62results about How to "Quick rollback" patented technology

The invention provides a continuous data protection method based on continuous snapshots and a system. A disk or a copy of a logic storage unit can be acquired by snapshots, and a plurality of copiesof disk can be continuously acquired by snapshots so as to realize continuous protection on data. The storage unit can be restored to the previous state by rollback snapshots, or a new logic disk canbe generated by cloning snapshots for user mounting. By continuously storing the state of a storage tank through automatic snapshot technology, continuous protection on data can be realized; when an accident happens, normal system and data state can be rolled back rapidly, and data error caused by soft disasters such as incorrect manipulation, virus attack and the like can be avoided by selectingvalid historical state point data.

The invention belongs to the technical field of distributed computation and discloses a distributed transaction processing method and a device thereof. The distributed transaction processing method includes the steps of application development, transaction management attribute definition, event definition, streaming calculation and transaction management. To be specific, the method includes the steps of acquiring and identifying event information in real time through the streaming calculation, acquiring transaction management attributes in real time, periodically verifying transaction status based on the event information and the transaction management attributes, initiating a verification request and / or a reversal request based on the transaction status, and sending transaction status verification instructions and / or reversal instructions through a transaction manager. With the above-mentioned scheme, the distributed transaction processing method and the device thereof have the advantages of allowing transaction control and the application development to be completely decoupled, not requiring development transaction control logic for the application development, greatly simplifying development difficulty, realizing fast rollback of transactions under abnormal scenarios, tolerating failure of any node and having no brain splitting problem.

The invention relates to a control method for software upgrading and failure rollback of hydrogen energy automobile controller, and belongs to the technical field of automobiles. The control method comprises the steps that upgrading push updating of automobile controller software is obtained, the upgrading software is downloaded and verified, whole automobile controller partition information and current version software information are obtained and sent to a remote server, upgrading flashing service is started, and the whole automobile controller responds to the flashing service to upgrade thesoftware; and the vehicle control unit finishes upgrading and feeds back an upgrading result, and if the Internet of Vehicles terminal receives upgrading failure information, a next upgrading failurerollback mechanism is entered: the vehicle control unit recovers the vehicle control unit software to version software before upgrading through a vehicle flashing technology. According to the controlmethod, the phenomenon that the automobile cannot be started due to failure of the automobile controller software in the upgrading process can be effectively avoided. The automobile using experienceof a user is improved, and both the cost and the time are saved.

The invention discloses a computer data backup and recovery method, belongs to the field of information safety and can be used for realizing the fact that a computer system can be recovered to the state of a happened random time point from a current state. The method is characterized in that back-up partitions are established on a computer hard disk, data manipulation events on sectors of conventional partitions of the hard disk are intercepted by a monitoring module, if the data manipulation events are a file system, which is capable of changing the conventional partitions of the hard disk, on the sectors, a hard disk operating log is established, and the hard disk operating log is stored in back-up partitions; and the time point corresponding to the current state of the computer is b, the computer returns to the time point a, a first recovery module takes the hard disk operating logs corresponding to all time points between the time point b and the time point a, and all hard disk operating logs taken by the first recovery module are overlaid in sequence according to the chronological order from the time point b to the time point a and are covered on a current hard disk .

A flashback method of a distributed object-based storage system is characterized by comprising the following steps: performing writing and reading of data of the distributed object-based storage system, data deleting data flashing bac, and deleting of outdated version data; constructing three data storing units, namely, a primary database, a back database and a servering database, based on a storage engine of a key-value data model; and performing data writing and reading, data deleting, data flashing back and deleting of outdated version data based on the three data storage units when a storage system receives a data operation request. With the adoption of the flashback method, the system complexity, increased by the flashback property, of the data can be greatly reduced, and a user can avoid the problems of additional data mirror server, serializing and deserializing of mirror files; a record-level previous version recovery mode is treated as the flashback mechanism, so that the data flashback process is only performed on a specified data set, and as a result, mis-operation can be avoided.

The invention discloses a method for implementing the batch configuration pre-activation and a device. The method of the invention comprises the processes that: a backup database is updated according to the static snapshot of a system configuration data file so as to result in that the data of the backup database is consistent with the data of a main database; the currently operated database is switched to the backup database and the configuration command in a configuration script when the current operation is executed is resolved; the configuration command after being resolved is executed and the backup database is updated according to the data in the configuration command, and the record for updating the backup database is submitted to the backup database. The invention avoids the successive inconsistency of the status of the configuration data of the network element after being pre-activated and the phenomenon that the configuration data of the network element is inconsistent with the configuration data of the network management system after the network element is pre-activated, and shortens the rollback time of the configuration data under the circumstance of having no effect on the services.

