Server upgrade faults and Solutions

Upgrade principles

1. The purpose of server upgrade is to solve the bottleneck in one aspect of the system, but the upgrade result cannot cause new system bottlenecks.

During the upgrade, you should consider the bottleneck of the server. However, in terms of applications, other performance bottlenecks cannot be caused by simple upgrading of a certain part. For example, a single CPU server can configure MB of memory, but if the Dual CPU server is configured with MB of memory, the memory will definitely become a bottleneck of the system.

2. The premise of server upgrade is that the system hardware configuration can provide space for user upgrade.

Server upgrades are performed on the premise that the server hardware configuration can be upgraded. Therefore, in the initial stage of network planning, network administrators should have an initial estimate of application requirements and growth, when you select a model, you can reserve a certain amount of space for future expansion.

3. The upgrade must be tailored to specific applications, rather than simply adding hardware.

The application of a server is usually dedicated for a special plane. A server is mainly used for one type of application. In this way, when the server has a performance bottleneck, the upgrade should be for this application, instead of targeting a certain component. It cannot be considered that simply upgrading a single hardware can meet the performance requirements of applications.

Upgrade fault and Solution

Upgrades can be divided into online and offline upgrades.

1. Online upgrade: When the system is not shut down, the system will not be affected by upgrading a certain part. On the IA architecture server, upgrading the hot swapping hard disk and migrating the RAID level do not require a system restart.

2. Offline upgrade: the system needs to be shut down, and the upgrade needs to be performed without being electrified. This mainly refers to upgrading the CPU and memory. For example, we recommend that you reinstall the system to upgrade the CPU. During CPU expansion, especially when a single CPU is extended to multiple CPUs, the system needs to re-install the SMP kernel. In this way, the UNIX system needs to upgrade the system kernel and re-compile it. For Windows systems, you need to reinstall the operating system. Another example is the memory upgrade. After the memory is increased, the system performance parameters and database application parameters must be adjusted.

Upgrade case

A single-CPU Database Application Server supports dual-CPU Systems). As the data volume increases and the number of clients increases, the system needs to be upgraded and the Unixware operating system is installed. How can this problem be solved?

According to your application analysis, you need to upgrade many components in the system, such as CPU, memory, and hard disk. In this case, the upgrade should include the following steps:

1. Back up system data to prevent system data loss due to misoperations ).

2. Power-off: the hardware is upgraded offline, and the CPU, memory, and hard disk are installed. Ensure that the specifications of the hardware are the same as those of the original motherboard.

3. Restart the operating system and add the CPU License.

4. Use the RAID feature to integrate new hard disks.