CN115878377A - Power failure protection method and device for embedded Linux system and computer equipment - Google Patents

Abstract

The application provides a power-down protection method, device and computer equipment for an embedded Linux system. The method includes the following steps: dividing the non-volatile storage medium of the Linux system into a system partition, a configuration partition, a first application partition, a second application partition, and a second application partition. 2. Application partition and data partition; use overlayfs to jointly mount the system partition to the root path, the system partition is a read-only layer, and the memory file system is a read-write layer, which not only realizes the root path can be read and written, but also protects the system partition; The configuration partition is mounted with a power-failure safe file system, redundancy and verification measures are applied, and space and efficiency are sacrificed in exchange for the power-failure security of writing configuration files at runtime, ensuring the atomicity of adding, deleting, modifying and checking configuration data, and using user space The file system provides a friendly user interface; the two application partitions are redundant backups of each other, and the backup application partition can be used for software upgrades, even if the upgrade is interrupted unexpectedly, it will not affect the operation of the device.

Description

Embedded Linux system power failure protection method, device and computer equipment

technical field

The present application relates to the technical field of embedded computer system data security, in particular to a power-down protection method, device and equipment for an embedded Linux system.

Background technique

For most industrial application computer equipment, when shutting down the equipment, the power supply of the equipment is directly cut off, instead of first executing the software shutdown and then cutting off the power supply like a home computer. Or shut down the device abnormally due to abnormal power failure, instantaneous undervoltage and other reasons during operation. For the embedded Linux system running in the industrial computer equipment, if the data is being written when the power of the equipment is directly cut off, the file system may be damaged, and if the key files are damaged, the equipment will not be able to start and run. At present, most power-down protection mechanisms use hardware means such as backup batteries, large capacitors, and undervoltage detection circuits. These methods not only require additional hardware costs, but also make the system hardware more complicated.

Contents of the invention

In order to solve the above-mentioned technical problems, this application provides a power-down protection method, device and computer equipment for an embedded Linux system, which can be realized through pure software means without backup battery, under-voltage detection circuit, and large capacitor energy storage. To realize the power-down protection of the embedded Linux system. At the same time, it supports the safe writing of important configuration data during runtime. Even if an abnormal power failure occurs during the writing of configuration files, integrity, security, and atomicity can be guaranteed. This method does not need to put the root file system into the memory file system as a whole, so it is suitable for Linux systems with relatively large root file systems.

In the first aspect, the application provides a power-down protection method for an embedded Linux system, the method comprising the following steps:

Step S1: Divide the non-volatile storage medium into a system partition, a configuration partition, a first application partition, a second application partition, and a data partition, and perform initialization;

Step S2: Jointly mount the system partition and the memory file system to the root path;

Step S3: Check and repair the data partition, mount the data partition;

Step S4: establishing a configuration partition index table, repairing the configuration partition, constructing a user space file system, and mounting the user space file system;

Step S5: Check and repair the first application partition or the second application partition, and mount the first application partition and the second application partition.

In the above scheme, the non-volatile storage medium of the Linux system is divided into system partition, configuration partition, first application partition, second application partition, and data partition; the system partition is jointly mounted to the root path using overlayfs, and the system partition It is a read-only layer, and the memory file system is a read-write layer, which not only realizes the root path can be read and written, but also protects the system partition; mount the configuration partition with a power-fail safe file system, apply redundancy and verification measures, and sacrifice Space and efficiency, in exchange for the power-off security of writing configuration files at runtime, ensuring the atomicity of adding, deleting, modifying and querying configuration data, and using the user space file system to provide a friendly user interface; the two application partitions are redundant backups of each other, and the backup application partition It can be used for software upgrade, even if the upgrade is interrupted unexpectedly, it will not affect the operation of the device.

In a specific implementation, the step S2 includes:

Step S21: Build a system partition mount program, disguised as a system init program, wherein the path of the init program includes at least /sbin/init;

Step S22: Using overlayfs in the above-mentioned system partition mount program to jointly mount the system partition and the memory file system to a temporary path, wherein the read-only layer of the joint mount is the system partition, and the read-write layer is the memory file system ;

Step S23: calling the pivot_root command to switch the temporary path to the root path;

Step S24: call the mount command and use the remount, ro parameters to switch the system partition to read-only;

Step S25: Execute the real system init program.

