Linux Interview Quiz 10

Interview , the interviewer may ask you what work you've done before. Through your answer, he can determine whether you have ever done operation. Below I give some answers, hope can help you.

1. Look at the monitoring center, to deal with the warning problem, note that the warning here is not the problem of failure, failure problems usually once received the alarm message we have been processed the first time. The problem here can not be solved immediately, but there is also a potential risk problem. For example, for example, the disk partition more than 80%, a machine load for a long time (more than 20), a MySQL slow query log number more, an Nginx server occasionally appear 502 status code, etc.

2. Write or optimize operations scripts. what script? such as backing up MySQL scripts, monitoring a service script, automating the release of the script as long as it is convenient for our operations to work the script is written.

3. Deploy or configure the environment according to the needs of your development or operations colleagues, and of course write some scripts. For example, today's development colleagues on-line a new business, then it means to add a virtual host, Nginx and PHP-FPM to configure, as well as to modify the DNS resolution and so on. For example, development colleagues say that a recent business visit is slow, then we need to cooperate with the development of colleagues to find the cause of the business is slow, you can check the MySQL slow query log can also view php-fpm slow execution log and so on. For another example, operations colleagues say that recent visits are a bit too much to see if they are normal access or unusual access, and we need to analyze the access logs. Wait a minute

4. Security and optimization. One day, the official PHP said that the version has a loophole, the impact is very large, we need to be based on the official patch to deal with the vulnerability. This requires us to pay more attention to some of the information in the IT field, focus on some of the public number of Daniel or the organization, the first time to know the loopholes. optimization, it is necessary to develop colleagues to work together to do. Many of the Apache-related configurations We've talked about have been optimized, such as log cutting, static file cache time, access control, and so on.

5. Plan and deploy the company's business and decide when to expand or shrink. Business access is increasing, the server load is increasing, once the support is not enough, you need to expand. The general enterprise architecture is to support horizontal expansion (LVS, nginx load Balancing), how many machines to increase, also need our measurement, the purchase of machines, shelves, deployment work is also we do. Now more and more enterprises use cloud host, in the machine purchase, shelves, deployment is quite simple. Environment deployment, we need to test, test no problem to really go online.

6. Focus on the first-line information and learn new skills. If not busy, we need to strictly manage their own, learning new skills is necessary, not learning will be eliminated. Open source China, 51cto, InfoQ, TechTarget and so on can be concerned about, we can understand the first-line operational trends and new technologies.

7. Some large enterprises have an asset management platform (CMDB), which also requires our OPS personnel to manage the records and changes of various server resources.

8. Some enterprises may be involved in employee account, authority Management (GIT/SVN), may also need us to allocate.

9. Fault handling. This needless to say, encounter problems, you have to deal with. Or a list of common failures: site access is slow or inaccessible, server downtime, a service outage, DDoS attacks, CC attacks, disk corruption, storage corruption, disk space is full, the site is black, MySQL master never sync, MySQL query slow and so on.

10. Analysis of PV UV data combined with business and product line out charts and reports (Web-based)

11. On-line release status statistics and problem entry

12. Basic APM Analysis Request troubleshooting trend analysis and response status

13. Automatic operation and maintenance platform development, to central control, certification centralization, work order integration

14. Troubleshooting the problem of scripting, problem-solving methods scripted, at least to achieve semi-automatic;

15. Responsible for the System Information batch scanning, batch chemical development and maintenance;

16. Continuous improvement and optimization of monitoring, including monitoring projects, monitoring scripts, if necessary, to monitor the development of two times;

17. Continuous optimization of the script, the goal is to use as little code to achieve maximum results, merging and eliminating unnecessary judgments and loops ;

Linux Interview Quiz 10