these expensive backup resources will be redundant. If you have a dedicated server package with a web host, it is possible that they will have a mirrored server already in place at another site and the switch can be done relatively quickly without much of a delay and without the expense of having to run your own alternative system.

不利的一面是，这意味着您必须在发生故障时大量投资于资源。 在绝大多数时间内，这些昂贵的备份资源将是多余的。 如果您有一个带有Web主机的专用服务器软件包，则可能它们已经在另一个站点上安装了镜像服务器，并且可以相对快速地完成切换，而不会造成太多延迟，而无需花费运行服务器的费用。自己的替代系统。

云灾难恢复方法 (Cloud disaster recovery methods)

There are two ways you can use the cloud for hardware disaster recovery, depending on what kind of system you have. If you have a private network, you can use the cloud just as a backup system so that you do not need to go to the expense of investing in a second data centre. In this case, you can store your data in the cloud and should a disaster occur you can failover to a virtual machine to run your network until your own servers are restored. This will require you to configure your cloud environment in advance so that if you need to failover, getting back online can be done much quicker. One of the other advantages of cloud disaster recovery is that you can scale your requirements as needed, thus saving money over the long term.

您可以使用两种方法将云用于硬件灾难恢复，具体取决于您所使用的系统类型。 如果您有专用网络，则可以将云用作备份系统，这样您就不必花钱投资第二个数据中心。 在这种情况下，您可以将数据存储在云中，如果发生灾难，则可以故障转移到虚拟机以运行网络，直到还原自己的服务器。 这将需要您预先配置云环境，以便在需要进行故障转移时，可以更快地恢复联机。 云灾难恢复的另一个优势之一是您可以根据需要扩展需求，从而长期节省资金。

If your system is entirely hosted in the cloud, then the virtual machines that run your operations can be migrated between clustered servers to avoid any hardware issues that have been detected, all whilst maintaining the state of your VM.

如果您的系统完全托管在云中，则可以在群集服务器之间迁移运行您的操作的虚拟机，从而避免检测到任何硬件问题，同时保持VM的状态。

Here at eUKhost, for example, we leverage the latest technologies and hyper-converged enterprise cloud design, to produce a system that guards against anything from hard disk failure to an entire server failure. This helps us ensure that your mission critical applications will be available 100% of the time and that downtime will be a thing of the past.

例如，在eUKhost，我们利用最新的技术和超融合的企业云设计来生产一个可以防止从硬盘故障到整个服务器故障的系统。 这有助于我们确保您的关键任务应用程序将在100％的时间内可用，并且停机时间已成为过去。

数据丢失的灾难恢复 (Disaster recovery from data loss)

Besides hardware failure, the other cause of a disaster is data loss. Again, this can be caused by many things; accidental deletion, corrupted storage, hacking and even ransomware encryption. There are, however, a number of simple precautions and solutions to this problem.

除硬件故障外，灾难的另一个原因是数据丢失。 同样，这可能是由许多原因引起的。 意外删除，存储损坏，黑客甚至勒索软件加密。 但是，有许多简单的预防措施和解决方案。

解决方案：远程备份数据 (The solution: remotely backup your data)

Quite simply, if you back up your data and your software, then if you lose it, it is only a simple fix to restore your website and system. Obviously, for some organisations, there will be a need for continuous backups to take place. If you are an ecommerce site that is continuously taking orders from customers, then it is essential that you do not lose vital sales information.

很简单，如果您备份数据和软件，那么如果丢失了数据，那仅仅是恢复网站和系统的简单方法。 显然，对于某些组织而言，将需要进行连续备份。 如果您是一个不断接受客户订单的电子商务网站，那么一定不要丢失重要的销售信息。

Thanks to developments in technology, backing up data can now be automated, making it a much simpler process. eUKhost’s backup service provides a whole range of benefits including; continuous backup, high speed backup windows and file restoration, low server load, multipoint replication and industrial strength storage. As a result, if you system does go down, putting it back online can be achieved rapidly, keeping you trading and protecting your reputation.

由于技术的发展，现在可以自动备份数据，从而使备份过程变得更加简单。 eUKhost的备份服务提供了一系列好处，包括； 连续备份，高速备份窗口和文件还原，低服务器负载，多点复制和工业强度存储。 因此，如果您的系统确实出现故障，则可以快速实现将其重新上线，从而保持交易并保护您的声誉。

防止数据丢失 (Preventing data loss)

You can help prevent data loss from threats like malware, code injections and cross-site scripting attacks simply from using a service like an MTvScan. It is also possible to monitor your site for intrusion and to block those who are making attempts to break in using brute force and other techniques – using a secure firewall and an Intrusion Prevention System can make your network far more secure. A key part of defending against intrusion this is to ensure that you use strong passwords and have a robust password policy in force.

您可以通过使用MTvScan之类的服务来帮助防止由于恶意软件，代码注入和跨站点脚本攻击等威胁而造成的数据丢失。 还可以监视您的站点是否受到入侵，并阻止使用暴力破解和其他技术进行入侵的人 -使用安全的防火墙和入侵防御系统可以使您的网络更加安全。 防止入侵的一个关键部分是确保您使用强密码并具有有效的强密码策略。

Another way to protect data is to make sure that you only give access to it to those who need it. This limits the number of people who can accidentally delete or overwrite your data.

保护数据的另一种方法是确保仅将访问权限授予需要它的人。 这限制了意外删除或覆盖您的数据的人数。

结论 (Conclusion)

From reading this article you should now be aware that although disasters can happen both with data and hardware, there are ways to avoid some of the causes. Regardless of precautions, all businesses should have a disaster recovery plan in place that ensures that if the worst does happen you have alternative hardware and software and backed up data in place to ensure a swift return to normality. The most effective and cost efficient way to do this is to use cloud hosting or to have a cloud disaster recovery package together with remote data backups.

通过阅读本文，您现在应该知道，尽管数据和硬件都可能发生灾难，但是有一些方法可以避免某些原因。 无论采取何种预防措施，所有企业都应制定一个灾难恢复计划，以确保如果发生最坏的情况，您将拥有备用的硬件和软件，并已备份数据以确保Swift恢复正常。 实现此目的的最有效和最具成本效益的方法是使用云托管或将云灾难恢复软件包与远程数据备份一起使用。

翻译自: https://www.eukhost.com/blog/webhosting/disaster-recovery-and-how-to-avoid-needing-it/

灾难恢复