Method and apparatus for implementing single node power-off in a multi-node server

Abstract

The application provides a kind of multi-node server implementation single node power-off method and device, belong to server power control technical field, the method steps are as follows: S1. after receiving the power-off request of a certain single node in middle backboard, notify each node in multi-node server;S2. after receiving the power-off information of single node that service related node returns in middle backboard, allow the single node to power off from multi-node server alone.This application provides a kind of multi-node server implementation single node power-off method and device, through middle backboard broadcast power-off notice, request power-off node and each node of multi-node server complete business handover, disconnect own power supply, realize the quick response of single node power-off request, meet the demand of single node power-off of user under the condition of not affecting the whole machine service of multi-node server.

Description

Technical Field

[0001] This invention belongs to the field of server power control technology, specifically relating to a method and apparatus for powering down a single node in a multi-node server. Background Technology

[0002] Currently, the power-on process of the server system is handled by the motherboard, which distributes and converts power. The PCIe daughter cards and backplanes draw power from the motherboard via cables or connectors. Once the corresponding power lines are established on the motherboard, each daughter card will complete the power-on process simultaneously.

[0003] Since each server operating as a single-node server performs its own operations and services, the current server power control strategy is sufficient to meet the power supply requirements of a single-node server. However, for multi-node servers, by configuring a backplane, any node requiring a power outage must send a power-off / reboot command to the backplane. Upon receiving the command, the backplane determines its priority. If the command has a high priority and no other nodes are consuming resources, the entire multi-node server can be powered down. If other nodes are consuming resources, and those nodes have a high priority, then the entire server cannot be powered down, thus failing to meet the needs of the node requiring a power outage. Powering down the entire server to meet the needs of the node requiring a power outage would cause other nodes' services to be suspended or interrupted.

[0004] In summary, the current single-node service requires a power outage on the entire server to be shut down, which may conflict with the business operations of other nodes.

[0005] To address the aforementioned shortcomings, it is essential to provide a method and apparatus for implementing single-node power failure in a multi-node server. Summary of the Invention

[0006] To address the shortcomings of current single-node services where a power outage of a single node requires a power outage of the entire server, which can conflict with the services of other nodes, this invention provides a method and apparatus for implementing single-node power outages in multi-node servers, thereby solving the aforementioned technical problems.

[0007] In a first aspect, the present invention provides a method for implementing single-node power failure in a multi-node server, comprising the following steps: S1. After receiving a power failure request from a single node, the backplane notifies all nodes in the multi-node server. S2. After receiving permission to power off from the relevant nodes of a single node, the backplane allows that single node to power off independently from the multi-node server. The single node notifies all nodes in the multi-node server of its power-off restart request through the backplane, and after receiving permission from each node, it powers off its own node independently to prevent impact on the services of other nodes.

[0008] Furthermore, the specific steps of step S1 are as follows: S11. The user sends a power-off restart command to a single node in a multi-node server via out-of-band or in-band methods. S12. Upon receiving a power-off restart command, the single node sends a power-off request to the backplane. S13. After receiving a power-off request, the backplane sends a power-off notification to each node in the multi-node server. A single node can receive in-band power-off restart commands, as well as power-off restart commands sent back from out-of-band by the BMC. The backplane broadcasts the power-off notification.

[0009] Furthermore, the specific steps of step S12 are as follows: S121. A single node that receives a power failure restart command obtains its own node number; S122. Upon receiving a power-off restart command, a single node encapsulates its own node number into a power-off request and sends it to the backplane. When sending the power-off request, the single node also sends its own node number.

[0010] Furthermore, the specific steps of step S13 are as follows: S131. After receiving a power-off request, the backplane in the middle parses out the node number of the single node requesting the power-off. S132. The backplane generates a power outage notification based on the node number of the single node requesting the power outage and sends it to other nodes in the multi-node server. The backplane sends the node number of the single node requesting the power outage and the power outage notification together to notify other nodes in the server which single node requested the power outage.

