Linux notes-disk and file systems, compression and archiving

• Block: The system logically divides the area of the disk into a number of blocks, each of which can be 1024, 2048, 4096 (bytes) device files do not take up disk blocks, and each block has an inode number

• Super Block super blocks data for block storage:

• 1. How many block groups are there?

• 2. How many blocks are included in each block group

• 3. Size of the Block

• 4. Free disk block, used disk block, used inode, idle Inode

• The first block of any partition will be reserved by the system: used by multiple systems to load systems on the boot partition

• A partition contains: Boot block (the first block of the partition)--block Group0--block group1--block group2 ....

• Block Group contains: Super Block---GDI (Block Group description table)---data block bitmap (block bitmap)---inode bitmap (inode bitmap)---inode table (inode)-- -Data Block

• GDI Block Group Description table: Block group name, start block, end block, end block number for each block group

• Directory Mapping Table:

• Inode Rec_len Name_len file-type Name

• 0 21 12 1 2.

• 12 22 12 2 2..

• Disk reference format

• Mode

• Owner-info

• Timestamps

• (primary citation) DIROCT block-----------Data Direct Reference Pointer Limited

• (Level two reference) INDIROCT block-----------------blocks group-----------{data Data}

• (Level three reference) DOUBLC INDIROCT--------Block Group-----{block Group block Group}-------{data data Data data} (common to large files, also limited to a single maximum file) indirect reference

• TRIPLC indiroct

• File System Management:

• Re-creating the file system will damage the original file

• To view the file system types supported by the current kernel: cat/proc/filesystems

• Mkfs:make File System

• Mkfs-t fstype/dev/sda#

• Manage Ext Series Files:

• Mke2fs-j ext3/dev/sda# directly created as Ext3 file system (Mke2fs-j/dev/sda#)-B: Specifies that the block size default to 4096 bytes can be used with a 1024x768 2048 4096-l: Specify the partition label- M #: Specifies the percentage of blocks reserved to the root user (without the percent sign)-I: Used to specify how many bytes of space to create an inode, the default is 8192 the value given here should be 2 N times the block size N #: Specifies the number of inode to create-f: Forces the creation of the file system ( Partition used for mounting)-E: User specified additional file system properties

• Blkid: View disk device related properties blkid device name (UUID: Uniform global Unique identifier)

• Eslabel: Used to view or define a volume label Definition volume Label: E2label Device Name volume label name

• TUNE2FS: Used to adjust the relevant properties of the file system,-j ext3: Upgrade ext2 to ext3 does not damage the original data, but not degraded (ext2 and ext3 zone ratio: Ext3 added a log file system)-L: Set or modify the volume label-M #: Adjust the reservation percentage-R #: Specify the number of reserved blocks-O: Set default mount Option-C #: Set the number of mounts, after # times to self-test, 0 or 1 To turn off this feature-I #: Do not mount the number of days after the self-test, 0 or 1 To turn off this feature-L: Show information in the Super block

• DUMPE2FS: Display file property Information-H: Show only Super block information

• FSCK: Check and Repair file system-T fstype: Specify File System Type-A: automatic repair (do not ask, auto-complete)

• E2fsck: Specialized check and repair ext2 ext3-f: Force Check-P: Automatic repair can also-u

• Mount: Associate a new file system to the current root file system

• Uninstall: Remove the file system from the associated relationship of the current root file system

• Mount:mount no option to display devices and mount points that are currently mounted on the system, mount [options] [-o options] Device mount_point-a: Represents all file systems defined in the Mount/etc/fstab file-N: Default  The mount command saves the mounted information to the/etc/mtab, using the-N to not write the information to the file, T-fstype: Specifies the type of file system on the device being mounted When you do not use this option, Mount calls the Blkid command to get the corresponding file system type-r: File system, read-only (Mount disc is commonly used)-W: Read-write Mount-o: Refers to the fixed plug-in option

• Mount Mount device name (device file, volume label, or UUID) mount point (directory)

• Directory requirements: 1. This directory device is not being used by another process 2. Directory must exist 3. The original file in the directory is temporarily hidden (can be seen after uninstallation)

