Updating method and device of cluster global routing table, equipment and storage medium

Abstract

The invention discloses a cluster global routing table updating method, device and equipment and a storage medium, and relates to the technical field of distributed storage, the method is applied to client equipment, and the method comprises the following steps: when it is monitored that communication with target object storage equipment of a distributed storage system is abnormal, updating a cluster global routing table; sending a global routing table request to a monitor of the distributed storage system; detecting whether the current main object storage device is a target object storage device or not according to a global routing table returned by the monitor; if not, generating a resending request corresponding to the target sent request, and sending the resending request to the current main object storage device; when the client device monitors that the communication with the target OSD in the distributed storage system is abnormal, the client device actively subscribes to the latest global routing table from the MON and applies the latest global routing table to regenerate and send the request, the situation that the request is blocked for a long time is avoided, and the stability and reliability of the distributed storage system are improved.

Description

Technical Field

[0001] The present invention relates to the field of distributed storage technology, and in particular to a method, device, equipment and storage medium for updating a cluster global routing table. Background Art

[0002] In a distributed storage system, after an external client device connects to the storage system, objects are ultimately stored in the storage pool. The storage pool is divided into several placement groups (PGs) for object storage. During object storage, objects undergo a mapping process, ensuring that objects with the same feature values are mapped to the same PG. The PG then uses an algorithm to calculate a mapping, storing all objects in that PG on an object storage device (OSD). Distributed storage systems typically utilize replication or erasure redundancy mechanisms to ensure data security and reliability. Therefore, a PG may have multiple corresponding OSD members, which serve as backups for each other to ensure data security and reliability. The dependencies between PGs and OSDs are recorded in the global routing table (Object Storage Device Map, OSDMap). When a client device sends a request, it relies on the global routing table to resolve the primary OSD corresponding to the object, establish a connection to that OSD, and send the request.

[0003] Currently, in conventional applications, client devices typically subscribe to the cluster's global routing table from the distributed storage system's MON (Monitor) upon registration, which is then periodically pushed to client devices. Since client devices need to send requests from the failed OSD to the new master OSD when the master OSD fails, this mechanism of regularly pushing the global routing table prevents client devices from receiving the latest global routing table in a timely manner and from responding quickly to changes in the global routing table. When the master OSD fails, this inability to respond quickly causes client device requests to be blocked, impacting the stability and reliability of the distributed storage system. Therefore, how to enable client devices to promptly obtain and apply the latest global routing table, avoid requests being blocked for extended periods, and improve the stability and reliability of the distributed storage system, is an urgent problem that needs to be addressed. Summary of the Invention

[0004] The purpose of the present invention is to provide a cluster global routing table update method, device, electronic device and computer-readable storage medium, so that client devices can obtain the latest global routing table and apply it in a timely manner, avoid long-term request blocking, and improve the stability and reliability of the distributed storage system.

[0005] To solve the above technical problems, the present invention provides a method for updating a cluster global routing table, which is applied to a client device and includes:

[0006] When detecting that communication with a target object storage device of a distributed storage system is abnormal, sending a global routing table request to a monitor of the distributed storage system;

[0007] Detecting, according to the global routing table returned by the monitor, whether the current primary object storage device is the target object storage device;

[0008] If not, a resending request corresponding to the target sent request is generated, and the resending request is sent to the current primary object storage device; wherein the target sent request is a request sent to the target object storage device.

[0009] On the other hand, before sending the global routing table request to the monitor of the distributed storage system, the method further includes:

[0010] Detecting whether an IO request sent to the target object storage device has timed out;

[0011] If the timeout has occurred, it is determined that an abnormality has occurred in the communication between the client device and the target object storage device, and the step of sending a global routing table request to the monitor of the distributed storage system is performed.

[0012] On the other hand, detecting whether the IO request sent to the target object storage device has timed out includes:

[0013] Detecting, at a preset time interval, whether there is an IO request exceeding a preset timeout period among the IO requests sent to the target object storage device;

[0014] If there is an IO request that exceeds the preset timeout period, it is determined that an abnormality occurs in the communication between the client device and the target object storage device, and the step of sending a global routing table request to the monitor of the distributed storage system is executed.

