CN110737501A - Method and system for realizing functions of check point and recovery point in Docker container - Google Patents
Method and system for realizing functions of check point and recovery point in Docker container Download PDFInfo
- Publication number
- CN110737501A CN110737501A CN201810788726.4A CN201810788726A CN110737501A CN 110737501 A CN110737501 A CN 110737501A CN 201810788726 A CN201810788726 A CN 201810788726A CN 110737501 A CN110737501 A CN 110737501A
- Authority
- CN
- China
- Prior art keywords
- module
- pid
- mirror
- point
- target process
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 223
- 230000006870 function Effects 0.000 title claims abstract description 95
- 238000011084 recovery Methods 0.000 title claims abstract description 74
- 230000008569 process Effects 0.000 claims abstract description 201
- 230000003071 parasitic effect Effects 0.000 claims abstract description 30
- 238000007689 inspection Methods 0.000 claims description 7
- 230000008014 freezing Effects 0.000 claims description 4
- 238000007710 freezing Methods 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 claims description 3
- 230000005012 migration Effects 0.000 abstract description 16
- 238000013508 migration Methods 0.000 abstract description 16
- 230000010076 replication Effects 0.000 abstract 1
- 238000001824 photoionisation detection Methods 0.000 description 22
- 238000005516 engineering process Methods 0.000 description 9
- 238000001514 detection method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 239000000725 suspension Substances 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Retry When Errors Occur (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention relates to a method and a system for realizing functions of a check point and a recovery point in a Docker container, wherein the system comprises a kernel space and a user space, wherein: the kernel space includes: a basic process and application program module for realizing the functions of the computer, a parasitic code module for acquiring the mirror image information of the process running in the user space through the basic process and application program module, and a mirror image process and application program module for saving the mirror image information collected from the suspended process; the user space includes: the system comprises a process and application program module for running corresponding programs, a checkpoint function module for suspending the process and collecting mirror image information of the suspended process, and a mirror process and application program module for storing the mirror image information. The invention makes the functions of container pause, recovery, dynamic migration, container integral migration, seamless kernel upgrade, network load balance, desktop environment pause and recovery, remote debugging of application programs, process replication and the like become possible.
Description
Technical Field
The invention relates to the technical field of virtual machine data recovery, in particular to a function realization method and a function realization system for detection points and recovery points in Docker containers.
Background
Cloud computing is a emerging method for sharing infrastructure, which distributes computing tasks on a resource pool formed by a large number of computers, so that various application systems can obtain computing power, storage space and various software services according to requirements, the bottom layer of the cloud computing needs virtualization technical support, compared with the famous virtualization technology of VMware corporation, the method provides software and services for cloud computing and hardware virtualization, products comprise VMware work and VMware Server, users are allowed to create and run a plurality of X86 virtual machines, and the virtual machines can be suspended and resumed.
Docker technology is based on cloud computing, is open source code software projects, and enables work of deploying application programs under software containers to be carried out automatically, so that additional software abstraction layers and an automatic management mechanism of operating system layer virtualization are provided on a Linux operating system, the Docker packages and isolates processes based on cgroup of a Linux kernel, namespace, Union FS of AUFS class and other technologies, belongs to virtualization technology of the operating system layer, the isolated processes are also called containers because the isolated processes are independent of hosts and other isolated processes, the Docker packages in steps on the basis of the containers, and the like from file systems, network interconnection to process isolation, and the like, greatly simplifies the creation and maintenance of the containers, so that the Docker technology is lighter and quicker than a virtual machine technology, and the relationship between the Docker and check points and restore points is shown in FIG. 1.
The traditional virtual machine technology is that after sets of hardware are virtualized, complete operating systems are operated on the hardware, required application processes are operated on the systems, the application processes in containers are directly operated in a kernel of a host, the containers are not provided with own kernels, and hardware virtualization is not performed, so that the containers are lighter and lighter than traditional virtual machines, the occupied space is smaller, some containers are only dozens of M, the space of the traditional virtual machines needs GB, and thousands of containers can be installed on computers provided with dozens of virtual machines.
