Shulou(Shulou.com)05/31 Report--

This article shows you what are the knowledge points of IBM minicomputer maintenance, which are concise and easy to understand, which will definitely brighten your eyes. I hope you can get something through the detailed introduction of this article.

IBM minicomputer maintenance

A brief introduction to AIX operating system

The AIX operating system is a Unix operating system developed by IBM, with many enhancements in technology and functions:

Modular kernel

Dynamically adjust the kernel

Multi-thread kernel

Efficient input / output

Real-time processing

Safety performance

Log file system

Storage management (including mirroring technology and fragmentation technology)

System management

Online help query.

II. Equipment management

2.1pieces of equipment (Block Device):

A buffered device with structured random access. Block devices are usually disk file systems (Disk File System).

2.2 character device (Character Device):

Bare equipment (Raw Device) is an unbuffered continuous, directional flow device. Character devices are usually peripherals of the system, such as terminals, printers and so on.

2.3 main equipment number (Major Number):

Identifies the device driver number in the system kernel.

2.4device number (Minor Number):

Identify a special device type.

The ODM database consists of a predefined database (Predefined Database) and a custom database (Customized Database). Used to store information and its properties for all logical devices in the system. Where:

Predefined database: contains configuration data for all devices (Supported Devices) supported by the system based on the system configuration.

Customize the database:

Contains all defined (Defined Devices) and available (Available Devices) device configuration data in the current system.

Class of device (Class): identifies the device

Subclass of the device (Subclass): identifies how the device connects to the system

Type of device (Type): identifies the type of device

Name of the device (Name): identifies the logical device name

# lsdev device list

# lsattr device attribute list

# chdev change device properties

# lscfg displays the configuration information of devices in the current system

2.5 configuration of the device:

# mkdev-c Class-s Subclass-t Type-p Parent_Adapter-w Connection_Location

-an Attribute=value

-c Class: specifies the class to which the device belongs (e.g. tape)

-s Subclass: specifies the subclass to which the device belongs (e.g. scsi)

-t Type: specifies the type to which the device belongs (e.g. 150mb)

-p Parent_Adapter: specify the parent card number of the device (e.g. scsi0)

-w Connection_Location: specify the location or address where the device is connected to the parent card (for example: 5pc0)

-an Attribute=value: specifies the properties of the device (e.g. ret=no)

2.6 status of the device:

Undefined: there is no record of the existence of this device in the system (device information is stored in a predefined database) (Supported)

Defined: the system has assigned a logical device name and port to this device, but the system currently cannot access this device (device information is stored in a custom database) (Defined)

Available: the device is defined in the system and can be used (device information is stored in a custom database) (Available)

# Adaptive device (Self-Configuring Devices) in cfgmgr automatic configuration system

III. Storage management

Restrictions on logical Storage in 3.1AIX

Volume group (VG) 255s / per system

Physical volumes (PV) 32 / per volume group

Physical partitions (PP) 1016 per physical volume

Logical volumes (LV) 256x / per volume group

Logical partitions (LP) 32512 per logical volume

AIX system logical volumes:

Paging Space:/dev/hd6, a fixed disk space for storing information in virtual memory

Journal Log:/dev/hd8, used to record changes in the file system structure in the system

Boot LV:/dev/hd5, the physically contiguous disk space for the system boot image

AIX system file system:

/ (root): important system equipment information and application storage space needed by the system to start the process

/ usr:/dev/hd2, storage space for system commands, information bases, and applications

/ var:/dev/hd9var, storage space for system log files and print data files

/ home:/dev/hd1, the storage space for user data in the system

/ tmp:/dev/hd3, temporary files of the system and storage space for users' work

3.2LVM operation command:

# lsvg View the information of the LV belonging to the active VG in the system

# lslv when viewing the information belonging to the specified LV, the system will give the number of LP and the number of copies of the specified LV.

# lspv View the information of PV in the system

# mkvg create a new VG

# chvg changes the startup feature of VG

# extendvg extends the size of VG

# reducevg reduces the size of VG

# reorgvg reschedule the allocation policy of LV in VG

# varyonvg activate a VG

# varyoffvg shuts down a VG

# importvg connects to a VG

# exportvg outputs a VG

Note: the correct steps to move a VG from system A to system B are:

System aura # umount all

# varyoffvg VG_Name

# exportvg VG_Name

System BRV # impoertvg

# varyonvg VG_Name

# rmlv Delete a LV

# chlv change the name of LV

# migratepv moves data from one PV to another PV

Note:

1. When migrating data on PV, the size of the destination PV (Target_PV) is larger than that of the source PV (Souce_PV), and the destination PV and the source PV must be in the same VG.

1. When increasing the number of copies of a LV (increasing the mirror of LV), you must use the # syncvg command to make LV and its copies

Shell synchronization (synchronous mirror).

Management of 3.3FS:

The structure of JFS:

Super block (Superblock):

The superblock contains the size of the FS, identity, list of free space, size of disk fragments (Fragment), and nbpi (which determines the number of inodes in the FS).

