Linux Operations (Application software principles) Interview topics

1. What is operation and maintenance? What is Game ops?

1) operation and maintenance refers to the large-scale organization has established a good network of hardware and software to maintain, is to ensure that the business of the line and operation of the normal,

In the process of his operation, he was maintained, he assembled network, system, database, development, security, monitoring in a technology

Operations include a number of types, DBA operations, website operations, virtualization operations, monitoring operations, game operations and so on

2) Game operation and division, divided into development operations, application operations (business operations) and system operation and maintenance

Development operation and maintenance: is to the application operations and maintenance tools and operation and maintenance platform

Application operations: is to the business on-line, maintenance and troubleshooting, with the development of operations developed tools to the business on-line, maintenance, troubleshooting

System operation and Maintenance: is to provide operational infrastructure for application operations, such as: system, network, monitoring, hardware and so on

Summary: DevOps and operational maintenance provides support for application operations in "Tools" and "infrastructure"

DevOps, application operations and systems operations their work is interlocking.

2, in the work, operation and maintenance personnel often need to deal with operators, what is the operator to do what work?

Game operation to do a thing in addition to the coordination of work, but also need to communicate with the platform, to open the service time, open the number of services, user guidance, activities and other programs

3, now give you 300 servers, how do you manage them?

How to manage 300 servers:

1) Set up a springboard, login with a unified account, easy to security and login considerations.

2) using salt, ansiable, puppet Unified management of the unified scheduling and configuration of the system.

3) Establish a simple server system, configuration, application of the CMDB information management. Easy access to various information records on each server.

4. Brief introduction to the working principle and characteristics of three kinds of working modes of RAID0 RAID1 RAID5

1) RAID0

How it works: continuous data is distributed across multiple disks for access

Features: The highest read and write performance, but does not allow the bad hard disk, a hard drive is broken, all the data will be lost.

2) RAID1

How it works: two drives mirror each other

Features: one hard drive for another hard disk backup, a hard disk after the broken, service and data are not affected. Half of total capacity

3) RAID5

How it works: continuous data and redundant data spread across multiple disks

Feature: Allow bad piece of hard disk, read and write performance based on RAID0 and RAID1, total capacity is n-1

5. Describe the advantages and disadvantages of Apache two modes of operation respectively.

Apache has two main modes of operation: prefork (default installation mode for Apache) and worker (can add parameter--with-mpm-worker to select working mode at compile time)

Prefork is characterized by: (Pre-derivation)

1) This mode reduces system overhead by eliminating the need to generate new processes when the request arrives

2) to prevent accidental memory leaks

3) automatically reduces the number of sub-processes when the server load drops

The characteristics of the worker are:

1) Multi-threading multi-process module with mixed multi-thread support

2) If the worker mpm is a good choice for a high-traffic HTTP server, the worker mpm consumes less memory than prefork.

6. Please write down the two working modes of the apache2.x version, and how to view the modules currently supported by Apache, and see which mode is working in?

Prefork process (multi-process, each process produces child processes)

Work (multi-process, each process produces multiple threads)

Httpd-l can view the current module of Apache. If with WORKER.C is working in worker mode, if there is

PREFORK.C is working in PREFORKK.C mode.

7, LVS, Nginx, haproxy What is the difference? How do you choose in your work?

LVS: is based on four-layer forwarding

HAproxy: is based on four-tier and seven-tier forwarding, is a professional proxy server

Nginx: is a Web server, cache server, but also reverse proxy server, can do seven-tier forwarding

The difference: LVS because it is based on four-layer forwarding so can only do port forwarding, and URL-based, directory-based such forwarding LVS can not do,

Work selection: Haproxy and Nginx because can do seven layer of forwarding, so the URL and directory forwarding can do, in a very large amount of concurrency when we have to choose LVS

Like small and medium-sized company words concurrency is not so big, choose Haproxy or nginx foot already, because Haproxy by is a professional proxy server, configuration is simple, so small and medium-sized enterprises recommend the use of Haproxy

8, Squid, Varinsh and nginx What is the difference between work how do you choose?