• To access files on the corresponding file system by mount point when Mount is complete

• Umount: Uninstalling a file system Umount device name (device file, volume label, or UUID) or mount point (directory) (mounted device not used by process)

• Remout: Re-mount the current file system, RO: Read-only mount RW: Read-write mount-o remount,ro/dev/sda#

• Free: View the swap space information in m for the usage of physical memory and swap space on the current system

• To create a swap partition:

• 1. Adjust the partition type: The FDISK command adjusts to the 82 partition type

• 2.mkswap/dev/sda# (Create swap partition)

• 3.swapon-a/dev/sda# Open Mount (-A: Enable all switching devices defined in the/etc/fstab file) swapoff/dev/sda# off Mount

• Loopback devices:

• Loopback, using software to simulate the implementation of hardware

• Mount command to mount an ISO image through a loopback device

• Mount-o Loop ~/centos6.ios/media/(-o loop local loopback device)

• DD if= source file of= data target bs=1024 (specify size can be in m) count=2 (Specify number)

• Can be used to back up the MBR:

• DD IF=/DEV/SDA of=/mnt/usb/bf.backup bs=512 count=1

• To restore the MBR:

• DD If=/mnt/usb/bf.backup OF=/DEV/SDA bs=512 count=1

• Make disc Image: Cat/dev/cdrom > ~/centos6.iso

• Use files to increase the size of swap space without partitioning:

• 1.dd/dev/zero of=/var/swfile/bs=1m count=1024 Create a file of 1G size

• 2.mkswap/var/swfile

• 3.swapon/var/swfile

• (DD can also be followed by seek=1023 to indicate that the space after the 1023M has been created is false and can be viewed with du)

• File system configuration file/etc/fstab: Automatically mounts every file system defined in this file at system startup

• Device to mount (UUID or device name) mount point file system type Mount option dump screen rate self-test order

• Mount option: Default is Defaulfs

• Switching screen rate: Every number of days to do a full backup 1: Every day to do Backup 2: Do a full backup every two days 0: do not back up

• Self-Test order: Only the root is 1, the other can be 2 0: indicates no check

• Fuser: Verifying that the process is using a file or socket file-V: Viewing a running process on a file-k-m: Terminating all processes that are accessing this mount point

• Compression decompression

• Compression format:. gz. bz2. xz. zip

• Compression algorithms are different, the compression ratio will be different

• Gzip bzip2 XZ can only compress files cannot be compressed directory by default source delete files

• Gzip:gzip file1 file2-d: equivalent to gunzip-#: Specify compression level (1-9), default = 6

• Decompression: gunzip file1 file2 (source files are also deleted)

• Zcat file.gz: View the contents of a file without being pressed

• BZIP2: More compression than gzip, using format with gzip approximation,-K: can keep source files also supports-D-# bunzip: Decompression bzcat: no pressure, view

• XZ file1 ... : also supports the-D-#-K unxz Xzcat

• That is, archive and compress:

• No source files are deleted from the zip unzip

• Zip-Compressed file name source file

• Archive: Archiving does not mean compression

• Tar: Archive tool, do not compress-C: Create archive file-F: Action Archive-x: Restore archive-cf file-xf ... Source files are not deleted by default--xattrs: Archive retains its extended attribute information-TF: View files in Archive XZ compression **.tar file-ZCF: Archive also compress (call gzip compression)-ZXF: Unzip, expand archive-JCF: Call bzip2 compress-JXF: Solution Compression (BZIP2)-JCF: Call XZ compression-JXF: Extract (XZ)-JTF: View--exclude filename: Do not include the filename file when packaging or compressing

• READ: Interactive-P: Prompt message-t wait time

• While loop:

• Suitable for scenes with unknown cycle times, while Cowditiow;do

• Statement

• ....

• Done

• 
  

This article is from the "stupid Bird first-FAI" blog, please be sure to keep this source http://benniaoxianhui.blog.51cto.com/10463960/1679207

Linux notes-disk and file systems, compression and archiving