In a specific implementation, the step S3 includes:

Step S31: calling the fsck command to use the -y parameter to check and try to repair the data partition;

Step S32: judging whether there is an abnormality that cannot be repaired according to the return value of fsck, and if so, formatting the data partition;

Step S33: call the mount command to mount the data partition.

In a specific implementation, the configuration partition includes file system metadata, Page Bitmap, multiple pages storing key-value pairs, redundant Page Bitmap, and redundant file system metadata area, wherein,

Each sector of the Page Bitmap area includes a Bitmap of 496 bytes and a check value of 16 bytes, the check value is used to check the integrity of the Bitmap, and the Bitmap is used to mark whether the Page is occupied, and each Page corresponds to a Bit;

Described Page is made up of header, header verification and data area, and wherein, described header area is made of key, reserved segment, data length, data verification, and the key of header area and the file path in Linux system Correspondingly, the header verification is used to check the integrity of the header itself, and the data verification is used to check the integrity of the data.

In a specific implementation, the step S4 includes:

Step S41: Open the block device corresponding to the configuration partition in O_DIRECT and O_SYNC modes;

Step S42: Quickly scan the configuration partition to establish a file index, wherein the quick scan uses file system metadata to locate the Page location, and uses PageBitmap to determine whether the Page is valid;

Step S43: check the integrity of the data through the verification mechanism of the Page, and use redundant data to repair when abnormalities are found;

Step S44: Construct and mount a user space file system, wherein the user space file system is a window for the user space to access the configuration partition.

In a specific implementation, the step S5 includes:

Step S51: mark the first application partition and the second application partition as default application partitions or backup application partitions, wherein the basis of the marking operation is the configuration file of the configuration partition;

Step S52: Check whether there is an abnormality in the backup application partition that cannot be repaired, and if so, use the data of the default application partition to repair the backup application partition;

Step S53: mount the backup application partition in read-write mode;

Step S54: Mount the default application partition in read-only mode.

In a second aspect, the embodiment of the present application provides a power-down protection device for an embedded Linux system, the device comprising:

The partition management module is used to divide the non-volatile storage medium into a system partition, a configuration partition, a first application partition, a second application partition, and a data partition, and perform initialization;

The system partition mount module, which jointly mounts the system partition and the memory file system to the root path;

Data partition mount module, used to check and repair data partition, mount data partition;

The configuration partition management module is used to establish the configuration partition index table, repair the configuration partition, construct the user space file system, and mount the user space file system;

The application partition mounting module is used to check and repair the first application partition or the second application partition, and mount the first application partition and the second application partition.

In the above scheme, the non-volatile storage medium of the Linux system is divided into system partition, configuration partition, first application partition, second application partition, and data partition; the system partition is jointly mounted to the root path using overlayfs, and the system partition It is a read-only layer, and the memory file system is a read-write layer, which not only realizes the root path can be read and written, but also protects the system partition; mount the configuration partition with a power-fail safe file system, apply redundancy and verification measures, and sacrifice Space and efficiency, in exchange for the power-off security of writing configuration files at runtime, ensuring the atomicity of adding, deleting, modifying and querying configuration data, and using the user space file system to provide a friendly user interface; the two application partitions are redundant backups of each other, and the backup application partition It can be used for software upgrade, even if the upgrade is interrupted unexpectedly, it will not affect the operation of the device.

In a third aspect, the embodiment of the present application provides a computer device, the computer device includes a processor and a non-volatile memory, the non-volatile memory stores a computer program, and the computer program runs on the processor The method described in any one of the embodiments of the first aspect is realized when executed.

Description of drawings

In order to illustrate the technical solution of the present invention more clearly, the following drawings will be briefly introduced in the embodiments. It should be understood that the following drawings only show some embodiments of the present invention, and therefore should not be regarded as It is regarded as limiting the protection scope of the present invention.