In the use of the virtual machine, the virtual machine can be suspended from running, a snapshot is created, snapshots are previewed, and whether the system is to be restored to the state when the snapshot is created is determined, i.e., storage functions, which allow an administrator to create restoration points for the operating system, applications and data of virtual machines at a specific time.
At present, under a Windows platform, a VMware work, VMware Server virtual machine series, Oracle VirtualBox is part of the virtualization platform technology of Oracle language company XVM, provides users with operating systems of other x86 virtualized on 32-bit or 64-bit Windows, Solaris and Linux operating systems, and all the virtual machines provide functions of pause, snapshot, migration and recovery.
At present, the functions of container checkpoint recovery are realized in Redhat, Centos, Ubuntu and Kangsu operating systems under an X86 architecture, while the functions of container checkpoint and recovery point are not realized under an MIPS Loongson platform.
In addition, the pause and restore speed of the current virtual machine is low, minute level is needed, the checkpoint restoration in the container can be completed within second level, and the response speed is improved by more than 90%.
Therefore, it is necessary to design methods and systems for implementing functions of check points and recovery points in a Docker container.
Disclosure of Invention
In order to solve the defects in the prior art, the invention provides kinds of function realization systems of detection points and recovery points in a Docker container, wherein the function realization systems comprise a kernel space and a user space, and the function realization systems comprise:
the kernel space includes base process and application modules, native code modules, and mirror process and application modules, wherein,
the basic process and the application program module are used for realizing the basic functions of the computer;
the parasitic code module is connected with the basic process and the application module and is used for acquiring the mirror image information of the process running in the user space through the basic process and the application module;
the mirror process and application module is used for saving mirror information collected from the suspended process;
the user space includes process and application modules, a system call module, a checkpoint function module, and a recovery point function module, wherein,
the process and application program module is connected with the basic process and application program module of the kernel space through the system calling module and is used for calling a program needing to be operated in the kernel space through the system calling module;
the checkpoint function module is connected with the basic process and the application program module, the parasitic code module and the mirror process and application program module in the kernel space, and is used for suspending the running process in the user space, collecting the mirror information of the suspended process through the parasitic code module, and storing the mirror information to the mirror process and the application program module;
and the recovery point function module is connected with the mirror image process and the application program module and is used for recovering the suspended process according to the mirror image information.
And the parasitic code module is formed by injecting the parasitic code into the checkpoint function module through a ptrace interface of the kernel space.
And the checkpoint function module suspends the running process in the user space through a ptrace interface of the kernel space.
The checkpoint function module suspends the running process and acquires the mirror image information of the process through the following steps:
step Sa: freezing a target process through a ptrace interface, and injecting a parasitic code into the target process;
and Sb: the parasitic code invades a corresponding specific target process, and the process is transparently monitored to obtain a file descriptor;
step Sc: acquiring socket and netns information of a target process;
step Sd: acquiring the content of a specific PID in a target process procfs;
step Se: obtaining and storing the system call of the target process through a parasitic code module;
step Sf: storing a graphic file of a memory page of a target process;
step Se: the CRIU checkpoint of all child processes in the target process is checked.
Wherein the functions of the checkpoint function module further comprise: and in the process of suspending the running process, collecting the mirror image information of the process running, unlocking the mirror image information and returning a result after the running process is suspended, so that the recovery point functional module can recover the corresponding process.
The invention also provides Docker container check point and recovery point function realization methods, which comprises the following steps:
implementation of checkpoint functions:
step S1: tracking and freezing a suspended target process through a ptrace interface;
step S2: injecting a parasitic code into a target process, and performing transparent monitoring on the target process to obtain a file descriptor;
step S3: acquiring information of a target process to generate corresponding mirror image information;
step S4: checking CRIU check points of all sub processes in a target process;
and (3) realizing the recovery point function:
step S5: the target process is restored by the mirror information collected in step S3.
In step S1, tracking the target process through the ptrace interface includes the following steps:
step S11: acquiring a process number $ pid of a process group father node of a target process through a command line;
step S12: the backup device traverses the/proc/$ pid/task directory through the process number $ pid and collects thread information;
step S13: the backup continues to traverse/proc/$ pid/task/$ tid/children iteration collection child processes.
