System problem solving after thinking, dealing with the problem of thinking (Oracle can't shut down, not to be)

Objective

Application support A, to the business unit to reflect the company's Mister Production Kanban data display, contact Development Group B, said the program has not changed recently, the problem must be in the database let me check the database

Check the error ora-12541, check the database monitoring, 1521 ports, are normal, the production system on the line (and production Kanban is the same database), Plsql are normal. Reaction to the development group B,b said that the program did not change, the problem must be in the database.

Since the data schema of the production Kanban is not understood, the configuration parameters to B to apply the connection database have not been given to me

Because it is the general manager of the company, the leader is concerned about, application support A has been said to be a database problem, let restart the database first

I have always insisted on the first understanding of the architecture, but leadership, a and a few colleagues are talking about the database problem, think about the next factory, here are 15 minutes to restart the database for me.

Sql>shutdown immediate;

Wait 2 minutes or here, it's probably a database hang up.

Open a different window:

Sql> Conn/as SYSDBA

Connected to an idle instance.

Sql> startup;

Ora-10997:another Startup/shutdown operation of this instance inprogress

Ora-09968:unable to lock file

Linux-x86_64 Error:11:resource temporarily unavailable

Additional information:23096

Well, you can't shut it off.

The leader at this time asked how the matter, I said hang live, call to ask the factory still have a few minutes of the involute, answer 3 minutes

Kuanghan ...

All right, just kill the local=no. The process of forced shutdown (risking the risk of data loss, the factory stop the line of responsibility is too big, magnanimous)

Sql>shutdown abort;

sql>startup;

Okay, database's up.

Check the database status:
Sql>select Open_mode from V$database

Read_write

Normal Open

Landing MES system

System normal psychological big stone put down

The issue of production Kanban has not been resolved, which means that the database restart is invalid

Asked B,b that the problem was found.

Production Kanban data is summarized to me here the database shows, I here the database is from the group Server DB1 Summary, the production data of each base is summarized to the group server DB1 first.

Group Server DB1 Downtime causes this problem,-_-| | |

Summarize:

A: When encountering a system problem, check whether the system has an abnormal error, if not, check whether other aspects of the impact;

Second: In the absence of understanding of the system architecture, do not arbitrarily do some of the greater risk of operation, after all, to solve a problem requires more communication, more understanding, closed doors certainly not;

Three: When the need to restart the server, to distinguish the size of things, priorities, not affected by the surrounding people, after all, restart the server when there is a sudden state of affairs, there is a major failure, you are the first responsible person, and rushes people do not have much relationship.

This article is from "one van" blog, please be sure to keep this source http://1336014.blog.51cto.com/1326014/1661523

System problem solving after thinking, dealing with the problem of thinking (Oracle can't shut down, not to be)