Shulou(Shulou.com)05/31 Report--

This article introduces how to achieve random recovery in MySQL, the content is very detailed, interested friends can use for reference, I hope it can be helpful to you.

1) the configuration of database parameters is not standard, and the configurations of / etc/my.cnf and / data/mysql_xxx/my.cnf do not match, resulting in the failure of instance startup

2) the database version is different, for example, the mainstream support is 5.7, and a 5.6 version suddenly appears.

3) an error occurred in binlog parsing, resulting in subsequent recovery failure

4) an error occurred in the recovery of the backup set, causing the overall recovery to fail

There are countless cases like this. If you are not careful, it will be difficult to recover. Problems like configuration can be solved, but in an emergency, the recovery process fails, and it is difficult to ensure that you have a good mentality that can be solved quickly. So testing the quality of recovery is a mistake we have been making in the past: we have been improving the backup, but we have paid little attention to the recovery side, and we think it should be possible. This is exactly what should put us in a passive position.

So I came up with an idea of random recovery, or suppose there are 500 instances, then if we restore these instances one by one, the daily workload is huge, and the load on the system is very high. So if we make a combination of risk and cost, the efficiency and significance of this work will be very obvious.

At present, the recovery is mainly based on backup set recovery, point-in-time recovery, object granularity recovery and table structure recovery. what we usually call system layer recovery is mainly based on backup set recovery and point-in-time recovery.

To this end, I have designed and implemented the following basic process:

It should be added that the random time is within the time period of the backup set, while the random timestamp is based on a random point in time within nearly 24 hours.

So many times random, can make the judgment of this matter will be more clear, the quality of recovery at a glance.

On this basis, a series of things are needed:

1) Random needs to ensure that all instances can be covered within a certain time range

2) Linear expansion of the recovery machine, such as providing a recovery server group, which can run some recovery tasks in parallel to improve the efficiency of recovery response.

3) Visualize the recovery results in the daily newspaper, what is restored and how efficient it is, and summarize and review the recovery results in a certain period of time.

4) according to the thinking of inference statistics, take a certain sample of data, through hypothesis testing, establish a corresponding data model for testing and analysis.

On how to achieve random recovery in MySQL to share here, I hope that the above content can be of some help to you, can learn more knowledge. If you think the article is good, you can share it for more people to see.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.