Fig. 1 is the overall flow of the method provided by the embodiment of the present application;

Fig. 2 is the partition situation of the non-volatile storage medium of the embodiment of the present application, and the file system type, mount mode, and mount path of each partition;

Fig. 3 is the configuration partition file system general framework of the embodiment of the present application;

Fig. 4 is the configuration partition file system PageBitmap area composition of the embodiment of the present application;

FIG. 5 shows the Page composition of the configuration partition file system according to the embodiment of the present application.

Detailed ways

In order to facilitate the understanding of the methods provided in the embodiments of this application, first introduce the terms involved:

Linux: The name of an operating system kernel. The operating system based on this kernel is generally referred to as the Linux operating system.

Root path: Everything under the Linux operating system is a file, each file corresponds to a path, and the root path is the root of all paths.

aufs: The full name is Advanced Multi-layered unification filesytem, the main function is to combine multiple directories into one directory for external use.

overlayfs: A file system that comes with the Linux kernel. Its main function is to combine multiple directories into one directory for external use.

ext4: The fourth extended file system (English: Fourth extended file system), is the most commonly used file system under Linux system, and it is a file system with log function.

tmpfs: Temporary file system is a memory-based file system, which is similar to the virtual disk ramdisk, but not exactly the same. Like ramdisk, tmpfs can use RAM, but it can also use swap partition for storage, and the traditional ramdisk is a block device, it must be formatted with mkfs before it can be really used. And tmpfs is a file system, not a block device, just install it, you can use it. tmpfs is the best memory based file system.

Mount: The process by which an operating system makes computer files and directories on a storage device accessible to users through the computer's file system.

SDK: Software development kit is a collection of development tools for building application software for specific software packages, software frameworks, hardware platforms, operating systems, etc.

bitmap: A data structure that represents a dense set in a finite field, each element appears at least once, and no other data is associated with the element. It is widely used in indexing, data compression, etc.

init: One of the indispensable programs in Linux system operation. The so-called init process, which is a user-level process started by the kernel. The kernel looks for it in several places where init has been used in the past, and its correct location (for Linux systems) is /sbin/init.

VFS: (virtual File System) uses standard Unix system calls to read and write different file systems located on different physical media, and provides a unified application programming interface for various file systems. VFS is a glue layer that allows system calls such as open(), read(), and write() to work regardless of the underlying storage medium and file system type.

User space file system: (Filesystem in Userspace, referred to as FUSE) is a concept in the operating system, which refers to a file system completely implemented in user mode. Currently Linux supports this through kernel modules.

eMMC: Embedded Multi Media Card, embedded multimedia card.

fsck: file system check, a tool in the Linux system, used to check and maintain the file system. If the system is powered off or the disk has problems, you can use the fsck command to check and try to repair the file system.

MD5: (MD5 Message-Digest Algorithm), an information digest algorithm that can generate a 128-bit (16-byte) hash value to ensure the integrity and consistency of information.

DDR: Double Data Rate SDRAM, double-rate synchronous dynamic refresh random access memory, generally called a memory stick.

In order to have a further understanding of the technical solution of the present application, a specific implementation case of the technical solution of the present application will be described in detail below in conjunction with the accompanying drawings. The hardware platform information involved in this case is as follows:

Processor: 6-core ARMv8.2

Storage: 16GB eMMC, 8GB DDR

OS: Ubuntu 18.04.5LTS

Kernel version: Linux4.19.140

With reference to Fig. 1, this concrete implementation case comprises the following steps:

Step S1: Divide the non-volatile storage medium into a system partition, a configuration partition, a first application partition, a second application partition, and a data partition, and perform initialization.

As shown in Figure 2, the non-volatile storage medium eMMC of the Linux system is divided into five areas, which are system partition, configuration partition, first application partition, second application partition, and data partition, and format the above partitions , and populate the initial file.

The file system type of the system partition is ext4. The initial files mainly include the kernel image, device tree, root file system, SDK library files, etc. The above files are related to whether the device can be started, so they need to be protected.

The file system type of the configuration partition is a custom file system. The initial files mainly include operating system configuration files and application software configuration files. The configuration files need to be updated during runtime, so the configuration partition must be mounted in a readable and writable manner It is safe to power off when entering.

The file system type of the first application partition and the second application partition is ext4, and the initial files mainly include business software executable files, algorithm software executable files, dynamic library files, and algorithm model data files.