I node (inodes):

The I node contains the file size, file owner, file permissions, file creation and access

Time and a pointer to the data block.

Data blocks (indirect blocks):

IV. Process management

AIX is a multi-user and multi-task operating system that manages multiple user requests and tasks. Most systems have only one CPU and one main memory, but a system may have multiple secondary storage disks and multiple input / output devices. The operating system manages these resources and shares them among multiple users, giving you the illusion that the system is occupied only by you when you make a request. In fact, the operating system monitors a queue of tasks waiting to be executed, including user jobs, operating system tasks, mail, and print jobs. The operating system assigns appropriate time slices to each task according to the priority of each task, and each time slice has about a few tenths of a second. Although it looks very short, it is actually enough for the computer to complete thousands of instruction sets. Each task is run by the system for a period of time, then suspended, and the system turns to other tasks; come back to deal with this task later until a task is completed and removed from the task queue.

4.1 processes and Operations

Everything that runs on an AIX system can be called a process. Each user task, each system management daemon, can be called a process. Aix uses time-sharing management to make all tasks share system resources. What we care about is how to control these processes so that they can serve users well. A more formal definition of a process is that ∶ runs a separate program in its own virtual address space.

There is a difference between a process and a program. A process is not a program, although it is generated by a program. A program is only a static set of instructions and does not occupy the running resources of the system, while a process is a dynamic program that can change at any time and use the running resources of the system. And a program can start multiple processes.

There is also a difference between the concept of process and job. An executing process is called a job, and a job can contain one or more processes, especially when pipes and redirect commands are used.

Job control refers to the behavior of controlling a running process. For example, a user can suspend a process and continue to execute it later. Shell records all started processes, and during each process, the user can suspend or restart the process at will. Job control is a feature of many shell, including bash and tcsh, that enables users to switch between multiple independent jobs.

4.2 start the process

Type the program name of the program you need to run, execute a program, that is, start a process. In the aix system, each process has a process number, which is used to identify and schedule the process. There are two main ways to start a process: ∶ manual startup and scheduling startup, the latter is set up in advance and start itself according to the user's requirements.

1. Start it manually

By the user input command, directly start a process is to start the process manually. However, the manual startup process can be divided into foreground startup and background startup. Foreground startup is the most common way to start a process manually. In general, the user types a command "ls-l", which already starts a process and is a foreground process. At this time, the system is already in a multi-process state.

2. Scheduling startup

Sometimes it is necessary to do some time-consuming and resource-consuming maintenance work on the system, which is suitable to be carried out late at night, when the user can schedule in advance and specify the time or occasion for the task to run. then the system will automatically complete all the work. To use the ability to start a process automatically, you need to master the following startup commands.

The first important command is the at command user uses the at command to execute a specified sequence of commands at a specified time. In other words, the command needs to specify at least one command and one execution time to run properly. The at command can specify only the time, or it can be specified together with the date. It should be noted that there is a systematic discrimination problem at a specified time. For example, when the ∶ user now specifies an execution time, ∶ 3:20, and when the at command is issued at 20:00 the night before, on which day will the command be executed? If the user is still working before 3:20, the command will be completed by that time; if the user exits work before 3:20, the command will not be executed until the wee hours of the next morning.

3. Cron command

Both of the commands described above will complete certain tasks within a certain amount of time, but be aware that they can only be executed once. That is, when the run command is specified, the system completes the task at the specified time, and it's all over. But in many cases, you need to repeat some commands over and over again, such as ∶, a company automatically reports to employees about its activities in the first week every Monday, so you need to use the cron command to complete the task. In fact, the cron command should not be started manually. The cron command is automatically started by a shell script when the system starts and goes into the background (so you don't need to use the & symbol). The average user does not have permission to run this command, and although the superuser can start cron manually, it is recommended that it be put into the shell script and started by the system itself.

First, the cron command searches the / var/spool/cron directory for crontab files named after the user name in the / etc/passwd file, and the files found will be loaded into memory. For example, for a user whose user name is foxy, the corresponding crontab file should be / var/spool/cron/foxy. That is, the crontab file named after the user is stored in the / var/spool/cron directory. The cron command also searches for the / etc/crontab file, which is written in a different format. After cron starts, it will first check whether any user has set up the crontab file, and if not, it will go into the "dormant" state and release system resources. So the background process takes up very few resources. It wakes up every minute to see if there are currently any commands that need to be run. After the command is executed, any output is sent as a message to the owner of the crontab, or to the user specified in the MAILTO environment variable in the / etc/crontab file. The above briefly describes how some cron works, but the execution of the cron command does not require user intervention; what the user needs to modify is the sequence of commands to be executed in the crontab, so the crontab command is described below.

4. Crontab command

The crontab command is used to install, delete, or list the tables used to drive cron background processes. That is, the user puts the sequence of commands that need to be executed into the crontab file for execution. Each user can have their own crontab file. Let's see how to create a crontab file. Crontab files under / var/spool/cron cannot be created or modified directly. The crontab file is obtained through the crontab command. Now suppose you have a user name of foxy and you need to create your own crontab file. You can first create a new file using any text editor, and then write to it the commands you need to run and the time you want to execute periodically. Then save and exit. Assume that the file is / tmp/test.cron. Then use the crontab command to install the file and make it the user's crontab file.