[0011] Furthermore, the specific steps of step S2 are as follows: S21. After receiving the power outage notification sent by the backplane, each node in the multi-node server determines whether the single node requesting the power outage has any business dealings with its own node. If so, proceed to step S22; If not, return the agreed power-off information to the backplane and proceed to step S23; S22. Nodes that have business dealings with the single node requesting power outage initiate business handover, and after the business handover is completed, return the power outage consent information to the backplane. S23. The backplane allows a single node that requests a power outage to perform a power outage restart; S24. The single node requesting a power outage initiates a separate power outage within the multi-node server based on its own power supply mode. Nodes within the multi-node server that have business relationships with the single node requesting the power outage need to initiate business handover to avoid business disruptions during the power outage of the single node, ensuring that the single node remains independent within the server node and has no business connections with other nodes, thus preparing for subsequent separate power outages.

[0012] Furthermore, the specific steps of step S21 are as follows: S211. After receiving the power failure notification sent by the backplane, each node of the multi-node server parses out the node number of the single node requesting the power failure. S212. Within a multi-node server, each node determines, based on its node number, whether the single node requesting a power outage has any business dealings with itself. If so, proceed to step S22; If not, encapsulate its own node number into the agreed power failure information and return it to the backplane, proceeding to step S23. Business communication between nodes of a multi-node server requires data transmission via node numbers to ensure transmission accuracy.

[0013] Furthermore, the specific steps of step S22 are as follows: S221. Initiate service handover with nodes that have business dealings with the single node requesting power outage, and suspend or transfer the services of that single node; S222. After the service handover is completed, the relevant nodes will encapsulate their own node number in the power outage consent information and return it to the backplane. After the service handover is completed, the power outage consent information needs to be replied to in a timely manner so that the single node can promptly initiate the power outage process.

[0014] Furthermore, the specific steps of step S23 are as follows: S231. The backplane receives the agreed power-off information returned by each node in the multi-node server and parses out the node number; S232. The backplane determines whether the number of nodes returning the agreed power outage information is equal to the number of nodes in the multi-node server based on the node number; If so, proceed to step S235; If not, proceed to step S233; S233. The backplane positioning node did not return the power outage information and sent the power outage notification to the positioning node again; S234. The backplane determines whether the number of times it sends a power failure notification to the positioning node exceeds a set threshold. If so, issue a timeout alarm and proceed to step S235; If not, return to step S21; S235. The backplane sends a power-off permission command to the single node requesting the power-off. The backplane confirms the number of nodes that have replied agreeing to the power-off, and determines whether all nodes other than the single node itself have agreed to the power-off. If no node has replied, the backplane locates that node and resends the power-off notification. This ensures that the node waits if the service handover time is too long, and also avoids failure to send the power-off notification to that node. The limit on the number of power-off notifications is to exclude the situation where the power-off notification link failure causes the failure to return the power-off permission information in a timely manner, ensuring that the single node's power-off request can be executed.

[0015] Furthermore, the specific steps of step S24 are as follows: S241. A single node requesting a power outage obtains its own power supply mode; When the self-powered mode is external power supply mode, proceed to step S242; When its own power supply mode is the adjacent node power supply mode, proceed to step S243; S242. The single node requesting a power outage sends a power cut-off command to cut off the motherboard's power supply from the external power source and sends a command to prohibit the motherboard from drawing power from neighboring nodes, thereby achieving a power outage restart for the single node and ending the process. S243. The single node requesting a power outage sends a power cut-off command, cutting off the motherboard's power supply from the neighboring node's motherboard and prohibiting it from drawing power from the neighboring node's motherboard, thus achieving a power-off restart for that single node. In a multi-node server, power is supplied either by each node's motherboard drawing power from an external power source, or by the motherboards of other nodes within the server. Regardless of the power supply method, the motherboard power supply must be disconnected. Due to the power supply stability requirements of a multi-node server, if one motherboard's power supply fails, it will autonomously draw power from other nodes to achieve primary / backup power supply for the node, while prohibiting power supply to the motherboards of other nodes. This ensures that a single node's power-off restart can be truly achieved and prevents situations where one power supply is cut off and power is drawn from other node motherboards again.