Squid, Varinsh and Nginx are proxy servers

What is a proxy server:

Users to access the public network, and can be access to the data cache to the server local, and so on the next time users access the same

Source, the proxy server directly from the local response to the user, when there is no local, I instead of you to access the public network, I answered

To receive your request, I will first look in my own local cache, and if I have a local cache, I will reply to you directly from my local cache,

If I do not find the cached data you want to access in my local, then the proxy server will replace you to access the public network

Difference:

1) Nginx is a reverse proxy/web server, with the plug-in can do this sideline, but it does not support many features, can only cache static files

2) from these functions. Varnish and squid are professional cache services, and nginx These are third-party modules to complete

3) Varnish's own technical superiority is higher than squid, it uses the Visual page cache technology, in the memory utilization, varnish than squid has the superiority, the performance is higher than the squid.

There is also a powerful management port through the varnish, you can use regular expressions to quickly and bulk clear part of the cache

It is a memory cache, the speed is first-class, but the memory cache also limits its capacity, cache pages and pictures are generally very good

4) Squid's advantage lies in the complete huge cache technical data, and a lot of application production environment

Choice of work:

To do the cache service, we must choose the professional cache service, prefer squid or varnish.

9. What is the difference between Tomcat and resin, and how do you choose to work?

Difference: Tomcat user number, can refer to more documents, resin fewer users, can consider less documentation

The main difference is that Tomcat is a standard Java container, but performance is worse than resin.

But stability and Java program compatibility, should be better than resin

Work choice: Now large companies are using resin, the pursuit of performance, and small and medium-sized companies are using Tomcat, the pursuit of stability and program compatibility

10. What is the difference between Apache and Nginx? Which one does the job choose? Why?

1) The advantages of Nginx compared to Apache:

Lightweight, same as Web service but consumes less memory and resources than Apache

concurrency, Nginx processing requests are asynchronous non-blocking, and Apache is blocking, in high concurrency, nginx can keep low resource consumption

Highly modular design, relatively simple to write modules

Community active, various high-performance modules export rapidly

2) The advantages of Apache with respect to Nginx:

Rewrite is stronger than Nginx's rewrite,

More modules, Nginx bug is relatively more

Ultra-Stable

3) Choice of work:

In general, if you need performance Web services, use Nginx, if you do not need performance as long as the stability of the Apache

11. What is middleware? What is a JDK?

Middleware Introduction:

Middleware is a kind of independent system software or service program, and distributed application software uses this software to share resources among different technologies. The middleware is located on the client/server operating system,

Manage computer resources and network traffic. Is software that connects two standalone applications or standalone systems. Connected systems, even if they have different interfaces, but through the middleware can still

Exchange information. One of the key ways to implement middleware is information delivery. With middleware, applications can work in multi-platform or OS environments.

JKD:JDK is the SDK for Java. It is a development environment for building applications, applets, and components that are published on the Java platform.

12, tell the meaning of Tomcat8005, 8009, 80,803 ports?

8005== "use when closed

8009== "for AJP port, which is used by the container, such as Apache to access Tomcat's 8009 port via the AJP protocol

8080== "General application use

13, tell the meaning of Resin8080, 6911, 69,213 ports?

8080== "Resin Web server port, if you do not need to Resin do Web server, you can turn off this configuration

6911== "true resin port, this must be up. In the future, in Apache, I'm going to throw the request to this port parsing.

6921== "Watchdog Port

14. What is CDN?

-Content Distribution Network

-The goal is to publish the content of the site to the web by adding a new layer of network architecture to the existing Internet

Closest to the user's network edge, enabling users to get the content they need, increasing the speed of access to the site

15, what is called the site grayscale publishing?

Grayscale Publishing is a way to publish a smooth transition between black and white. AB Test is a grayscale publishing method, allowing a user to continue with a, a subset of users start with B,

If the user has no objection to B, then gradually expand the scope and move all users to B. Gray-scale publishing can ensure the stability of the overall system, in the initial gray level can be found, adjust the problem to ensure its impact.