The file system type of the data partition is ext4, which stores data generated during the operation of the device, such as log files and temporary files, without initial files.

Step S2: Jointly mount the system partition and the memory file system to the root path;

Specifically, this step includes the following steps:

Step S21: Construct a system partition mount program disguised as a system init program, wherein the path of the init program includes at least /sbin/init.

Step S22: Use overlayfs in the above system partition mounting procedure to realize the joint mounting of the system partition and the memory file system tmpfs to a temporary path, wherein the system partition is used as the read-only layer of overlayfs, and the memory file system tmpfs is used as the read-write layer of overlayfs , which not only realizes the root path can be read and written, but also achieves the read-only protection of the system partition.

Step S23: calling the pivot_root command to switch the temporary path to the root path;

Step S24: Call the mount command and use the remount, ro parameters to switch the system partition to read-only, thereby prohibiting any data from being written into the system partition at the operating system level, and preventing damage during power failure.

Step S25: Execute the real system init process. The init program is one of the indispensable programs in the operation of the Linux system. It is the first user-level process started by the kernel, and its path is generally /sbin/init.

Step S3: Check and repair the data partition, mount the data partition.

Specifically, the following steps are included:

Step S31: Invoke the fsck command with the -y parameter to check and try to repair the data partition.

Step S32: Determine whether there is an abnormality that cannot be repaired according to the return value of fsck, and if so, format the data partition.

Step S33: Step S33: calling the mount command to mount the data partition.

Specifically, after the inspection and repair are completed in step S32, the data partition is mounted to the data subdirectory of the user directory.

The data partition stores the data generated during the operation of the device, including log files and temporary files. Even if these files are lost, the normal operation of the device will not be affected. Therefore, when an unrepairable abnormality is detected in the data partition, the data partition can be formatted directly.

Since the subsequent steps will generate operation logs, these logs need to be stored in the data partition, so the data partition must be mounted before the configuration partition and application partition.

Step S4: establishing a configuration partition index table, repairing the configuration partition, constructing a user space file system, and mounting the user space file system.

Specifically, during specific mounting, the partition index table, repair configuration partition, and build user space file system are mounted to the config subdirectory of the user directory.

Referring to Figure 3, the configuration partition shown includes at least the following areas: file system metadata, PageBitmap, multiple Pages for storing key-value pairs, redundant Page Bitmap, redundant file system metadata, configured for user program access Partition, combined with the Linux user space file system, the user program can read and write the files of the configuration partition just like reading and writing ordinary files.

Referring to Figure 4, in the PageBitmap area, 512 bytes of each disk sector includes a 496-byte Bitmap and 16-byte checksum data, and the integrity of the Bitmap is checked using the MD5 checksum. The Bitmap is used to mark whether the Page is occupied, and each Page corresponds to a Bit.

Referring to Figure 5, each Page consists of a 133-byte header, a 16-byte header checksum, and a variable-length data area, where the header area consists of a key, reserved segment, data length, and data checksum. The key in the header area corresponds to the file path in the Linux system, and the maximum data length is determined when the file system is formatted. The header verification is used to check the integrity of the header itself, and the data verification is used to check the integrity of the data, both of which use MD5 value verification.

The same file is repeatedly saved in 3 pages. If a page loses power when it is written, when the file is accessed, it is checked that the duplicate consistency of the file is abnormal, and the redundant page will be used. To restore the abnormal Page. Through the data verification and redundancy backup mechanism, the atomicity of adding, deleting, modifying, and checking operations is ensured at the expense of space and efficiency, so that even if a power failure occurs during the configuration file writing process, the configuration file can be automatically restored after the power is turned on again. Roll back to the full state before the modification.

Specifically, the following steps are included:

Step S41: Open the block device corresponding to the configuration partition in O_DIRECT and O_SYNC modes;

Step S42: Quickly scan the configuration partition to establish a file system index, wherein the quick scan uses the file system metadata to locate the Page position, and uses the Page Bitmap to determine whether the Page is valid;

Step S43: check the integrity of the data through the verification mechanism of the Page, and use redundant data to repair when abnormalities are found;

Step S44: Construct a user space file system and mount it to the config subdirectory of the user directory, wherein the user space file system is a window for the user space to access the configuration partition.