[0015] On the other hand, detecting whether the IO request sent to the target object storage device has timed out includes:

[0016] Detecting whether the number of timed-out IO requests sent to the target object storage device reaches a number threshold;

[0017] If the number threshold is reached, it is determined that an abnormality occurs in the communication between the client device and the target object storage device, and the step of sending a global routing table request to the monitor of the distributed storage system is performed.

[0018] On the other hand, after detecting whether the current primary object storage device is the target object storage device, the method further includes:

[0019] If the current primary object storage device is the target object storage device, then this process ends.

[0020] On the other hand, before sending the global routing table request to the monitor of the distributed storage system, the method further includes:

[0021] Detecting whether a network connection with the target object storage device is reset;

[0022] If a reset occurs, it is determined that an abnormality occurs in the communication between the client device and the target object storage device, and the step of sending a global routing table request to the monitor of the distributed storage system is performed.

[0023] On the other hand, after detecting whether the current primary object storage device is the target object storage device, the method further includes:

[0024] If the current primary object storage device is the target object storage device, a network connection is re-established with the target object storage device, and a resend request corresponding to the target sent request is generated, and the resend request is sent to the target object storage device through the re-established network connection.

[0025] The present invention also provides a device for updating a cluster global routing table, which is applied to a client device and includes:

[0026] A routing request module, configured to send a global routing table request to a monitor of the distributed storage system when detecting an abnormality in communication with a target object storage device of the distributed storage system;

[0027] a change detection module, configured to detect whether the current primary object storage device is the target object storage device according to the global routing table returned by the monitor;

[0028] The resending module is used to generate a resending request corresponding to the target sent request if it is not the target object storage device, and send the resending request to the current primary object storage device; wherein, the target sent request is a request sent to the target object storage device.

[0029] The present invention also provides an electronic device, comprising:

[0030] memory for storing computer programs;

[0031] The processor is configured to implement the steps of the above-mentioned method for updating the cluster global routing table when executing the computer program.

[0032] In addition, the present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the computer program implements the steps of the above-mentioned method for updating the cluster global routing table.

[0033] A cluster global routing table updating method provided by the present invention is applied to a client device, comprising: when an abnormality is detected in the communication with the target object storage device of the distributed storage system, sending a global routing table request to the monitor of the distributed storage system; based on the global routing table returned by the monitor, detecting whether the current primary object storage device is the target object storage device; if not, generating a resend request corresponding to the target sent request, and sending the resend request to the current primary object storage device; wherein the target sent request is a request sent to the target object storage device.

[0034] As can be seen, in the present invention, when a client device detects an anomaly in communication with a target OSD in a distributed storage system, it proactively subscribes to the latest global routing table from the MON and applies it. This allows the client device to regenerate and resend requests when the target OSD changes, thus avoiding long-term request blockage and improving the stability and reliability of the distributed storage system. Furthermore, the present invention also provides a device for updating a cluster global routing table, an electronic device, and a computer-readable storage medium, all of which also have the aforementioned beneficial effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0036] Figure 1 A flowchart of a cluster global routing table updating method provided by an embodiment of the present invention;

[0037] Figure 2 A schematic flow chart of another method for updating a cluster global routing table provided by an embodiment of the present invention;

[0038] Figure 3 A schematic flow chart of another method for updating a cluster global routing table provided by an embodiment of the present invention;

[0039] Figure 4 A structural block diagram of a device for updating a cluster global routing table provided by an embodiment of the present invention;

[0040] Figure 5The present invention provides a schematic structural diagram of an electronic device. DETAILED DESCRIPTION

[0041] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0042] Please refer to Figure 1 , Figure 1 A flowchart of a cluster global routing table update method provided by an embodiment of the present invention. The method is applied to a client device and may include:

[0043] Step 101: When an abnormality is detected in the communication with the target object storage device of the distributed storage system, a global routing table request is sent to a monitor of the distributed storage system.