A method for managing a WeChat applet third-party platform comprises the steps of keeping a public basic service system of a WeChat system independent, and the basic service system achieves an encryption and decryption protocol for packaging a WeChat interface; obtaining, refreshing and maintaining Token of the WeChat developer account and the WeChat applet account; packaging a functional interface of the WeChat; the public basic service system is provided by the following programs: one account is supported to manage a plurality of WeChat developer accounts and a plurality of WeChat applet accounts by grouping and segmenting data; wherein fifty WeChat applet accounts can be bound under each WeChat developer account; wherein the WeChat applet accounts are minimum units of data segmentation,and the information of each WeChat applet account must be consistent with the account information of the WeChat developer to which the WeChat applet account belongs and the account information of thesystem user. According to the method, the complexity of applying for and managing the applet account is greatly reduced, and qualification authentication does not need to be carried out on a WeChat party. When a large number of similar small programs are needed, the code template can be quickly reused.

The invention discloses a roll-back method, a device and a system of a non relational database. The method comprises: judging whether operations can not be carried out to a non relational database, if the operations can not be carried out to the non relational database, replacing the original data of the non relational database with the mirror data of a mirror database, and the mirror data are the backup copy of the data of the non relational database before the current time period, modifying the mirror data of the non relational database according to the correct log content before a first time which enables the operations not be carried out to the non relational database in the current time period in a log database. By adopting the roll-back method, the device and the system of the non relational database, the mirror data of the mirror database is enabled to replace the original data in the non relational database, then modification can be carried out on the correct mirror data according to the content of the correct log, and therefore the roll-back function of the non relational database is achieved.

The invention discloses an access method of user equipment (UE). The access method comprises the following steps: determining the initial value of a rollback timer; determining the number of resource blocks (RBs) to be occupied by the UE; determining the rollback step length of the UE on a time domain according to the number of the RBs to be occupied by the UE, monitoring the number of idle RBs on a current channel, decreasing the numerical value of the rollback timer by one rollback step length when the monitored number of the idle RBs is greater than or equal to the number of the RBs to be occupied by the UE, and repeatedly executing the step till the rollback timer is reset; and transmitting data. Corresponding to the method, the invention further discloses an internal structure of the UE.

The invention discloses a method of rapidly rolling back a list. The method of rapidly rolling back the list comprises the following steps: judging whether a sliding operation enters an area corresponding to a list interface; judging the sliding direction of the sliding operation is upward or downward; and rolling back to the top row or the terminal row according to the sliding direction. Compared with the prior art, the method of rapidly rolling back the list judges whether the sliding operation enters the area corresponding to the list interface at first, if the judged result is yes, the sliding direction of the sliding operation is judged to be upward or downward, when the sliding direction is upward, the top row of the list interface is rolled back to, when the sliding direction is downward, the terminal row is rolled back to, and therefore rapid rolling-back of the list is achieved. Besides, tools like a sliding block are not needed to occupy the list interface, and therefore the visual effect of the entire list interface is not affected. The invention further discloses a terminal of rapidly rolling back the list.

The invention discloses a method for sharing models between scheduling automatic systems. The method is characterized by comprising the following steps of: 1) reading model files of a local system, and establishing topological graphs of the local system; 2) reading the model files of an opposite-end system, and establishing the topological graphs of the opposite-end system; 3) selecting part of the topological graphs in which a central station of the local system serves as a central node and stations in direct connection with the central station through alternating current lines serve as other nodes; 4) selecting part of the topological graphs in which the central station of the opposite-end system serves as the central node and the stations in direct connection with the central station through the alternating current lines serve as other nodes; 5) replacing the part of the topological graphs which are selected in the step 3) with the part of the topological graphs which are selected in the step 4) or combining the part of the topological graphs which are selected in the step 3) and the part of the topological graphs which are selected in the step 4); and 6) synchronously receiving the real-time data and alarm data of each node in the part of the topological graphs of the opposite-end system. The method has the advantages of simplicity in operation, convenience for use and the like.