How to enter the command and time you need to execute in the crontab file. Each line in the file contains six fields, the first five of which are the specified time when the command is executed, and the last is the command to be executed. Each field is separated by spaces or tabs.

Job control allows the process to be suspended and can be resumed when needed, and the suspended job will continue to run from the point where it is resumed. Just press ctrl+z on the keyboard to suspend the current foreground job. When you press ctrl+z on the keyboard, the currently executed command cat is suspended. Use the jobs command to display a list of jobs for shell, including the specific job, the job number, and the current status of the job. When the recovery process executes, there are two options: ∶ uses the fg command to put the pending job back to the foreground for execution, and the bg command to put the suspended job in the background. Flexible use of the above commands will bring great convenience to yourself.

5. Process view

Who command

This command is mainly used to view the status of users currently online. This command is very useful. If a user wants to establish instant messaging with other users, such as using the talk command, the first thing to make sure is that the user is actually online, otherwise the talk process cannot be set up. As another example, the system administrator also uses the who command to monitor what each logged-in user is doing at this moment.

6. Top command

The basic function of the top command and the ps command is the same, showing the current progress and other conditions of the system, but top is a dynamic display process, that is, you can constantly refresh the current state by pressing the user key. If the command is executed at the foreground, it will monopolize the foreground until the user terminates the program. More accurately, the top command provides real-time monitoring of the state of the system processor. It will display a list of CPU's most "sensitive" tasks in the system. This command can be used by CPU. Memory usage and execution time sort tasks; and many of the features of this command can be set through interactive commands or in personal customization files.

4.3 process scheduling

When a foreground process needs to be interrupted, Ctrl+c key combination is usually used; but for a background process, it may not be solved by a single key combination, so you must resort to the kill command. This command terminates the background process. There are many reasons for terminating the background process, perhaps because the process takes up too much CPU time, or because the process is dead. In short, this happens all the time.

Fifth, system backup / recovery and availability

Tartar this command is used to archive files.

Cpio this command copies files in and out of memory and directories.

Dd this command reads the input file or standard input to make the specified changes, and then copies the converted data to the output file or standard output. You can specify the block size of the Iripple O to better match the physical Ipicuro.

Pax can archive more than 2G system dump and mirrors.

Makesysb creates a rootVG installation mirror to a file or a bootable tape.

Savevg this command saves non-rootVG.

Backup creates file backups on backup media in two ways: specify file backups and entire system backups (as I nodes).

Restore this command restores archive files created with the backup command.

Tctl this command controls the tape device.

Backup and recovery, including rootvg backup and user data backup:

5.1 backup of operating systems and system programs:

Insert a new or useless tape into the tape drive

# tctl-f / dev/rmt0 rewind

# smit mksysb

Enter "/ dev/rmt0" in "backup device or file" and the system will run for a long time, wait until the screen shows OK and take out the tape. At this point, the system backup is complete. Mksysb backs up only mounted file systems in rootvg.

5.2 user data backup

Common Tape Drive option

/ dev/rmt0: if / dev/rmt0 is selected, the tape drive will rewind the tape to the head when inserting the tape and writing the tape once, so the next backup file will overwrite this backup.

/ dev/rmt0.1: if / dev/rmt0.1 is selected, the tape drive does not rewind the tape when the tape is inserted and written once, so a tape can back up several files or file systems in succession.

# smit fs

Select backup File system

Add the File system name to be backed up

Add "/ dev/rmt0.1"

Repeat the above to back up multiple file systems on the same tape.

5.3 recovery of rootvg

Start the machine into maintenance mode

See the installation manual, when "Welcome to Base Operating System Installation and Maintanence" appears

Select 3 "Start Maintenance Mode for System Recovery"

Restore the system

Continue to choose 4 "Install from a System Backup"

When the "Choose mksysb Device" screen appears, select "/ dev/rmt0" and insert the tape and enter. At this time, the system automatically restores the operating system.

5.4 user data recovery

# tctl-f / dev/rmt0 rewind

# smit fs

Select restore File system

Add "device name" and "destination directory"

The system will automatically find the appropriate directory for recovery.

VI. Daily maintenance

6.1 whether the file system is full

Df-k can check file system usage in units of K. (over 90%, need to be adjusted)

6.2 check the system error log using errpt | more to check

Clear the existing log: Errclear 0

6.3 check the legal / illegal login of the system

Use the Last command to check where you came from.

6.4 check whether the system has huge Core file generation

Use find /-name core-print to check. For Core files, you can usually delete them directly.

6.5 system performance check:

A) CPU performance: use Vmstat, topas to check

B) memory usage: also use topas and vmstat to check

C) check IO balance usage: use iostat to check

D) swap space usage: use lsps-a to check

What are the knowledge points of IBM minicomputer maintenance? have you learned any knowledge or skills? If you want to learn more skills or enrich your knowledge reserve, you are welcome to follow the industry information channel.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.