In step S3, the acquiring information of the target process includes:
socket and netns information, which is obtained through a/proc file system;
the contents of a particular PID in procfs;
the system call of the target process is obtained through the parasitic code;
and an image file of the target process memory page.
Wherein, the content of the specific PID includes: /PROC/PID/maps,/PROC/PID/map _ files// PROC/PID/status/PROC/PID/mountinfo,/PROC/$ PID/SMAPS,/PROC/$ PID/MMASPORE and/OF/PROC/$ PID/PIPMAP,/PROC/$ PID/FD, network channel parameters or memory mapping.
The image file of the target process memory page comprises: open file, certificate, register, or task state.
In step S5, when restoring the process, the recovery point is restored only when the initial checkpoint has the same process number.
In step S5, the CRIU restores the target process to the state before the checkpoint through the mirror information collected in step S3, thereby restoring the target process.
According to the method and the system for realizing the functions of the check point and the recovery point in the Docker container, provided by the invention, through the setting of the functions of the check point and the recovery point, the functions of container pause, recovery, dynamic migration, container integral migration, seamless kernel upgrade, network load balance, desktop environment pause, recovery, remote debugging of an application program, process copying and the like become possible under a cloud computing system.
Drawings
FIG. 1: the existing docker container and the checking point and the reduction point are in a relationship schematic diagram.
FIG. 2: the functions of the detection point and the recovery point in the Docker container provided by the invention realize the system architecture diagram of the system.
FIG. 3: the invention provides a container heat transfer tool under a dragon core platform and an implementation process thereof.
Description of the reference numerals
10-kernel space, 11-base process and application module, 12-parasitic code module, 13-mirror process and application module;
20-user space, 21-process and application module, 22-system call module, 23-checkpoint function module, 24-recovery point function module.
Detailed Description
In order to further understand the technical solution and the advantages of the present invention , the following describes the technical solution and the advantages thereof in detail with reference to the accompanying drawings.
Fig. 1 is a system architecture diagram of a function implementation system for a check point and a recovery point in a Docker container according to the present invention, as shown in fig. 1, in view of the fact that functions of the check point and the recovery point in the Docker container provided by the present invention are designed in a kernel space, a Linux kernel community does not achieve in design, and the function implementation of the recovery point in the kernel is too complex to be integrated into a kernel primary branch.
Specifically, as shown in fig. 1, the system for implementing functions of an inspection point and a recovery point in a Docker container provided by the present invention includes a kernel space 10 and a user space 20, wherein,
the kernel space 10 includes a base process and application module 11, a native code module 12, and a mirror process and application module 13, and the user space 20 includes a process and application module 21, a system call module 22, a checkpoint function module 23, and a recovery point function module 24.
The checkpoint function module 23 in the user space 20 completes interaction with the kernel space 10 through the ptrace interface, the parasitic code module 12 in the kernel space 10 is formed by injecting a parasitic code into the checkpoint function module 23 through the ptrace interface, and when the checkpoint function module 23 realizes the function thereof, the checkpoint function module 23 completes information acquisition of the target process mainly through the parasitic code module 12.
In the present invention, the detailed relationship of the functional modules is as follows.
The base process and application module 11 provides the underlying application to enable the computer to implement the series of functions, in which case the process and application module 21 in the user space 20 calls the required program from the base process and application module 11 through the system call module 22.
The checkpoint function module 23 is connected to the base process and application module 11, and is configured to suspend a process running in the user space through a ptrace interface; meanwhile, the checkpoint function module 23 is also connected to the parasitic code module 12, and the parasitic code module 12 acquires the mirror image information of the process running in the user space through the basic process and the application program module 11 under the control of the checkpoint function module 23; the checkpoint function 23 then saves the mirrored information to the mirrored process and application module 13 for the later recovery point function 24 to resume the suspended process.