[0016] Secondly, the present invention provides a device for implementing single-node power failure in a multi-node server, comprising: The power outage notification module is used to notify all nodes in the multi-node server after receiving a power outage request from a single node. The negotiated individual power-off module is used to allow a single node to independently power off from the multi-node server after receiving consent to power off from the business-related nodes of that single node.

[0017] Furthermore, the power outage notification module includes: The power-off restart command sending unit is used to send a power-off restart command to a single node in a multi-node server via out-of-band or in-band methods. The power failure request issuing unit is used to send a power failure request to the backplane from a single node that receives a power failure restart command. The power outage notification sending unit is used to send power outage notifications to each node in the multi-node server after receiving a power outage request on the backplane.

[0018] Furthermore, the power outage request issuing unit includes: The self-node number acquisition sub-unit is used for a single node to obtain its own node number upon receiving a power failure restart command. The node number encapsulation and power failure request sending subunit is used for a single node that receives a power failure restart command to encapsulate its own node number into a power failure request and send it to the backplane.

[0019] Furthermore, the power outage notification sending unit includes: The node number parsing subunit is used to parse out the node number of the single node requesting power failure after receiving a power failure request on the backplane. The power outage notification broadcast subunit is used to generate a power outage notification on the backplane based on the node number of the single node requesting a power outage, and send it to other nodes of the multi-node server.

[0020] Furthermore, the negotiated individual power-off module includes: The business interaction judgment unit is used to determine whether there is any business interaction between the single node requesting the power outage and its own node after each node in the multi-node server receives a power outage notification sent by the backplane. The power outage information return unit is used to initiate business handover when the single node requesting the power outage has no business dealings with its own node, or when the single node requesting the power outage has business dealings with its own node, and after the business handover is completed, to return the power outage information to the backplane. The power-off restart permission unit is used to perform a power-off restart on a single node that requests a power-off in the backplane. A separate power-off unit is used by a single node requesting a power-off to initiate a separate power-off within a multi-node server based on its own power supply mode.

[0021] Furthermore, the business transaction determination unit includes: The request node number parsing subunit is used to parse out the node number of the single node requesting power outage after each node of the multi-node server receives the power outage notification sent by the backplane. The business transaction judgment subunit is used to determine, within a multi-node server, whether the single node requesting a power outage has any business transactions with itself, based on the node number. The no-traffic node consent return sub-unit is used when a single node requesting a power outage has no business dealings with itself. It encapsulates its own node number into the consent power outage information and returns it to the middle backplane.

[0022] Furthermore, the power outage information return unit includes: The service handover initiation subunit is used to initiate service handover with nodes that have business dealings with a single node requesting a power outage, thereby suspending or transferring the services of that single node. The incoming and outgoing nodes agree to return the sub-unit, which is used to encapsulate the node number of the relevant nodes in the power-off information after the business handover is completed, and return it to the backplane.

[0023] Furthermore, the power-off restart permission unit includes: The node number parsing subunit is used to receive the agreed power-off information returned by each node in the multi-node server on the backplane and parse out the node number. The "Agree Node Count Determination Subunit" is used to determine, based on the node number, whether the number of nodes returning the "Agree Power Off Information" is equal to the number of nodes in the multi-node server. The power outage notification is resent to the sub-unit. When the number of nodes returning the power outage information is not equal to the number of nodes in the multi-node server, the node that did not return the power outage information is located in the backplane and the power outage notification is resent to the located node. The power failure notification sending count subunit is used to determine in the backplane whether the number of power failure notifications sent to the positioning node exceeds a set threshold. The timeout alarm sending subunit is used to send a timeout alarm when the number of times the power outage notification is sent to the positioning node exceeds a set threshold. The sub-unit that allows power-off command issuance is used to send a power-off permission command to the single node requesting the power-off when the number of nodes returning agreed power-off information is equal to the number of nodes in the multi-node server.

[0024] Furthermore, the individual power-off unit includes: The self-powered module obtains a sub-unit, which is used to obtain the self-powered mode of a single node requesting a power outage; The external power cut-off subunit is used when the power supply mode is external power supply mode. When a single node requests power cut-off, it sends a power cut-off command to cut off the motherboard from the external power supply and sends a command to prohibit the motherboard from drawing power from the adjacent node, thereby realizing the power cut-off restart of the single node. The neighboring node motherboard power cut-off subunit is used when the power supply mode is adjacent node power supply mode. When a single node requests power cut-off, it sends a power cut-off command to cut off the motherboard from the neighboring node motherboard and sends a command to prohibit power from the neighboring node motherboard, thereby realizing the power cut-off restart of the single node.