[0044] It is understandable that when the distributed storage system fails during the process of writing or reading data by the client device, some member OSDs of some PGs may go offline; generally, the master OSD of the PG receives the request of the client device, such as IO (input and output) request, and distributes it to the slave OSD members of the PG after analysis and processing, and then returns it to the client device after processing and aggregation; when the master OSD fails, the cluster will select a new master OSD for the PG to process the request of the client device based on the fault recovery mechanism; at this time, the client device is required to send the request sent to the original failed OSD (that is, the previous OSD) to the new master OSD to avoid long-term blocking of the request.

[0045] However, the mechanism used in related art whereby the distributed storage system's MON (monitor) regularly pushes the global routing table (OSDMap) to client devices prevents the client devices from receiving the latest global routing table in a timely manner and from responding quickly to changes in the global routing table. This results in a delay in responding quickly when the primary OSD fails, blocking client device requests and impacting the stability and reliability of the distributed storage system. Therefore, this embodiment utilizes the client device's monitoring of its communication with the distributed storage system's target OSD. Upon detecting a communication anomaly, the client device proactively sends a global routing table request to the distributed storage system's MON, allowing the client device to promptly receive and apply the latest global routing table.

[0046] Correspondingly, the target OSD in this embodiment can be the OSD of the distributed storage system that the client device is currently communicating with, that is, the main OSD to which the previous request was to be sent. The method provided in this embodiment may also include a process in which the client device monitors whether an abnormality occurs in the communication with the target OSD of the distributed storage system. The specific way in which the client device monitors whether an abnormality occurs in the communication with the target OSD of the distributed storage system can be set by the designer according to the practical scenario and user needs. For example, the client device can use the timeout detection of the IO request sent to the target OSD to determine whether an abnormality occurs in the communication with the target OSD; for example, the client device can detect whether the IO request sent to the target OSD has timed out; if it has timed out, it is determined that an abnormality occurs in the communication between the client device and the target OSD, and the process goes to step 101. The client device can also determine whether an abnormality occurs in the communication with the target OSD by detecting the network connection status with the target OSD. The client device can also simultaneously monitor whether an abnormality occurs in the communication with the target OSD through the above two methods, or use other methods to monitor whether an abnormality occurs in the communication with the target OSD. As long as the client device can monitor whether an abnormality occurs in the communication with the target SDD, it can detect the abnormality of the cluster-end device (such as a server) of the distributed storage system in a timely manner, thereby avoiding the situation where the client device request cannot perceive the cluster failure. This embodiment does not impose any restrictions on this.

[0047] Among them, this embodiment does not limit the specific method by which the above-mentioned client device uses the timeout detection of the IO request sent to the target OSD to determine whether the communication with the target OSD is abnormal. For example, the client device can detect whether there is an IO request exceeding the preset timeout period in each IO request sent to the target OSD at a preset time interval; if there is an IO request exceeding the preset timeout period, it is determined that the communication between the client device and the target object storage device is abnormal, and enter step 101; if there is no IO request exceeding the preset timeout period, the process can be ended or the process can be returned to the step of detecting whether there is an IO request exceeding the preset timeout period in each IO request sent to the target object storage device at a preset time interval to wait for the next detection. Figure 2 As shown, the client device uses a pre-set timer and a preset timeout period to regularly detect whether the IO request sent to the target OSD has timed out (i.e., exceeds the preset timeout period); if so, it can send an OSDMap request to the monitor of the distributed storage system to update the OSDMap stored in itself.

[0048] Accordingly, the client device may also detect whether the number of timed-out IO requests sent to the target OSD reaches a quantity threshold; if so, it is determined that a communication anomaly has occurred between the client device and the target object storage device, and the process proceeds to step 101. The client device may also detect whether the number of timed-out IO requests sent to the target OSD reaches a quantity threshold at a preset time interval; if so, it is determined that a communication anomaly has occurred between the client device and the target OSD, and the process proceeds to step 101.