16. Describe the process of DNS for domain name resolution?

Users to access www.baidu.com, will first find the host file, and then find the local DNS server settings, if not, go to the network to find root server, root

Server feedback results, said only a first-level domain name server. cn, go find a domain name server, first-class domain name server

It says that you can only provide a level two domain name server. com.cn, go to the level two domain name server, two level domain server can only provide three level domain name server

. baidu.com.cn, to find the level three domain name server, three level domain name server just have such a site www.wsyht.com, and then

Send the requested server, save one copy, then send to the client

17, why integrate Apache and resin?

1) Early Resin,tomcat support for HTTP service is not good

2) Tomcat, resin support for rewrite, expires, gzip function is not good

3) in order to do static and dynamic separation

18. What is RABBITMQ?

RABBITMQ is the message queue middleware, the message middleware is the container that holds the message during the transmission of the message. Message middleware acts as a middleman when it bids a message from its source to its target.

The primary purpose of the queue is to provide routing and to guarantee the delivery of messages, and if the recipient is unavailable when the message is sent, Message Queuing does not retain the message until it can be successfully delivered, and, of course, Message Queuing holds the message for a period of time

19, talk about the working principle of keepalived?

In a virtual router, only the VRRP router as master will always send VRRP notification information,

Backup does not preempt master unless it has a higher priority. When Master is not available (backup does not receive notification information),

The highest priority in more than one backup will be preempted to master. This preemption is very fast (<1s) to ensure continuity of service.

Due to security considerations, the VRRP package uses the encryption protocol for encryption. Backup does not send notification messages and only receives notification information.

20. Describe the working process of LVS three modes.

LVS has three load-balanced modes, namely Vs/nat (NAT mode) VS/DR (route mode) Vs/tun (tunnel mode)

Vs/nat Mode Introduction

When this mode is used to access the network, the load balancing opportunity is distributed to application Server A or application Server B, respectively, through NAT translations. The application server returns to the external access,

To be accessed through the load machine again to external access. This mode is usually the application server is IP is a private IP address, you need to use a NAT address to convert to a public IP address.

VS/DR Mode Introduction

Routing mode, the load balancer and application server belong to an IP range, and the load-balanced virtual IP also belongs to this IP range, usually is the configured public IP address, this way we can adopt the route mode. When the external

Network to access, the first must be distributed through the load balancer to this set of application server, but should be the server, returned to the external network is not through this load balancer out, and itself through the route back to the outside

Department network, Go Anza by yourself, go back.

Vs/tun Mode Introduction

Tunnel mode, load balancer and cluster collection server and load balancer virtual out IP not belong to the same network segment, distributed in different network segments, this time I need tunnel mode, like this load balancer and this cluster set of services

The device is recommended to be used specifically for load-balanced tunnels, tunnel mode first outside the network access time, the load balancer to distribute to the application server, the application server is also the same way as the routing mode, the route itself to return the results to

External access, nor is it returned to the external network through a load balancer.

21, tell me about the working principle of Fastdfs?

Fastdfs is an open source lightweight Distributed File system that manages files, including file storage, file synchronization, file access (file upload, file download), etc.

Solves the problem of mass storage and load balancing. Especially suitable for document-based online services, such as photo album sites, video sites, etc.

Fastdfs server side has two roles, Tracker (tracker) and storage node (storage). The tracker mainly does the dispatching work, and the function of load balancing on the access.

Storage node stores files to complete all the functions of file management: storage, synchronization, and provision of access interfaces.

Both the tracker and the storage node can be composed of multiple servers. Servers in trackers and storage nodes can be added or offline at any time without impacting online services. where all the servers in the tracker are peers,

Can be increased or decreased at any time depending on the pressure of the server. To support large capacity, storage nodes (servers) are organized in a sub-volume (or grouping) manner. The storage system consists of one or more volumes,

The files between volumes and volumes are independent of each other, and the cumulative file capacity of all volumes is the file capacity of the entire storage system. A volume can consist of one or more storage servers, and the files in a storage server under a volume are the same.

Multiple storage servers in a volume play a role of redundant backup and load balancing.

22. When you want to share your things to your leader, how will you tell him about your share?

1. What is shared?

2, why to use

3. How to use

4, using the problem how to maintain

This article is from the "Wsyht blog" blog, make sure to keep this source http://wsyht2015.blog.51cto.com/9014030/1786893

Linux Operations (Application software principles) Interview topics