A method and device for automatic maintenance of ultra-large-scale machines

Abstract

The purpose of the present invention is to provide a method and device for automatic maintenance of ultra-large-scale machines. Compared with the prior art, the present invention collects soft and / or hardware faults in ultra-large-scale machines; performs fault analysis on the soft and / or hardware faults to obtain corresponding fault data; based on the fault data, adopts maintenance status machine, complete the automatic maintenance of the ultra-large-scale machine for each state flow, wherein, the machine corresponding to the data that needs to be migrated is migrated and the whole machine is repaired, and the machine corresponding to the storage service is repaired online. For ultra-large-scale machines, the present invention can meet requirements such as fault detection, service migration, environment deployment, machine maintenance status flow, fast delivery, etc., saving manpower for operation and maintenance, and saving machines by improving flow efficiency; detection, maintenance, service migration and deployment are realized All automation does not require manual access; machine delivery is highly efficient and can achieve hourly and minute-level delivery.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a technology for automatic maintenance of ultra-large-scale machines. Background technique [0002] Existing machine maintenance is generally divided into the following scenarios: [0003] 1) When it takes shape (dozens of units), it is generally maintained and delivered by operation and maintenance personnel through monitoring and manual monitoring; [0004] 2) Medium-scale and large-scale (scale of 100 units, scale of 1,000 units) are generally realized through monitoring + scripts, or small automation systems. [0005] However, for ultra-large-scale (tens of thousands, hundreds of thousands) machines, there are problems such as labor costs and maintenance delivery efficiency. [0006] The existing automatic maintenance implementation schemes generally have the following types: [0007] 1) Script-based maintenance system: Generally, it is a way to deal with small-scale clu...

AI Technical Summary

Problems solved by technology

[0007] 1) Script-based maintenance system: Generally, it is a way to deal with small-scale clusters. This kind of cluster may not even be fully realized in virtualization. Generally, the system is simple and easy to operate through monitoring, migrating services through deployment tools, and triggering service API commands. Development, but the disadvantage is that there is no fixed collection and analysis system, and maintenance and deployment are generally aimed at simple scenarios with simple functions and cannot be applied to large-scale systems

[0008] 2) Trigger maintenance system: It can also be called a semi-automatic maintenance system. Generally, there is an independent collector to collect and classify faults, and then there is an independent fault pool and maintenance push system. This system meets the needs of large Part of the maintenance system needs, but the disadvantage is that there is no independent service migration and docking service, and the user needs to get the active push of the fault if there is a problem, and there is a lack of interactive process

[0009] However, these existing maintenance schemes cannot meet the diversity, nor can they meet the needs of ultra-large-scale machines. Most maintenance systems are relatively unified for models, systems, and environments

Images