The invention relates to a system for realizing functions of an inspection point and a recovery point in a Docker container, which comprises the following working processes:
, implementation of checkpoint function:
1. target process acquisition and suspension
The checkpoint function injects the parasitic code through the ptrace interface of the kernel space, and then tracks the running process of the user space and suspends all processes.
The checkpoint function module suspends all processes in complete process trees, and the specific tracking method is to obtain the process number ($ pid) of the parent node of process groups through a command line, and then by using the process number $ pid, the backup will traverse/proc/$ pid/task directory, collect thread information, and traverse/proc/$ pid/task/$ tid/children iterate to collect sub-process information, and suspend the running process by using the PTRACE _ segment.
The process tree itself and its processes are "paused" when dumped, the checkpoint function is transparent to the application we are DUMP (injecting), and DUMP specific application approaches are monitoring the application using the ghost code.
2. Extraction of mirror image information
The parasitic code can invade specific processes, transparently monitors the processes, and acquires the file descriptor, the dump memory content, therefore, before the program is normally executed, the parasitic code needs to be executed first, and specifically, the parasitic code can be executed by using system calls such as getitimer () and sign ().
The acquired information includes:
(1) the socket, netns and other information are acquired through a netlink;
(2) the contents of specific PIDs in procfs, e.g./proc/PID/maps,/proc/PID
The system comprises a plurality OF modules, wherein each module comprises a/map _ files/,/PROC/PID/status/PROC/PID/mourninfo,/PROC/$ PID/SMAPS,/PROC/$ PID/MMASPORE,/OF/PROC/$ PID/PIPMAP,/PROC/$ PID/FD, network channel parameters, memory mapping and the like;
(3) system calling of a target process;
(4) image file of target process memory page: containing additional information about the checkpoint process such as open files, certificates, registers, task status, etc.
It uses the/proc file system in Linux system to obtain the process data, where/proc is pseudo file systems (i.e. virtual file systems), and stores series special files of the current kernel running state, so that the user can view the information about the system hardware and the current running process through these files, and even change the running state of the kernel by changing some of them.
3. And finally, checking the CRIU point of each connected child process in the complete process tree (the parent process and all child processes) corresponding to the target process so as to ensure the smooth recovery of the target process in the later period.
Secondly, the realization of the recovery point function:
the recovery point function uses the mirror information collected by the checkpoint function to recover the process, noting that the process can only be recovered if the initial checkpoint has the same process number, if another processes are using this process number, the recovery fails.
The reason why the process is restored with the same process number is that the parent and child processes must be completely restored as they are, the parent process to be restored cannot create a parent-child relationship, and in the present invention, in order to restore the process with the same process number, the kernel is used to influence the process number of the next processes given by the kernel.
The file is opened and located, like state when the recovery point is running, memory is restored to the same state, and all other information from the image file is used to restore the process once the process is restored, the rest of the restorer is deleted, as is the snapshot, the restored process restores control, and continues from the location of the previous checkpoint.
The method and the system for realizing the functions of the check point and the recovery point in the Docker container are suitable for suspending and recovering processes based on an MIPS Loongson platform.
In order to facilitate the technical understanding of the present invention, the dragon core platform is taken as an example to describe two specific embodiments of the present invention in detail.
Example recovery Point function on Container opening
In this embodiment, series mirror image files and log files are generated and stored by using the container recovery point, and the specific flow is as follows:
(1) the user looks over the Loongson server container running state through the command line terminal, and the docker container receives the command, can go to check the state of virtual machine earlier: whether the container is in an abnormal state such as a quit state or a death state; if the container is in an abnormal state, the container is normally started through a multi-Docker command, normal operation of the container is ensured, an application program or a process is normally operated, the operation of the container is suspended by the checkpoint function module, the Docker service where the container is located is inquired, twenty or more image files such as cgroup.img, enentpoll-tfd.img, fifo.img and the like are called to generate the operation of the container, the image files save all states of the container, such as memory states, storage states, network connection states, parent-child process calling relations and the like, and meanwhile, a checkpoint log file and a dump file are generated.
