How to manage servers for routine software operation

Everyone should have this situation, that is, when software development is all done to develop the software. Once the software goes online and runs, we can relax and manage the server and data. At this time, a minor fault on the server can keep us busy for a while. After having gone through similar experiences, I feel more and more important about server management. How can we manage servers more effectively without dedicated personnel like ours, to ensure the smooth operation of enterprise information systems, there is no major problem. Here we will discuss the actual situation of such an enterprise environment, which is not a professional server management.

 

1. Server SETUP Configuration

The general situation is who develops the software and who will do the job. The developer builds the server and runs the software after installation, in the future maintenance work, it is very likely that there will be no one to manage, or no one will constrain this management. If there is no problem, no one may think of this server for a few months.

Therefore, in order to ensure that this work can be carried out in an orderly and stable manner, a detailed installation and configuration document should be prepared at the beginning of the server establishment to form an archive and distribute it to relevant management personnel. Determine daily management personnel and clarify their daily work content, such as database backup and solutions to common problems.

2. Disaster Recovery Plan

It is said that "if there is backup, there is no problem ". Only by constantly thinking about the importance of data and timely backup can we be able to be prepared. In addition, this plan should set a reasonable backup interval based on the actual business application of the software and regularly check and verify the backup data to ensure that the backup data is effectively available.

In addition, you must archive and query the configuration information of the backup files, test and record the backup files in a timely manner in case of problems or recovery.

3. Raid x

For servers, RAID 5 is the most commonly used form of data backup. The advantage of this backup is that the failure of three hard disks does not affect the normal operation of servers, replace the same hard disk with a bad one. However, after several verifications, it is difficult to ensure that the system is completely normal. For example, it takes a long time for a machine to be purchased, and it is difficult to match hard disks of the same model and capacity. This cannot be replaced, or the system crashes. It is a problem to install the system on the old server. Therefore, I still have a lot of questions about whether to perform raid. If there are not many users using the system, try to apply the RAID 0 + swap disk format, the two disks back up each other and regularly back up data to the swap disk through the software. In addition, if a problem occurs on the raid disk, the system can automatically replace it. RAID5 can also be operated in the same way, that is, when buying a server, do not buy only three disks, so that you can leave too little room for operation.

4. Backup Machine

Speaking of the backup machine, this is a luxury. Of course, this machine is not just waiting for the backup to be idle. It can also run small applications, and it can be topped up at critical moments. Recently, a server cannot be started. Fortunately, a server of the same type has taken down the hard disks of the faulty machine (RAID5) and placed them on a good machine, import the configuration from the raid configuration, OK, and the system can normally read the data in the hard disk. The data is the truth. The raid information was written in the raid card and hard disk at the same time, when the hard disk is placed on a new machine, the raid configuration information of the hard disk can be read out because the RAID card is of the same model. If the hardware is broken, you can repair it. If the data is lost, it is a big event. Therefore, if you have such a backup machine (preferably of the same model), you can leave one more choice for yourself.

5. Routine Management records

For normal server management, it is best to have a good management record so that relevant information can be found at critical moments. This requires the Administrator to develop good work habits. It is best to keep this document open to some extent, so that others can obtain relevant information when necessary.

How to manage servers for routine software operation