Cobbler Automated Deployment

One: The difference between Cobbler and PXE:

PXE (Preboot Execute environment, pre-boot execution Environment) is a technology developed by Intel Corporation, requires hardware support of network card, works in the network mode of C/s, supports client to download image from server through network, and support the client machine to boot the operating system through the network, during the boot process, the client needs to assign the IP address, mask, gateway, and associated configuration file path from the server, and then use TFTP (trivial file transfer Protocol) protocol to download a boot package from the server to the native memory execution, by this boot package to complete the client's basic software settings, thereby booting the pre-installed terminal operating system, PXE can boot a variety of operating systems, such as: CentOS, Redhat, Windows 7, and so on.

cobbler{[? kɑ:bl? ( R)]} translated to Chinese is a cobbler, but in Linux is a free open source software for system installation deployment, cobbler is two Python-based development of PXE technology, provides CLI (command line) and Web management mode, also provides API (application programming Interface) interface, convenient two times development use, compared to Pxe,cobbler support multi-system simultaneous boot, and PXE can only boot one system installation, cobbler usually with open source software puppet, saltstack with the use of Optimize the batch installation and deployment of the operating system, so as to realize the unattended installation and service deployment of the high-volume Linux system, and greatly improve the efficiency of operation and maintenance.

Process:

1. Depending on the boot order of the local BIOS setup, the client sends a protocol message based on the RARP Reverse Address Resolution protocol (Reverse addresses Resolution Protocol), which contains its own MAC address, but no IP address, if it is started from the network card.

After the 2.DHCP server receives the request, it returns to the client IP address, mask, gateway, and also contains a TFTP protocol-based file server address and the name of the file to or get

3. Get the file from the TFTP server

4. Put the downloaded file in memory to run

5. Give the system's boot to the file execution, which is equivalent to the bootloader process loaded from the hard disk

6. To load the kernel vmlinux and initrd.img, this procedure is as follows:

(1) bootloader is grub, which initializes initrd.img into a device/dev/intrd, and then bootloader the kernel and the contents of the/DEV/INITRD into memory.
(2) During kernel initialization, the kernel/dev/initrd the contents of the device and copies it to the/DEV/RAM0 device.
(3) The kernel mounts the/DEV/RAM0 device as the original root file system in a read-write manner.
(4) If/DEV/RAM0 is designated as a real root filesystem, then the kernel does not perform (5), (6), (7) operations, because this is done to help the kernel load the final root filesystem.
(5) Executing the/LINUXRC file on INITRD, LINUXRC is usually a script file that loads the kernel to access the required drivers for the root filesystem and loads the root file system.
(6)/LINUXRC execution is complete, the real root filesystem is mounted.
(7) If the real root filesystem exists in the/INITRD directory, then/DEV/RAM0 will move from/to/INITRD. Otherwise, if the/INITRD directory does not exist,/DEV/RAM0 will be uninstalled.
(8) Perform the normal startup process on the real root file system and execute the/sbin/init.

Cobbler Automated Deployment