(2) And after the docker container server receives the request, checking the integrity of the file according to the mirror image file, the log file and the dump file generated by the check point, and if the file is occupied, circularly waiting until overtime and returning a result. And if the file is not occupied, locking the file, generating a snapshot for the virtual machine image file, storing the unlocked image file and returning a result.
(3) And a recovery point functional module in the docker container server receives and stores the result, then returns information, and starts an automatic container recovery process. Specifically, the recovery point function module receives the recovery request, calls a container recovery for the container with the recovery function started, and recovers the container according to the mirror image file and the log. And checking whether the file is occupied, if not, locking the mirror image file of the virtual machine, performing recovery operation on the snapshot, unlocking the container mirror image file after the recovery operation is completed, and then returning the result to the controller. An automatic recovery module in the controller saves the results, and then executes recovery operations to run the container.
Example two: realization of thermomigration function of Loongson MIPS platform container
Container Live Migration (Docker Live Migration, also called container Live Migration, container Live Migration), i.e. container Save/Restore (Save/Restore): the running state of the whole container is completely stored, and meanwhile, the running state of the whole container can be quickly restored to an original hardware platform or even different hardware platforms. After recovery, the virtual machine still runs smoothly, the user can not detect any difference, on the traditional X86 virtual machine, the virtual machine hot migration technology is mature, almost all virtualization schemes have realized container hot migration, such as VMware, KVM, Hyper-V, Xen and the like, on an X86 container, the check point and the recovery point of X86 can be used, and on a Loongson MIPS platform, no solution in the aspect is available at present.
Fig. 3 is a container thermomigration tool under a dragon core platform and an implementation process thereof, which are proposed based on the method provided by the invention:
(1) before migration, the running state of the container of the Loongson host A is prepared to be checked, and the feasibility of the container check point is verified.
(2) And if the container runs normally, storing a basic state file of the container, a container state configuration, a root file system of the container, a mirror image file of the container and a log file. The sync and rsync tools may be used to synchronize files to the destination host.
(3) Before restoration, the feasibility of a restoration point is verified on the Loongson host B, and a container is created according to the configuration file.
(4) On the Loongson host A, the running state of the container process is saved, and common migration, iterative migration, incremental backup of a root file system, synchronous files and mirror images can be used.
(5) And on the Loongson host B, recovering a container root file system, recovering a network by the Libnetwork, recovering a data Volume by the Volume Driver and recovering a container process.
In the present invention, the "checkpoint" refers to the current state of a running process in the user space. The "checkpoint function" is able to obtain all relevant information of the current state in order to restore it later to its previous (prior to the checkpoint) state.
In the present invention, the "resume point function module" refers to a function implementation module that resumes the suspended process or migrates the suspended process from the host to another hosts.
In the present invention, the "kernel space" refers to a space in which a part of kernel software and a driver run, and runs at a higher privilege level independently of a general application program, and they reside in a protected memory space and have all the rights to access a hardware device, and will be referred to as kernel space
In the present invention, the term "user space" is distinguished from kernel space, and refers to a space in which an application is running.
In the present invention, the term "ptrace" refers to a kernel function called by a system under linux programming.
In the present invention, the term "socket" refers to a network socket.
In the present invention, the "PID and the process number $ PID" are both process ids, and can be seen by the command ps in the linux system.
In the present invention, a so-called "netlink" stores kernel-related code for use by CRIU.
When the checkpoint and recovery point function is used specifically, the Docker container can be selectively opened or closed according to needs, the current container state can be saved as a snapshot when the container checkpoint recovery point function is opened, and the snapshot can be recovered when the container checkpoint and recovery point function is started next time, so that the container recovery function is realized.
The invention has the following beneficial effects:
1. by setting functions of the check point and the recovery point, functions of container suspension, recovery, dynamic migration, container overall migration, seamless kernel upgrading, network load balancing, desktop environment suspension and recovery, remote debugging of application programs, process copying and the like become possible in the cloud computing system.
2. The check point and the recovery point are arranged in the user space, so that the whole system is stable and reliable, and the method is particularly suitable for the use of containers and other scenes needing the function of the check point recovery point under the Loongson platform.