Step S5: Check and repair the first application partition or the second application partition, and mount the first application partition and the second application partition.

Specifically, the following steps are included:

Step S51: mark the first application partition and the second application partition as default application partitions or backup application partitions, wherein the basis of the marking operation is the configuration file of the configuration partition;

Specifically, the first application partition and the second application partition are determined as default application partitions or backup application partitions according to the configuration items in the configuration file app-block-device of the configuration partitions.

Step S52: Check whether there is an abnormality in the backup application partition that cannot be repaired, and if so, use the data of the default application partition to repair the backup application partition;

Step S53: mount the backup application partition in read-write mode;

Step S54: mount the default application partition in read-only mode.

Specifically, check and repair the backup application partition, mount the default application partition to the app subdirectory of the user directory, and mount the backup application partition to the app-backup subdirectory of the user directory.

The default application partition is mounted in read-only mode, and any data writing is prohibited to prevent damage during power failure, so there is no need for detection and repair; the backup application partition is mounted in read-write mode for application software upgrades. The upgrade process is due to power failure Or when the network is terminated abnormally, when the system restarts, if an unrepairable abnormality or data integrity problem is detected in the backup application partition, the default application partition will be used to restore the backup application partition.

It can be seen from the above description that the method provided by the embodiment of the present application divides the non-volatile storage medium of the Linux system into a system partition, a configuration partition, a first application partition, a second application partition, and a data partition; the system partition uses overlayfs Jointly mount to the root path, the system partition is a read-only layer, and the memory file system is a read-write layer, which not only realizes the root path can be read and written, but also protects the system partition; mount the configuration partition using a power-failure safe file system, Apply redundancy and verification measures, sacrifice space and efficiency, in exchange for power-off security when writing configuration files at runtime, ensure the atomicity of adding, deleting, modifying and checking configuration data, and use user space file systems to provide friendly user interfaces; two application partitions Mutual redundant backup, the backup application partition can be used for software upgrade, even if the upgrade is interrupted unexpectedly, it will not affect the operation of the device.

Corresponding to the above method embodiment, the present application also provides a power-down protection device for an embedded Linux system, the device comprising:

The partition management module is used to divide the non-volatile storage medium into a system partition, a configuration partition, a first application partition, a second application partition, and a data partition, and perform initialization;

The system partition mount module, which jointly mounts the system partition and the memory file system to the root path;

Data partition mount module, used to check and repair data partition, mount data partition;

The configuration partition management module is used to establish the configuration partition index table, repair the configuration partition, construct the user space file system, and mount the user space file system;

The application partition mounting module is used to check and repair the first application partition or the second application partition, and mount the first application partition and the second application partition.

For the specific functions of the above modules, reference may be made to relevant descriptions in the method, and details will not be repeated here.

It can be seen from the above description that the device provided by this application divides the non-volatile storage medium of the Linux system into a system partition, a configuration partition, a first application partition, a second application partition, and a data partition; The system partition is a read-only layer, and the memory file system is a read-write layer, which not only makes the root path readable and writable, but also protects the system partition. Residual and verification measures, sacrifice space and efficiency, in exchange for power-off security of runtime configuration file writing, ensure the atomicity of adding, deleting, modifying and checking configuration data, and use user space file system to provide friendly user interface; two application partitions are mutually Redundant backup, the backup application partition can be used for software upgrade, even if the upgrade is interrupted unexpectedly, it will not affect the operation of the device.

In addition, a computer device is also provided, the computer device includes a processor and a non-volatile memory, the non-volatile memory stores a computer program, and when the computer program is executed on the processor, the above-mentioned A power-down protection method for an embedded Linux system.

One or more embodiments of this description are intended to embrace all such alterations, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omission, modification, equivalent replacement, improvement, etc. made within the spirit and principles of one or more embodiments of this specification shall be included within the protection scope of this application.