[0049] Likewise, this embodiment does not limit the specific method by which the client device determines whether an abnormality occurs in the communication with the target OSD by detecting the network connection status with the target OSD. For example, the client device may detect whether the network connection with the target OSD is reset; if a reset occurs, it is determined that an abnormality occurs in the communication between the client device and the target object storage device, and the process proceeds to step 101. Figure 3 As shown, the client device can record the mapping table of the target OSD and the corresponding network connection; since when an abnormality occurs in the OSD of the cluster-end device (such as the server of the distributed storage system), the network connection between the client device and the OSD will be reset, the client device can send an OSDMap request to the monitor (MON) of the distributed storage system after detecting the reset of the network connection, so as to update the OSDMap stored in itself according to the latest OSDMap replied by the MON.

[0050] Step 102 : Based on the global routing table returned by the monitor, detect whether the current primary object storage device is the target object storage device; if not, proceed to step 103 .

[0051] It is understandable that the current master OSD in this step can be the current master OSD determined by using the OSDMap returned by the monitor. In this step, the client device can use the global routing table (OSDMap) returned by the MON to detect whether the current master OSD is the target OSD; Figure 2 and Figure 3 As shown, the client device can use the OSDMap returned by the MON to update the OSDMap stored in itself, apply the updated OSDMap, and determine whether the target OSD has changed, that is, to detect whether the current master OSD is the target OSD (that is, the previous master OSD). When the target OSD has changed, that is, the current master OSD is not the target OSD, the client device enters step 103 and resends the request to the current master OSD to avoid the situation where the request is blocked for a long time.

[0052] Correspondingly, this step may also include a process of determining the current master OSD using the global routing table returned by the monitor, such as by using the same or similar method as the method of using OSDMap to parse the master OSD corresponding to the object in the related art. This embodiment does not impose any restrictions on this.

[0053] Step 103: Generate a resend request corresponding to the target sent request, and send the resend request to the current primary object storage device; wherein the target sent request is a request sent to the target object storage device.

[0054] It can be understood that in this step, the client device can resend the request to the current main OSD (i.e., resend the request) when the current main OSD is not the target OSD, so that the current main OSD can process the request normally and in a timely manner and return the corresponding request processing result, thereby avoiding the situation where the request is blocked for a long time.

[0055] Accordingly, the target sent request in this embodiment may be a request sent to the previous primary OSD (ie, the target OSD), such as a request sent to the target OSD for which no request processing result has been received.

[0056] Correspondingly, in the case where the current main OSD is the target OSD in this embodiment, that is, the target OSD has not changed, the designer can set it according to the practical scenario and user needs, such as Figure 2 As shown, when the client device uses the timeout detection of the IO request sent to the target OSD to determine that the communication with the target OSD is abnormal, when it detects that the current main OSD is the target OSD, it can directly end this process and continue to wait for the target OSD to return the corresponding request processing result. Figure 3 As shown, when the client device determines that an abnormality has occurred in the communication with the target OSD by detecting the network connection status with the target OSD, the client device can re-establish the network connection with the target OSD and directly send the target sent request to the target OSD through the re-established network connection, or generate a resend request corresponding to the target sent request and send the resend request to the target object storage device through the re-established network connection.

[0057] In this embodiment, when the client device detects an abnormality in the communication with the target OSD in the distributed storage system, it actively subscribes to the latest global routing table from the MON and applies it to regenerate and send requests when the target OSD changes, thereby avoiding long-term blocking of requests and improving the stability and reliability of the distributed storage system.

[0058] Corresponding to the above method embodiment, an embodiment of the present invention further provides a device for updating a cluster global routing table. The device for updating a cluster global routing table described below and the method for updating a cluster global routing table described above can refer to each other.

