Vous êtes ici : Formations en Ligne Gratuites » Catalogue » workbooks » Red Hat » RH124 - Course Overview » RH12401 - File Hierarchy System

Différences

Ci-dessous, les différences entre deux révisions de la page.

Lien vers cette vue comparative

Les deux révisions précédentesRévision précédente
Prochaine révision		Révision précédente
elearning:workbooks:redhat:rh124en:l100 [2024/10/10 15:00] – adminelearning:workbooks:redhat:rh124en:l100 [2024/11/22 17:46] (Version actuelle) – admin
Ligne 5: Ligne 5:
 Last update : ~~LASTMOD~~ Last update : ~~LASTMOD~~
  
-======RH12401 - File System======+======RH12401 - File Hierarchy System=====
  
-=====Module content=====+=====Contents=====
  
-  * **RH12401 - File System** +  * **RH12401 - File Hierarchy System** 
-    * Module content+    * Contents 
 +    * Set Up your Keyboard
     * LAB #1 - Linux File Hierarchy System     * LAB #1 - Linux File Hierarchy System
-    * 1.1 - File Types +      * 1.1 - File Types 
-    * 1.2 - The mount command +      * 1.2 - The mount command 
-    * 1.3 - The umount command +      * 1.3 - The umount command 
-    * 1.4 - The /etc/fstab file +      * 1.4 - The /etc/fstab file 
-  * Mount Options+        * Mount Options
     * LAB #2 - Unix File System     * LAB #2 - Unix File System
-    * 2.1 - Superblock +      * 2.1 - Superblock 
-    * 2.2 - Inodes +      * 2.2 - Inodes 
-    * 2.3 - Data blocks +      * 2.3 - Data blocks 
-    * 2.4 - Physical links +      * 2.4 - Physical links 
-    * 2.5 - Symbolic links+      * 2.5 - Symbolic links 
 + 
 +=====Set Up your Keyboard===== 
 + 
 +The current keyboard layout of the Red Hat 9.4 VM is **fr** : 
 + 
 +<code> 
 +[root@redhat9 ~]# localectl status 
 +System Locale: LANG=en_US.UTF-8 
 +    VC Keymap: fr 
 +   X11 Layout: fr,us 
 +  X11 Variant: , 
 +</code> 
 + 
 +Find the appropriate keymap for your host PC: 
 + 
 +<code> 
 +[root@redhat9 ~]# localectl list-keymaps 
 +ANSI-dvorak 
 +adnw 
 +al 
 +al-plisi 
 +amiga-de 
 +amiga-us 
 +apple-a1048-sv 
 +apple-a1243-sv 
 +apple-a1243-sv-fn-reverse 
 +apple-internal-0x0253-sv 
 +apple-internal-0x0253-sv-fn-reverse 
 +applkey 
 +at 
 +at-mac 
 +at-nodeadkeys 
 +atari-de 
 +atari-se 
 +atari-uk-falcon 
 +atari-us 
 +az 
 +azerty 
 +ba 
 +ba-alternatequotes 
 +ba-unicode 
 +ba-unicodeus 
 +ba-us 
 +backspace 
 +bashkir 
 +be 
 +be-iso-alternate 
 +be-latin1 
 +be-nodeadkeys 
 +be-oss 
 +be-oss_latin9 
 +be-wang 
 +bg-cp1251 
 +bg-cp855 
 +bg_bds-cp1251 
 +bg_bds-utf8 
 +bg_pho-cp1251 
 +bg_pho-utf8 
 +bone 
 +br 
 +br-abnt 
 +br-abnt2 
 +br-dvorak 
 +br-latin1-abnt2 
 +br-latin1-us 
 +br-nativo 
 +br-nativo-epo 
 +br-nativo-us 
 +br-nodeadkeys 
 +br-thinkpad 
 +by 
 +by-cp1251 
 +by-latin 
 +bywin-cp1251 
 +ca 
 +lines 1-58 
 +</code> 
 + 
 +Before you start, configure your keyboard accordingly. For example in the case of a **US** layout: 
 + 
 +<code> 
 +[root@redhat9 ~]# localectl set-keymap us 
 + 
 +[root@redhat9 ~]# localectl status 
 +System Locale: LANG=en_US.UTF-8 
 +    VC Keymap: us 
 +   X11 Layout: us 
 +    X11 Model: pc105+inet 
 +  X11 Options: terminate:ctrl_alt_bksp 
 +</code>
  