The invention discloses a control system and method of a car window. The system comprises a car window motor drive circuit, a key sampling circuit, a ripple sampling circuit, a power circuit and a carwindow ripple anti-pinch controller; the car window ripple anti-pinch controller comprises a processor, an operational circuit used for collecting motor lifting current value and a detection circuitused for detecting the number of motor current ripples, the ripple sampling circuit comprises a sampling resistor arranged at the powering-on end of a car window motor, the car window ripple anti-pinch controller is connected with the key sampling circuit, the ripple sampling circuit and the output end of the power circuit. Compared with an existing car window control system, the system has the beneficial effects that the automation degree and the user-friendly degree are higher, the anti-pinch performance is more stable and reliable, and the effect can be effectively improved.

The invention provides a version automatic unifying method and system. The version automatic unifying system is used for unifying versions of one service module in a plurality of user environments. Each user environment includes a plurality of terminal clusters, at least one version control server and at least one domain name server; one service module is corresponding to at least one terminal cluster in each user environment, and the domain names of the terminal cluster corresponding to the service module in the plurality of user environments are identical. The versions of the service module are unified in the plurality of user environments through analysis of domain names by means of at least one domain name server in each user environment, and then versions in each user environment can be easily and quickly unified; and the efficiency of version release can be improved, and quick fallback of versions can be achieved.

The invention relates to an Agent deployment upgrading method and system based on Git version control. The method comprises the steps of deploying an Agent configuration management server-side component and a Git warehouse to a target server and respectively starting; deploying an Agent Daemon service component for the target server; updating an Agent deployment packet to the Git warehouse and configuring an Agent deployment strategy; and synchronizing the Agent deployment strategy for the Agent Daemon service component, and executing a command operation corresponding to the Agent deployment strategy. According to the Agent deployment upgrading method and system, the deployment configuration difficulty is simplified, and the roll-back of a version is implemented; moreover, the deployment packet is separated from a configuration file, and a Git version mechanism is utilized, so a simple and easy configuration upgrade management mechanism is provided, and the introduction of a complex configuration management system is avoided.

The invention discloses an extensible efficient APP channel package packing platform. The extensible efficient APP channel package packing platform comprises a Web management backstage used for writing a packaged channel into a public task queue, setting storage space and a packing way of the APP and checking the packing schedule and a packing result; a public task queue used for storing a to-be-packaged channel list, providing a channel capturing interface for multiple distributed packing services to compete to consume the packed list; a distributed packing service module used for monitoringthe public task queue and competing to consume the public task queue and enabling multiple packing threads for parallel packing based on the packing task; a distributed storage module used for providing extensible distributed storage for a channel and storing the channel packages in various versions; a downloading module used for providing a resource output service for each APP; and a CDN storagemodule used for caching the channel package and providing an installation package downloading resource for a user. As the distributed packing service technology is adopted, each service can be packedsimultaneously by multiple threads, the packing efficiency is improved efficiently, and the storage performance is extended.

The invention relates to the technical field of finance technology (Finch), and discloses a gray release management method, which comprises the following steps: determining a gray verification example corresponding to gray release and a target APPID corresponding to the gray verification example when gray release is carried out; when a service request is received, obtaining an input parameter APPID corresponding to the service request; judging whether the service request is a grayscale verification request or not according to the target APPID and the input parameter APPID; and if the service request is a gray scale verification request, sending the service request to a gray scale instance machine to verify the service request through the gray scale instance machine. The invention further discloses a gray release management device, equipment and a computer readable storage medium. According to the invention, the grayscale verification request is verified by setting the grayscale instance machine, and the grayscale verification request is separated from a conventional service request, so that grayscale verification is more convenient and comprehensive.

The invention provides a business data emergency switching system and a processing method. The method includes the a step of establishing data connection with a production system, a step of allocating order numbers to business data according to abnormal signals of the production system and storing the order numbers to a table t1, a step of assembling the business data into message strings according to the abnormal signals of the production system, associating the message strings with the order numbers in the table t1 and storing the associated message strings to a table t2, and a step of scanning order numbers in the table t1 one by one according to the recovery signals of the production system and sending the corresponding message strings in the table t2 to the production system. The system and method support two systems including a production system and an emergency system, the two systems can be switched at any time without affecting business processing. The systems are transferable and configurable. Data in the emergency switching system can be timely and accurately switched back to the production system, and seamless connection and smooth switching between the emergency switching system and the production system are realized.