[0059] Please refer to Figure 4 , Figure 4 This is a block diagram of a device for updating a cluster global routing table provided by an embodiment of the present invention. The device is applied to a client device and may include:

[0060] The routing request module 10 is configured to send a global routing table request to a monitor of the distributed storage system when detecting an abnormality in communication with a target object storage device of the distributed storage system;

[0061] The change detection module 20 is used to detect whether the current primary object storage device is the target object storage device according to the global routing table returned by the monitor;

[0062] The resending module 30 is used to generate a resending request corresponding to the target sent request if it is not the target object storage device, and send the resending request to the current primary object storage device; wherein the target sent request is a request sent to the target object storage device.

[0063] In some embodiments, the apparatus may further comprise:

[0064] The timeout detection module is used to detect whether the IO request sent to the target object storage device has timed out; if it has timed out, it determines that the communication between the client device and the target object storage device is abnormal, and sends a start signal to the routing request module 10.

[0065] In some embodiments, the timeout detection module can be specifically used to detect whether there is an IO request that exceeds the preset timeout period among the IO requests sent to the target object storage device at a preset time interval; if there is an IO request that exceeds the preset timeout period, it is determined that an abnormality has occurred in the communication between the client device and the target object storage device, and a start signal is sent to the routing request module 10.

[0066] In some embodiments, the timeout detection module can be specifically used to detect whether the number of timed-out IO requests sent to the target object storage device reaches a quantity threshold; if the quantity threshold is reached, it is determined that an abnormality has occurred in the communication between the client device and the target object storage device, and a start signal is sent to the routing request module 10.

[0067] In some embodiments, the change detection module 20 may also be configured to end the process when the current primary object storage device is the target object storage device.

[0068] In some embodiments, the apparatus may further comprise:

[0069] The reset detection module is used to detect whether the network connection with the target object storage device is reset; if a reset occurs, it is determined that the communication between the client device and the target object storage device is abnormal, and a start signal is sent to the routing request module 10.

[0070] In some embodiments, the apparatus may further comprise:

[0071] The reconnection sending module is used to re-establish the network connection with the target object storage device when the current primary object storage device is the target object storage device, and generate a resending request corresponding to the target sent request, and send the resending request to the target object storage device through the re-established network connection.

[0072] In this embodiment, when the client device detects an abnormality in the communication with the target OSD in the distributed storage system through the routing request module 10, it actively subscribes to the latest global routing table from the MON and applies it, so as to regenerate and send the request when the target OSD changes, thereby avoiding long-term blocking of the request and improving the stability and reliability of the distributed storage system.

[0073] Corresponding to the above method embodiment, an embodiment of the present invention further provides an electronic device. The electronic device described below and the cluster global routing table updating method described above can refer to each other.

[0074] Please refer to Figure 5 , Figure 5 This is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device may include:

[0075] Memory D1, for storing computer programs;

[0076] The processor D2 is configured to implement the steps of the cluster global routing table updating method provided in the above method embodiment when executing a computer program.

[0077] The electronic device provided in this embodiment is specifically a client device of the distributed storage system, that is, a device running client software of the distributed storage system, such as a computer device.

[0078] Corresponding to the above device embodiment, an embodiment of the present invention further provides a distributed storage system. The distributed storage system described below and the cluster global routing table updating method described above can refer to each other.

[0079] A distributed storage system includes a client device and a cluster device.

[0080] The client device may be the electronic device provided in the previous embodiment, and the cluster-end device may be a server of a distributed storage system.

[0081] Corresponding to the above method embodiment, an embodiment of the present invention further provides a computer program product. The computer program product described below and the cluster global routing table updating method described above can refer to each other.

[0082] A computer program product includes a computer program / instruction, which, when executed by a processor, implements the steps of the method for updating a cluster global routing table provided in the above method embodiment.

[0083] Corresponding to the above method embodiment, an embodiment of the present invention further provides a computer-readable storage medium. The computer-readable storage medium described below and the cluster global routing table updating method described above can refer to each other.

[0084] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the method for updating a cluster global routing table of the above-mentioned method embodiment.