Although the present invention has been described with reference to the preferred embodiments, it should be understood that the scope of the present invention is not limited thereto, and those skilled in the art will appreciate that various changes and modifications can be made without departing from the spirit and scope of the present invention.
Claims (12)
1, kinds of Docker container check point and recovery point's function implementation system, characterized by that, function implementation system includes kernel space and user space, wherein:
the kernel space includes base process and application modules, native code modules, and mirror process and application modules, wherein,
the basic process and the application program module are used for realizing the basic functions of the computer;
the parasitic code module is connected with the basic process and the application module and is used for acquiring the mirror image information of the process running in the user space through the basic process and the application module;
the mirror process and application module is used for saving mirror information collected from the suspended process;
the user space includes process and application modules, a system call module, a checkpoint function module, and a recovery point function module, wherein,
the process and application program module is connected with the basic process and application program module of the kernel space through the system calling module and is used for calling a program needing to be operated in the kernel space through the system calling module;
the checkpoint function module is connected with the basic process and the application program module, the parasitic code module and the mirror process and application program module in the kernel space, and is used for suspending the running process in the user space, collecting the mirror information of the suspended process through the parasitic code module, and storing the mirror information to the mirror process and the application program module;
and the recovery point function module is connected with the mirror image process and the application program module and is used for recovering the suspended process according to the mirror image information.
2. The system for functional implementation of an inspection point and a recovery point in a Docker container of claim 1, wherein: the parasitic code module is formed by injecting parasitic code into the checkpoint function module through a ptrace interface of the kernel space.
3. The system for functional implementation of an inspection point and a recovery point in a Docker container of claim 1, wherein: and the checkpoint function module suspends the running process in the user space through a ptrace interface of the kernel space.
4. The system for functional implementation of an inspection point and a recovery point in a Docker container of claim 1, wherein: the checkpoint function suspends a running process and acquires mirror information of the process by:
step Sa: freezing a target process through a ptrace interface, and injecting a parasitic code into the target process;
and Sb: the parasitic code invades a corresponding specific target process, and the process is transparently monitored to obtain a file descriptor;
step Sc: acquiring socket and netns information of a target process;
step Sd: acquiring the content of a specific PID in a target process procfs;
step Se: obtaining and storing the system call of the target process through a parasitic code module;
step Sf: storing a graphic file of a memory page of a target process;
step Se: the CRIU checkpoint of all child processes in the target process is checked.
5. The system for functional implementation of an inspection point and a recovery point in a Docker container of claim 1, wherein: the functions of the checkpoint function module further include: and in the process of suspending the running process, collecting the mirror image information of the process running, unlocking the mirror image information and returning a result after the running process is suspended, so that the recovery point functional module can recover the corresponding process.
6, Docker container check point and recovery point function implementation method, characterized by, including the following steps:
implementation of checkpoint functions:
step S1: tracking and freezing a suspended target process through a ptrace interface;
step S2: injecting a parasitic code into a target process, and performing transparent monitoring on the target process to obtain a file descriptor;
step S3: acquiring information of a target process to generate corresponding mirror image information;
step S4: checking CRIU check points of all sub processes in a target process;
and (3) realizing the recovery point function:
step S5: the target process is restored by the mirror information collected in step S3.
7. The method for implementing functions of a search point and a recovery point in a Docker container according to claim 6, wherein in step S1, tracking the target process through a ptrace interface includes the following steps:
step S11: acquiring a process number $ pid of a process group father node of a target process through a command line;
step S12: the backup device traverses the/proc/$ pid/task directory through the process number $ pid and collects thread information;
step S13: the backup continues to traverse/proc/$ pid/task/$ tid/children iteration collection child processes.
8. The method for implementing functions of a check point and a recovery point in a Docker container according to claim 6, wherein the step S3 of obtaining the information of the target process includes:
socket and netns information, which is obtained through a/proc file system;
the contents of a particular PID in procfs;
the system call of the target process is obtained through the parasitic code;
and an image file of the target process memory page.