The embodiment of the invention provides a version switching method. The version switching method comprises the steps of receiving a feedback instruction for indicating abnormal operation of an application program of a first version; according to the feedback instruction, generating a first prompt message for indicating that the application program of the first version runs abnormally; receiving afirst switching instruction generated by the mobile terminal according to the first prompt message, the first switching instruction carrying a configuration file of a second version and an identifiercorresponding to the first type of port, and generating the application program of the first version based on optimization of the application program of the second version; and according to the identifier carried by the first switching instruction, identifying the first type of port, so as to call the configuration file of the second version through the first type of port, thereby enabling the mobile terminal to run the application program of the second version according to the configuration file of the second version. The invention further provides a version switching device, computer equipment and a computer storable medium, one-key switching and rapid rollback can be achieved, the influence on a user is reduced, and the version problem processing process is simplified.

The embodiment of the invention provides a service rollback method and device under a micro-service architecture, and belongs to the technical field of computers. The method comprises: under the condition that a first micro-service fault is detected, judging whether a second micro-service coupled with the first micro-service with the fault exists or not, wherein the first micro-service and / or thesecond micro-service comprise one or more examples; determining a first rollback time point in the case of having a second micro-service coupled to the failed first micro-service; determining available versions of each instance of the first micro-service and each instance of the second micro-service before the first rollback time point; and respectively rolling back the instances of the first micro-service and the second micro-service to available versions before the first rolling-back time point. Rapid rollback based on a time point is realized, so that the influence of a micro-service faulton the operation of a system for calling the micro-service is rapidly stopped.

The invention discloses a rapid cutover method based on a circuit configuration pair. The method comprising the steps of building a cutover task, building cutover pairs in the cutover task, assigning a circuit before cutover for each cutover pair, and building a corresponding circuit after cutover; forming network elements of the cutover pairs into the circuit configuration pair one by one, and setting the attribute of the circuit configuration pair, wherein the circuit configuration pair comprises configuration before cutover and configuration after cutover corresponding to the circuits before and after cutover; configuring each network element in the circuit configuration pair to a network element configuration library, performing incremental isolation on the configuration before cutover and the new configuration after cutover according to the configuration attribute in the circuit configuration pair by using the network element; executing a cutover task, and starting the circuit configuration pair so as to make the configuration before cutover be invalid, and bring the configuration after cutover into effect; after the cutover is successful, deleting the configuration before cutover in the circuit configuration pair. According to the method provided by the invention, service life of the original circuit is improved, switching time of the circuit is reduced, and uninterrupted cutover of the circuit is achieved, and a cutover risk is reduced,

The invention discloses an electronic evidence storage system and method based on containerization rapid deployment, and the method comprises the steps: generating a Tars common frame mirror image file and a Tars service node mirror image file in response to the compiling of a Tars common frame mirror image on a Tars common frame server and a Tars service node mirror image on a Tars node server by a first server; enabling the first server to upload the Tars common frame mirror image file and the Tars service node mirror image file to a mirror image warehouse; enabling the second server to download and run the Tars public frame mirror image file and / or the Tars service node mirror image file from the mirror image warehouse; and enabling the second server to execute the script related to the business service on the Tars node server to complete the deployment of the business service. According to the electronic evidence storage system and method based on containerization rapid deployment provided by the invention, the Tars public framework and the node service are containerized, the version consistency of each dependent component under the Tars framework is solved, and rapid and batch deployment of business services can be realized based on the script language.

The invention provides a gray-scale publishing method and a device. Firstly, a server receives a client's acquisition request for a designated new script and generates a scheduling file correspondingto the designated new script. Then the client receives the scheduling file sent by the server, and judges whether the gray range value contained in the scheduling file is a preset value; if the gray range value is not a preset value, the client cancels sending the interval time of the new script acquisition request, and generates a random value according to the preset algorithm in the scheduling file. If the random value falls within the gray range value, the client obtains the new script content from the acquisition address of the designated new script. Under the premise that the client pageuses the original configuration information, the application enables different client pages to load a new script file according to a specified gray scale in actual rendering, thereby realizing the function of gray scale publishing of the new script.

The invention relates to a default gray release method and device based on a Kubernetes platform and a load balancing component. The default gray release method comprises gray release configuration parameter information. In the implementation process, gray release configuration parameter information is issued to a micro-service gateway, when a user request passes through the gateway, the gateway carries out initial gray marking on the user request in combination with the gray release configuration parameter information, and the control right of flow switching and execution in the gray release process is placed in the micro-service gateway. According to the method, a service grid technology is not used, an online operation service and a gray scale test service are isolated by using Namespace, service communication in the two Namespaces is broken through at the same time, the limitation that the resource utilization rate is only 50% in a traditional blue-green publishing scheme is avoided, the utilization rate of hardware resources is maximized, the traffic borne by different versions of services can be switched by adjusting the gray release parameters, and fast rollback is realized when the gray test fails.