 =====LAB #1 - Linux File Hierarchy System===== =====LAB #1 - Linux File Hierarchy System=====
Ligne 54: Ligne 145:
 </code> </code>
  
-* **/afs**: Andrew File System (AFS) is a distributed file system that uses a set of trusted servers to present a consistent, location-transparent file namespace. +| **/afs** | Andrew File System (AFS) is a distributed file system that uses a set of trusted servers to present a consistent, location-transparent file namespace. | 
-**/bin**: is short for //**bin**ary// or binaries. It contains programs such as ls. Under RHEL 9 it is a symbolic link pointing to /usr/bin.  +| **/bin** | Contains user programs such as ls, cp e.t.c.. Note that under RHEL 9, this is a soft link (shortcut) to **/usr/bin**. | 
-**/boot**: contains the files needed to boot the system.  +| **/boot** | Contains bootloader files, kernels and initrd (INItial Ram Disk) files. | 
-**/dev**: contains the nodes used to access all types of hardware, such as /dev/fd0 for the floppy drive. The //udev// binary is responsible for dynamically creating and deleting nodes. +| **/dev** | Contains device nodes for accessing all the peripherals and devices connected to the system. The //udev// binary takes care of dynamically creating and deleting the relevant device nodes automatically. | 
-**/etc**: contains configuration files such as passwd for passwords and fstab, which is the list of file systems to be mounted when the system boots.  +| **/etc** | Contains static configuration files. | 
-**/home**: contains the directories for each user except root.  +| **/home** | Contains a directory for each registered user of the system except for root. | 
-**/lib**: contains the common 32-bit libraries used by programs and modules. Under RHEL 9 this is a symbolic link pointing to /usr/lib.  +| **/lib** | Contains shared 32 bit libraries for applications and modules. Note that under RHEL 9, this is a soft link (shortcut) to **/usr/lib**. | 
-**/lib64**: contains the common 64-bit libraries used by programs and modules. Under RHEL 9 this is a symbolic link pointing to /usr/lib64.  +| **/lib64** | Contains shared 64 bit libraries for applications and modules. Note that under RHEL 9, this is a soft link (shortcut) to **/usr/lib64**. | 
-**/media**: contains directories for each mounted file system (accessible to the Linux system) such as floppy, cdrom etc.  +| **/media** | Contains a folder for each of the mounted external file systems (CDRom DVDRom, USB Key e.t.c.). | 
-**/mnt**: contains directories for each file system temporarily mounted by root. +| **/mnt** | Contains a folder for each external file system mounted temporarily by root. | 
-**/opt**: contains optional applications. +| **/opt** | Contains optional application packages. | 
-**/proc**: contains a virtual file system which extracts the information being processed from memory. The contents of the files are created dynamically during consultation. Only root can view all the information in the /proc directory. +| **/proc** | Contains a virtual file system that documents kernel and process status information as text files. | 
-**/root**: the home of root, the system administrator. +| **/root** | The home directory of the root user. | 
-**/run**: replaces the /var/run directory. +| **/run** | Replaces the /var/run directory. Note that under RHEL 9, /var/run is a soft link (shorcut) to **/run**. | 
-**/sbin**: contains binaries, i.e. programmes, for administering the local system. Under RHEL 9 this is a symbolic link pointing to /usr/sbin. +| **/sbin** | Contains essential system administration binaries. Note that under 9, this is a soft link (shortcut) to **/usr/sbin**. | 
-**/srv**: contains data for the **s**e**r****v**ices hosted by the system, such as ftp, databases, web, etc. +| **/srv** | Contains site specific data **s**e**r****v**ed by the system (www,ftp,databases e.t.c.). |  
-**/sys**: contains a virtual file system whose role is to describe the hardware for udev. +| **/sys** | Contains a virtual file system that describes devices for //udev//. | 
-**/tmp**: stores temporary files created by programs.  +| **/tmp** | Contains the temporary files created by the system and by applications. | 
-**/usr**: contains user commands in /usr/bin, HOWTOs in /usr/share/doc, manuals in /usr/share/man and other major entries. +| **/usr** | Contains user commands in ///usr/bin//, HOWTOs in ///usr/share/doc//, manuals in ///usr/share/man// and is the //Secondary Hierarchy// for read-only user data. | 
-**/var**: contains files of variable size.+| **/var** | Contains variable files. i.e. files that continually change such as log files and spool files. |
  