9. The method of claim 8, wherein the contents of a particular PID comprise: /PROC/PID/maps,/PROC/PID/map _ files// PROC/PID/status/PROC/PID/mountinfo,/PROC/$ PID/SMAPS,/PROC/$ PID/MMASPORE and/OF/PROC/$ PID/PIPMAP,/PROC/$ PID/FD, network channel parameters or memory mapping.
10. The method of claim 8, wherein the image file of the target process memory page comprises: open file, certificate, register, or task state.
11. The method for implementing functions of an inspection point and a recovery point in a Docker container of claim 6, wherein in the step S5, the recovery point recovers when recovering the process only if the initial checkpoint has the same process number.
12. The method of claim 6, wherein in step S5, the CRIU restores the target process to the state before the checkpoint through the mirror information collected in step S3, thereby restoring the target process.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810788726.4A CN110737501A (en) | 2018-07-18 | 2018-07-18 | Method and system for realizing functions of check point and recovery point in Docker container |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810788726.4A CN110737501A (en) | 2018-07-18 | 2018-07-18 | Method and system for realizing functions of check point and recovery point in Docker container |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110737501A true CN110737501A (en) | 2020-01-31 |
Family
ID=69233706
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810788726.4A Pending CN110737501A (en) | 2018-07-18 | 2018-07-18 | Method and system for realizing functions of check point and recovery point in Docker container |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110737501A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112527461A (en) * | 2020-12-17 | 2021-03-19 | 中国人民解放军国防科技大学 | On-line migration method and system for micro virtual machine facing Feiteng platform |
| CN112698873A (en) * | 2020-12-29 | 2021-04-23 | 中车大连机车研究所有限公司 | Docker technology-based operation control method and system for rail transit system |
| CN112835683A (en) * | 2021-03-02 | 2021-05-25 | 杭州雅观科技有限公司 | Containerized deployment system of intelligent community software |
| CN113821333A (en) * | 2020-06-20 | 2021-12-21 | 华为技术有限公司 | Android application program migration method and device |
| CN115437889A (en) * | 2022-11-08 | 2022-12-06 | 统信软件技术有限公司 | Emergency processing method and system and computing equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101093453A (en) * | 2007-07-25 | 2007-12-26 | 湖南大学 | Method for implementing checkpoint of Linux program at user level based on virtual kernel object |
| CN101198930A (en) * | 2005-04-13 | 2008-06-11 | 艾利森电话股份有限公司 | Data value consistency in computer system |
| US20090183027A1 (en) * | 2008-01-11 | 2009-07-16 | International Business Machines Corporation | Checkpointing and restoring user space data structures used by an application |
| US20090252386A1 (en) * | 2008-04-04 | 2009-10-08 | Validity Sensors, Inc. | Apparatus and Method for Reducing Parasitic Capacitive Coupling and Noise in Fingerprint Sensing Circuits |
| CN105843554A (en) * | 2016-03-11 | 2016-08-10 | 中国联合网络通信集团有限公司 | Object storage-based method for data migration and system thereof |
| CN108090001A (en) * | 2017-12-11 | 2018-05-29 | 中标软件有限公司 | A kind of kernel DMA stable states dispatching method and device |
-
2018
- 2018-07-18 CN CN201810788726.4A patent/CN110737501A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101198930A (en) * | 2005-04-13 | 2008-06-11 | 艾利森电话股份有限公司 | Data value consistency in computer system |
| CN101093453A (en) * | 2007-07-25 | 2007-12-26 | 湖南大学 | Method for implementing checkpoint of Linux program at user level based on virtual kernel object |
| US20090183027A1 (en) * | 2008-01-11 | 2009-07-16 | International Business Machines Corporation | Checkpointing and restoring user space data structures used by an application |
| US20090252386A1 (en) * | 2008-04-04 | 2009-10-08 | Validity Sensors, Inc. | Apparatus and Method for Reducing Parasitic Capacitive Coupling and Noise in Fingerprint Sensing Circuits |
| CN105843554A (en) * | 2016-03-11 | 2016-08-10 | 中国联合网络通信集团有限公司 | Object storage-based method for data migration and system thereof |