[0085] The computer-readable storage medium may be a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, which may store program codes.

[0086] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. References to the common and similar parts between the various embodiments are sufficient. The devices, electronic devices, systems, computer program products, and computer-readable storage media disclosed in the embodiments are described briefly because they correspond to the methods disclosed in the embodiments. For relevant details, refer to the description of the methods.

[0087] The above is a detailed introduction to the cluster global routing table update method, device, electronic device and computer-readable storage medium provided by the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core ideas. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present invention, several improvements and modifications can be made to the present invention, and these improvements and modifications also fall within the scope of protection of the present invention.

Claims

1. A method for updating a cluster global routing table, characterized in that: Applied to client devices, including: When detecting that communication with a target object storage device of a distributed storage system is abnormal, sending a global routing table request to a monitor of the distributed storage system; Detecting, according to the global routing table returned by the monitor, whether the current primary object storage device is the target object storage device; If not, a resending request corresponding to the target sent request is generated, and the resending request is sent to the current primary object storage device; wherein the target sent request is a request sent to the target object storage device.

2. The method for updating the cluster global routing table according to claim 1, characterized in that: Before sending the global routing table request to the monitor of the distributed storage system, the method further includes: Detecting whether an IO request sent to the target object storage device has timed out; If the timeout has occurred, it is determined that an abnormality has occurred in the communication between the client device and the target object storage device, and the step of sending a global routing table request to the monitor of the distributed storage system is performed.

3. The method for updating the cluster global routing table according to claim 2, characterized in that: The detecting whether the IO request sent to the target object storage device has timed out includes: Detecting, at a preset time interval, whether there is an IO request exceeding a preset timeout period among the IO requests sent to the target object storage device; If there is an IO request that exceeds the preset timeout period, it is determined that an abnormality occurs in the communication between the client device and the target object storage device, and the step of sending a global routing table request to the monitor of the distributed storage system is executed.

4. The method for updating the cluster global routing table according to claim 2, characterized in that: The detecting whether the IO request sent to the target object storage device has timed out includes: Detecting whether the number of timed-out IO requests sent to the target object storage device reaches a number threshold; If the number threshold is reached, it is determined that an abnormality occurs in the communication between the client device and the target object storage device, and the step of sending a global routing table request to the monitor of the distributed storage system is performed.

5. The method for updating the cluster global routing table according to claim 2, characterized in that: After detecting whether the current primary object storage device is the target object storage device, the method further includes: If the current primary object storage device is the target object storage device, then this process ends.

6. The method for updating a cluster global routing table according to any one of claims 1 to 5, characterized in that: Before sending the global routing table request to the monitor of the distributed storage system, the method further includes: Detecting whether a network connection with the target object storage device is reset; If a reset occurs, it is determined that an abnormality occurs in the communication between the client device and the target object storage device, and the step of sending a global routing table request to the monitor of the distributed storage system is performed.

7. The method for updating the cluster global routing table according to claim 5, characterized in that: After detecting whether the current primary object storage device is the target object storage device, the method further includes: If the current primary object storage device is the target object storage device, a network connection is re-established with the target object storage device, and a resend request corresponding to the target sent request is generated, and the resend request is sent to the target object storage device through the re-established network connection.

8. A device for updating a cluster global routing table, characterized in that: Applied to client devices, including: A routing request module, configured to send a global routing table request to a monitor of the distributed storage system when detecting an abnormality in communication with a target object storage device of the distributed storage system; a change detection module, configured to detect whether the current primary object storage device is the target object storage device according to the global routing table returned by the monitor; The resending module is used to generate a resending request corresponding to the target sent request if it is not the target object storage device, and send the resending request to the current primary object storage device; wherein, the target sent request is a request sent to the target object storage device.

9. An electronic device, characterized in that: include: memory for storing computer programs; A processor, configured to implement the steps of the cluster global routing table updating method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for updating the cluster global routing table according to any one of claims 1 to 7 are implemented.

Images