-====.1 - File Types====+Graphically, this can be represented as follows ("Standard-unix-filesystem-hierarchy.svg" by Ppgardne, Wikimedia Commons is licensed under CC BY-SA 4.0):
  
-There are three main types of file on the Linux system:+{{ :elearning:workbooks:redhat:rh124en:standard-unix-filesystem-hierarchy.png |}}
  
-* ordinary files+====1.1 - File Types====
  
-* directories+The three major file types under Linux are :
  
-* special files (or Devices).+  * Ordinary files, 
 +  * Directories, 
 +  * Special files or Devices.
  
-Ordinary files are text files, tables or executables.+Note that :
  
-The length of the file name is limited to 255 characters.+  * Ordinary files can be anything from text files to binaries. 
 +  * The length of a file name is limited to 225 characters, including the file extension. 
 +  * Linux is case sensitive. 
 +  * If a file name starts with a dot (**.**), it is a hidden file.
  
-There is a distinction between upper and lower case. +====1.2 - The mount command====
  
-If a file name begins with **.**, the file becomes hidden.+In order to be able to use external file systems, such as a CDRom or DVDRom, Linux needs to be informed of their availability. This is accomplished by using the **mount** command:
  
-====.2 - The mount==== command+  # mount /dev/<special_file> /mnt/<directory_name> [Enter]
  
-In order for Linux to be informed of the presence of a file system, this system must be mounted. To mount a file system, use the **mount** command:+where **/dev/<special_file>** is the file system to mount and **/mnt/<directory_name>** is the target directory where the mounted file system will be available to the system. The directory **/mnt/<directory_name>** must exist prior to using the **mount** command.
  
-# mount /dev/<special_file> /mnt/<target_directory> +In the case where the **mount** command is used without options, the current mounted file systems are shown:
- +
-where **/dev/<special_file>** is the device to be mounted and **/mnt/<target_directory>** is the directory that will be used as a ‘window’ to view the contents of the file system. This directory must exist before attempting to mount the file system. +
- +
-If the **mount** command is used without options, the system returns a list of all the file systems currently mounted:+
  
 <code> <code>
Ligne 169: Ligne 261:
 </WRAP> </WRAP>
  
-====.3 - The umount===== command+====1.3 - The umount command====
  
 To unmount a file system, use the umount command: To unmount a file system, use the umount command:
  
-# umount /mnt/<target_directory>+  # umount /mnt/<target_directory>
  
 or or
  
-# umount /dev/cdrom+  # umount /dev/cdrom
  
-====.4 - The /etc/fstab==== file+====1.4 - The /etc/fstab file====
  
-If the **mount** command is used with the **-a** option, all the file systems listed in a special file called **/etc/fstab** will be mounted at the same time:+If the **mount** command is used with the **-a** option, all the file systems listed in the file called **/etc/fstab** will be mounted at the same time:
  