| CN108090001A (en) * | 2017-12-11 | 2018-05-29 | 中标软件有限公司 | A kind of kernel DMA stable states dispatching method and device |
Non-Patent Citations (2)
| Title |
|---|
| JONATHAN CORBET: "Checkpoint/restart(mostly) in user space", 《LWN.NET:HTTPS://LWN.NET/ARTICLES/452184/》 * |
| 房锦章等: "基于Docker技术的GUI应用的在线迁移研究", 《计算机系统应用》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113821333A (en) * | 2020-06-20 | 2021-12-21 | 华为技术有限公司 | Android application program migration method and device |
| CN112527461A (en) * | 2020-12-17 | 2021-03-19 | 中国人民解放军国防科技大学 | On-line migration method and system for micro virtual machine facing Feiteng platform |
| CN112527461B (en) * | 2020-12-17 | 2024-01-09 | 中国人民解放军国防科技大学 | Online migration method and system of micro virtual machine facing Feiteng platform |
| CN112698873A (en) * | 2020-12-29 | 2021-04-23 | 中车大连机车研究所有限公司 | Docker technology-based operation control method and system for rail transit system |
| CN112835683A (en) * | 2021-03-02 | 2021-05-25 | 杭州雅观科技有限公司 | Containerized deployment system of intelligent community software |
| CN112835683B (en) * | 2021-03-02 | 2023-12-12 | 杭州雅观科技有限公司 | Containerized deployment system of intelligent community software |
| CN115437889A (en) * | 2022-11-08 | 2022-12-06 | 统信软件技术有限公司 | Emergency processing method and system and computing equipment |
| CN115437889B (en) * | 2022-11-08 | 2023-03-10 | 统信软件技术有限公司 | Emergency processing method, system and computing equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110737501A (en) | Method and system for realizing functions of check point and recovery point in Docker container | |
| US11669409B2 (en) | Application migration between environments | |
| CA2810570C (en) | Private cloud replication and recovery | |
| US8219769B1 (en) | Discovering cluster resources to efficiently perform cluster backups and restores | |
| US10169173B2 (en) | Preserving management services with distributed metadata through the disaster recovery life cycle | |
| US8676762B2 (en) | Efficient backup and restore of a cluster aware virtual input/output server (VIOS) within a VIOS cluster | |
| US6795966B1 (en) | Mechanism for restoring, porting, replicating and checkpointing computer systems using state extraction | |
| US9323550B2 (en) | Mechanism for providing virtual machines for use by multiple users | |
| CN103970585B (en) | Create the method and device of virtual machine | |
| US8533164B2 (en) | Method and tool to overcome VIOS configuration validation and restoration failure due to DRC name mismatch | |
| US20190391880A1 (en) | Application backup and management | |
| US10936442B2 (en) | Simultaneous file level recovery from multiple backups using a proxy virtual machine | |
| US20150378839A1 (en) | Recovery system and method for performing site recovery using replicated recovery-specific metadata | |
| CN110088733A (en) | The layout based on accumulation layer of virtual machine (vm) migration | |
| US20140059375A1 (en) | Recovery system and method for recreating a state of a datacenter | |
| US20120246642A1 (en) | Management of File Images in a Virtual Environment | |
| US11880282B2 (en) | Container-based application data protection method and system | |
| CN102520881A (en) | Virtual machine snapshot management method and system of cloud computing platform | |
| KR102016095B1 (en) | System and method for persisting transaction records in a transactional middleware machine environment | |
| US11194673B2 (en) | System and method to enable file level recovery of virtual disks and dynamic disks | |
| US20220269414A1 (en) | Snapshotting a containerized application | |
| CN106612314A (en) | System for realizing software-defined storage based on virtual machine | |
| US11372731B2 (en) | System and method to enable file level recovery of virtual disks and dynamic disks | |
| CN110618884A (en) | Fault monitoring method, virtualized network function module manager and storage medium | |
| CN106371952A (en) | Emergency management system based on physical machine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200131 |
|
| WD01 | Invention patent application deemed withdrawn after publication |