The invention discloses a real-time road condition publishing method and device based on a container arrangement technology, and the method comprises the following steps: A, a system construction and publishing process: submitting a local code and a file to Gitlab through a git command, monitoring the code change through a webhook plug-in, automatically triggering jenkins for construction, enabling the jenkins to make the code into a docker mirror image, and enabling the push to a harbor mirror image warehouse; B, executing rc and service creation on the k8s-master, then creating Pod, and pulling a mirror image from a harbor mirror image warehouse; C, reading the yaml configuration file of the mirror image in k8s. According to the method, automatic construction and rapid release are achieved, rapid and automatic construction, deployment and release are achieved, a preset arrangement template is used for arranging multiple modules, automatic management and a one-key rollback mechanism of the multiple modules are achieved, the situation that abnormity occurs in the updating process and rapid rollback occurs is achieved, the continuous optimization and updating requirements of a road condition release system are met, the system version is updated more rapidly, and the manual publishing cost is reduced.

The present application provides an automated publishing method. The method comprises: receiving a publishing package; parsing content in the publishing package, and determining a type of the content in the publishing package; and executing a corresponding publishing operation according to the type of the content in the publishing package. According to the method provided by the present application, in terms of the publishing process, the existing publishing steps are simplified, and the publishing process is unified; in terms of the publishing time, the average time of each publishing time is shortened to be less than 15 minutes, and especially in the case of complex publishing content and a large amount of data, the effects are significant; and in addition, the publishing errors can be reduced to zero, and the publishing quality is improved.

The invention discloses a virtual cluster rollback method and a virtual cluster rollback system based on a multicast technology. The method comprises the following steps: (1) a storage cluster identifies virtual machine pages with the same data from a snapshot file, determines a virtual machine page needing multicast or unicast according to the corresponding relationship between virtual machines and physical machines, and packages the virtual machine page into a data packet; and (2) the storage cluster judges whether a current data packet to be sent is a multicast data packet; if the current data packet to be sent is a multicast data packet, the storage cluster notifies a physical machine to which the multicast data packet is to be sent to join a multicast group and then sends the multicast data packet to the multicast group, and the physical machine analyzes the received multicast data packet and recovers the virtual machine page in the multicast data packet to a corresponding address; and if the current data packet to be sent is a unicast data packet, the data packet is sent to a corresponding physical machine in a unicast mode, and the physical machine recovers the virtual machine page in the unicast data packet to a corresponding address. Through the method and the system, a virtual cluster can be rolled back quickly, and the amount of data transmitted is reduced.

The invention relates to a background data processing method and system. The method comprises the following steps of: storing a piece of to-be-effective state data to a secondary node; when a data effective instruction aiming at the to-be-effective state data is received, storing the to-be-effective state data stored in the secondary node into a primary node and configuring the state of the data stored in the primary node to be an effective state. According to the method and system, the background data setting correctness is improved and the aim of safely operating background data is realized.

The invention discloses a ball screw capable of achieving rapid separation and reunion. The ball screw is connected with a clutch and comprises a screw shaft and a sliding component capable of doing rectilinear back and forth movement on the screw shaft. The outer ring of the screw shaft is sleeved with the sliding component. The outer circumference of the screw shaft is of a thread structure. The sliding component comprises a contact part, a nut, a cylinder and a fixing part, wherein the outer ring of the screw shaft is sleeved with the contact part, the outer ring of the contact part is sleeved with the nut, the outer ring of the nut is sleeved with the cylinder, and the end of the nut is sleeved with the fixing part. The end, close to the fixing part, of the contact part is of a convex structure and connected with the fixing part through a reset spring. The outer ring of the other end of the contact part is provided with a groove a, a supporting part is clamped to the groove a, the outer ring of the contact part is provided with multiple rows of ball hole sets, each ball hole set is composed of multiple ball holes, balls are arranged in the ball holes, the cylinder is arranged between the supporting part and the fixing part, and the fixing part is limited and clamped to one end of the nut. The ball screw has the advantages that the rotation transmission way of the screw shaft and the nut is changed, and rapid separation and reunion and rapid retreating can be achieved.