 <code> <code>
Ligne 204: Ligne 296:
  
 ^ Field 1 ^ Field 2 ^ Field 3 ^ Field 4 ^ Field 5 ^ Field 6 ^ ^ Field 1 ^ Field 2 ^ Field 3 ^ Field 4 ^ Field 5 ^ Field 6 ^
-| Special block file or UUID or virtual file system or a label | Mount point | File system type | Comma separated options | Used by //dump// ( 1 = to dumper, 0 or empty = to ignore ) | The order in which //fsck// checks file systems at boot time | The **UUID** ( 1 = to dumper, 0 or empty = to ignore ) | The order in which //fsck// checks file systems at boot time | The **UUUID** ( 1 = to dumper, 0 or empty = to ignore ) | The order in which //fsck// checks file systems at boot time+| Special block file or UUID or virtual file system or a label | Mount point | File system type | Comma separated options | Used by //dump// ( 1 = to dump, 0 or empty = to ignore ) | The order in which //fsck// checks file systems at boot time |
  
 The **UUID** ( //Universally Unique Identifier// ) is a 128-bit string. UUIDs are created automatically and at random when the filesystem is created on the partition. They can be modified by the administrator. The **UUID** ( //Universally Unique Identifier// ) is a 128-bit string. UUIDs are created automatically and at random when the filesystem is created on the partition. They can be modified by the administrator.
Ligne 212: Ligne 304:
 The most important mount options are : The most important mount options are :
  
-^ Option ^ Filesystems ^ Description ^ Default | +^ Option ^ Filesystem ^ Description ^ Default Value | 
-| defaults | All | Equal to rw, suid, dev, exec, auto, nouser, async | N/A +| defaults | All | Use default options: rw, suid, dev, exec, auto, nouser, and async. | N/A ((Not Applicable)) | 
-| auto/noauto | All | Automatic mount/no automatic mount when using **mount -a** | auto | +| auto/noauto | All | Do or do not mount when "mount -a" is given. | auto | 
-| rw/ro | All | Mount read-write/read-only | rw | +| rw/ro | All | Mount the filesystem read-write/read-only. | rw | 
-| suid/nosuid | All | SUID and SGID bits are/are not taken into account | suid | +| suid/nosuid | All | Allow/disallow set-user-identifier or set-group-identifier bits to take effect. | suid | 
-| dev/nodev | All | Interprets/does not interpret special device files | dev | +| dev/nodev | All | Interpret/do not interpret character or block special devices on the filesystem. | dev | 
-| exec/noexec | All | Allow:do not allow programs to run | exec | +| exec/noexec | All | Permit/do not permit execution of binaries. | exec | 
-| sync/async | All | Synchronous/asynchronous mount | async | +| sync/async | All | All I/O to the filesystem should be done synchronously/asynchronously. | async | 
-| user/nouser | All | Allow/disallow a user to mount/dismount the file system. The mount point is the one specified in /etc/fstab. Only the user who mounted the file system can unmount it. +| user/nouser | All | Allow/disallow a user to mount. The mount point is read from the /etc/fstab file. Only the user that mounted the filesystem can unmount it. | N/A | 
-| users | All | Authorise all users to mount/unmount the file system | N/A | +| users | All | Allow  every  user  to  mount and unmount the filesystem.  | N/A | 
-| owner | All | Allows the owner of the device to mount it | N/A | +| owner | All | Allow device owner to mount. | N/A | 
-| atime/noatime | POSIX standard | Write/unwrite access date | atime | +| atime/noatime | POSIX | Do not use noatime feature, then the inode access time is controlled by kernel defaults/Do not update inode access times on this filesystem | atime | 
-| uid=value | Non-Linux formats | Specifies file owner number for non-Linux file systems | root | +| uid=value | Non-Linux filesystems | Set the owner of the root of the filesystem. | root | 
-| gid=value | Non-Linux Formats | Specifies the owner group number | N/A | +| gid=value | Non-Linux filesystems | Set the group of the root of the filesystem. | N/A | 
-| umask=value | Non-Linux Formats | Specifies permissions (access/read/write rights) | N/A | +| umask=value | Non-Linux filesystems | Set the umask. The default is the umask of the current process. The value is given in octal. | N/A | 
-| dmask=value | Non-Linux Formats | Specifies folder usage rights (Obsolete, prefer dir_mode) | umask current | +| dmask=value (Obsolete) | Non-Linux filesystems | Set the umask applied to directories only. The value is given in octal. | Current processes' umask | 
-| dir_mode=value | Non-Linux Formats | Specifies folder usage rights | umask current | +| dir_mode=value (Replaces dmask)| Non-Linux filesystems | Set the umask applied to directories only. The value is given in octal. | Current processes' umask | 
-| fmask=value | Non-Linux Formats | Specifies file usage rights (Obsolete, prefer file_mode) | umask current | +| fmask=value (Obsolete) | Non-Linux filesystems | Set the umask applied to regular files only. The value is given in octal. | Current processes' umask | 
-| file_mode=value | Non-Linux Formats | Specifies file usage rights | umask current |+| file_mode=value (Replaces fmask)| Non-Linux filesystems | Set the umask applied to regular files only. The value is given in octal. | Current processes' umask |
  