The above is only the specific implementation of the application, but the scope of protection of the application is not limited thereto. Any changes or substitutions that can be easily imagined by those skilled in the art within the technical scope disclosed in the application should be covered within the scope of protection of this application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims (8)

1. A power failure protection method for an embedded Linux system is characterized by comprising the following steps:

step S1: dividing the nonvolatile storage medium into a system partition, a configuration partition, a first application partition, a second application partition and a data partition, and initializing;

step S2: jointly mounting a system partition and a memory file system to a root path;

and step S3: checking and repairing the data partition and the mounting data partition;

and step S4: establishing a configuration partition index table, repairing a configuration partition, constructing a user space file system, and mounting the user space file system;

step S5: and checking and repairing the first application partition or the second application partition, and mounting the first application partition and the second application partition.

2. The power down protection method for embedded Linux system according to claim 1,

the step S2 includes:

step S21: constructing a system partition mounting program, and disguising the system partition mounting program into a system init program, wherein the path of the init program at least comprises/sbin/init;

step S22: in the system partition mounting program, adopting overlayfs to realize that a system partition and a memory file system are jointly mounted to a temporary path, wherein a read-only layer of the joint mounting is the system partition, and a read-write layer is the memory file system;

step S23: calling a pivot _ root instruction to switch the temporary path into a root path;

step S24: calling a mount instruction to use a remount and ro parameter to switch the system partition into read-only;

step S25: and executing a real system init program.

3. The embedded Linux system power-fail protection method according to claim 1,

the step S3 includes:

step S31: invoking the fsck instruction to check and attempt to repair the data partition using the-y parameter;

step S32: judging whether an irreparable abnormality exists according to the fsck return value, and if so, formatting the data partition;

step S33: and calling a mount instruction to mount the data partition.

4. The embedded Linux system power-fail protection method according to claim 1,

the configuration partition includes file system metadata, a Page Bitmap, a plurality of pages storing key-value pairs, a redundant Page Bitmap, a redundant file system metadata region, wherein,

each sector of the Page Bitmap area comprises a Bitmap with 496 bytes and a check value with 16 bytes, wherein the check value is used for checking the integrity of the Bitmap, the Bitmap is used for marking whether the Page is occupied or not, and each Page corresponds to one Bit;

the Page comprises a header, a header check sum data area, wherein the header area comprises a key, a reserved section, a data length and a data check, the key of the header area corresponds to a file path in the Linux system, the header check is used for checking the integrity of the header, and the data check is used for checking the integrity of the data.

5. The embedded Linux system power-fail protection method according to claim 4,

the step S4 includes:

step S41: opening the block device corresponding to the configuration partition in the O _ DIRECT mode and the O _ SYNC mode;

step S42: rapidly scanning a configuration partition and establishing a file index, wherein the rapid scanning uses file system metadata to position a Page position, and uses a Page bitmap to judge whether the Page is effective;

step S43: checking the integrity of data through a checking mechanism of a Page, and repairing by using redundant data when abnormality is found;

step S44: and constructing and mounting a user space file system, wherein the user space file system is a window of the user space access configuration partition.

6. The power down protection method for embedded Linux system according to claim 1,

the step S5 includes:

step S51: marking the first application partition and the second application partition as a default application partition or a backup application partition, wherein the marking operation is based on a configuration file of a configuration partition;

step S52: checking whether the backup application partition has an abnormality which can not be repaired, and if so, repairing the backup application partition by using the data of the default application partition;

step S53: mounting the backup application partition in a read-write mode;

step S54: the default application partition is mounted in read-only mode.

7. The power failure protection device for the embedded Linux system is characterized by comprising:

the partition management module is used for dividing the nonvolatile storage medium into a system partition, a configuration partition, a first application partition, a second application partition and a data partition and initializing the system partition, the configuration partition, the first application partition, the second application partition and the data partition;

the system partition mounting module is used for mounting the system partition and the memory file system to a root path in a combined manner;

the data partition mounting module is used for checking and repairing the data partitions and mounting the data partitions;

the configuration partition management module is used for establishing a configuration partition index table, repairing a configuration partition, constructing a user space file system and mounting the user space file system;

and the application partition mounting module is used for checking and repairing the first application partition or the second application partition and mounting the first application partition and the second application partition.

8. A computer arrangement, characterized by at least a processor and a non-volatile memory, in which a computer program is stored which, when executed on the processor, implements the method of any one of claims 1 to 6.

Images