 =====LAB #2 - Unix File System===== =====LAB #2 - Unix File System=====
  
-Each partition on a Unix system can host one of the following structures:+Each file system contains the following :
  
-* superbloc +  * Superblock 
-* inode +  * Inodes 
-* data block +  * Data blocks 
-* indirection blocks+  * Indirect blocks
  
 ====2.1 - Superblock==== ====2.1 - Superblock====
Ligne 245: Ligne 337:
 The superblock contains: The superblock contains:
  
-* the size of the blocks +  * the size of the blocks 
-* the size of the file system +  * the size of the file system 
-* the number of mounts performed for this file system +  * the number of mounts performed for this file system 
-* a pointer to the root of the file system +  * a pointer to the root of the file system 
-* pointers to the list of free inodes +  * pointers to the list of free inodes 
-* pointers to the list of free data blocks+  * pointers to the list of free data blocks
  
 ====2.2 - Inodes==== ====2.2 - Inodes====
  
-Each file is represented by an **inode**. The inode contains :+Each file is represented by an **inode**. An inode contains the following information:
  
-* the type of file, i.e. **-**, **d**, **l**, **b**, **c**, **p**, **s** +  * the file type : **-**, **d**, **l**, **b**, **c**, **p**, **s**, 
-* access rights, for example **rwx rw- r--** +  * file permissions, for example : **rwx  rw-  r--**, 
-* the number of physical links, i.e. the number of names +  * the number of hard links, 
-* the creator's UID or the UID assigned by the **chown** command if there has been a modification +  * the UID of the file creator or the current UID attributed by the **chown** command, 
-* the GID of the creator process or the GID assigned by the **chgrp** command +  * the GID of the creating process or the current GID attributed by the **chgrp** command, 
-* the size of the file in bytes +  * the file size in bytes, 
-* the date on which the inode was last modified, i.e. **ctime**. +  * the date of the last modification of the file's inode content : **ctime**, 
-the last modification date of the file, i.e. **mtime** * the last access date, i.e. +  * the date of the last modification of the file contents : **mtime**, 
-* the date of the last access, i.e. **atime**. +  * the date of the last access : **atime**, 
-* the addresses which point to the data blocks in the file. +  * allocation addresses that point to the data blocks used by the file.
- +
-Graphically, this organisation can be represented as follows: +
- +
-{{ :free:inodes.png?400 |}}+
  
 To understand this better, type the following command: To understand this better, type the following command:
Ligne 283: Ligne 371:
 The first character of each line can be one of the following: The first character of each line can be one of the following:
  
-**-** - a file +  * - - an ordinary file, 
-**d** - a directory +  * d - a directory, 
-* l** - a symbolic link +  * l - a symbolic link, 
-**b** - a block device +  * b - a bloc type peripheral, 
-* c** - a character device +  * c - a character type peripheral, 
-* p** - a named pipe for inter-process communication +  * p - a named pipe for communication between processes, 
-* s** - a socket in a network context+  * s - a network socket.
  
 To view the inode number, use the **-i** option: To view the inode number, use the **-i** option:
Ligne 303: Ligne 391:
 ====2.3 - Data blocks==== ====2.3 - Data blocks====
  
-Data is stored in data blocks. In the case of a directory, the data block contains a table which references the inodes and file names in the directory. This table is called a **catalogue table**.+Data is stored in data blocks. In the case of a directory, the data block contains a table which references the inodes and file names in the directory. This table is called a **catalog table**.
  
-The name of a file is not stored in the inode but in a **catalogue table**. This feature allows us to give two different names to the same file. To add a new name to a file, a **physical link** must be created.+The name of a file is not stored in the inode but in a **catalog table**. This feature allows us to give two different names to the same file. To add a new name to a file, a **physical link** must be created.
  
-====2.4 - Physical links====+====2.4 - Physical (hard) links====
  
-A physical link is created using the following command:+A physical or hard link is created using the following command:
  
-* ln file_name additional_name+  * ln file_name additional_name
  
 To illustrate this point, type the following command line: To illustrate this point, type the following command line:
Ligne 325: Ligne 413:
 Note the inode number of the file **file1**. Also note that the number in the third field of the file1 line has the value **1** : Note the inode number of the file **file1**. Also note that the number in the third field of the file1 line has the value **1** :
  
-33625154 -rw-r--r--. 1 root root 0 Sep 25 13:08 file1+  33625154 -rw-r--r--. 1 root root 0 Sep 25 13:08 file1
  
-Now create a physical link :+Now create the hard link and check the result:
  
 <code> <code>
Ligne 346: Ligne 434:
 Both files, file1 and file2, are referenced by the same inode. The number of links is therefore increased by 1 (the number in the third field). Both files, file1 and file2, are referenced by the same inode. The number of links is therefore increased by 1 (the number in the third field).
  
-<WRAP centre round important 60%>+<WRAP center round important 60%>
 **Important**: A physical link can only be created if the two files are in the same filesystem and the source file exists. **Important**: A physical link can only be created if the two files are in the same filesystem and the source file exists.
 </WRAP> </WRAP>
Ligne 354: Ligne 442:
 A symbolic link is a **shortcut** to another file or directory. A symbolic link is created using the following command: A symbolic link is a **shortcut** to another file or directory. A symbolic link is created using the following command:
  
-* ln -s filename shortcut_name+  * ln -s filename shortcut_name
  
-To illustrate this point, type the following command line:+To illustrate this point, type the following command:
  
 <code> <code>
Ligne 372: Ligne 460:
  
 <WRAP center round important 60%> <WRAP center round important 60%>
-**Important**: A symbolic link can be created even if the two files are in two different filesystems and even if the source file does not exist.+**Important**: A symbolic link can be created even if the two files are in two different file systems and even if the source file does not exist.
 </WRAP> </WRAP>
  
 ----- -----
 Copyright © 2024 Hugh Norris. Copyright © 2024 Hugh Norris.

Piste : • RH12414 - Validation des Acquis • RH12404 - Commandes de Base et de Manipulation de Fichiers Texte • DOF200 - Présentation de la Formation • SER704 - La Sauvegarde, la Restauration et la Réplication • workbook2 • LDF705 - OHAI et Attributs Automatiques, Databags • SER406 - Sauvegardes et Replication • Notions de Base de TCP/IP v4 • LCF903 - Scripting Shell • SER400 - Administration des serveurs MySQL-MariaDB

Menu