[0025] Thirdly, the present invention provides an apparatus including a processor and a memory; The memory is used to store a computer program, and the processor is used to retrieve and run the computer program from the memory, causing the device to perform the method described in the first aspect.

[0026] Fourthly, the present invention provides a storage medium, The storage medium stores instructions that, when run on a computer, cause the computer to perform the method described in the first aspect above.

[0027] The beneficial effects of this invention are as follows: The present invention provides a method and apparatus for implementing single-node power failure in a multi-node server. By broadcasting a power failure notification through the backplane, the power failure node requests the other nodes of the multi-node server to complete the business handover before disconnecting its own power supply, thereby achieving a rapid response to single-node power failure requests and meeting the user's need for single-node power failure without affecting the overall business of the multi-node server.

[0028] Furthermore, the design principle of this invention is reliable, the structure is simple, and it has a very wide range of application prospects.

[0029] Therefore, it is evident that the present invention has outstanding substantive features and significant progress compared with the prior art, and the beneficial effects of its implementation are also obvious. Attached Figure Description

[0030] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] Figure 1 This is a flowchart illustrating Embodiment 1 of the method for implementing single-node power failure in a multi-node server according to the present invention.

[0032] Figure 2 This is a flowchart illustrating Embodiment 2 of the method for implementing single-node power failure in a multi-node server according to the present invention.

[0033] Figure 3 This is a flowchart illustrating Embodiment 3 of the method for implementing single-node power failure in a multi-node server according to the present invention.

[0034] Figure 4 This is a schematic diagram of the device for implementing single-node power failure in a multi-node server according to the present invention. Detailed Implementation

[0035] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0036] For multi-node servers, the ability for each node to independently power down is crucial for the entire server. Independent power-down of a node not only ensures the normal operation of other nodes but also guarantees that a single node can respond promptly to urgent power-down requests from customers. Currently, the power-down method for multi-node servers involves sending a power-down / restart command to a backplane. Upon receiving the command, the backplane determines its priority and only powers down and restarts the server if no other nodes are consuming resources. Furthermore, this involves a complete power-down of the entire server. If several other nodes are consuming resources and the command has a higher priority, the machine cannot be powered down, making it impossible to respond promptly to user requests for power-down of a single node. Forcibly powering down the entire server may disrupt the normal operation of other nodes, causing their services to be suspended or interrupted. The following implementation addresses these issues.

[0037] Example 1: like Figure 1 As shown, the present invention provides a method for implementing single-node power failure in a multi-node server, comprising the following steps: S1. After receiving a power failure request from a single node, the backplane notifies all nodes in the multi-node server. S2. After receiving the power-off consent information returned by the business-related nodes of a single node, the backplane in the middle allows that single node to perform a power-off independently from the multi-node server.

[0038] For multi-node servers, each node can represent an independent server, but at the same time, it is connected to other nodes of the entire server through the design of some hardware cables. Therefore, they are both independent and interconnected. If a single node needs to be powered off and restarted due to business needs or emergencies, the single node needs to control its own power-on and power-off. After receiving the power-off and restart command, the single node waits for the backplane to send a message to other nodes informing that the single node is about to be powered off. After completing the handover of business with other nodes within a reasonable time period and completing the business cut-off, the single node can independently complete the power-off and restart.

[0039] The present invention provides a method for implementing single-node power failure in a multi-node server. By broadcasting a power failure notification through the backplane, the power failure node requests the other nodes of the multi-node server to complete the business handover before disconnecting its own power supply. This enables a rapid response to single-node power failure requests and meets the user's need for single-node power failure without affecting the overall business of the multi-node server.

[0040] Example 2: like Figure 2 As shown, the present invention provides a method for implementing single-node power failure in a multi-node server, comprising the following steps: S1. After receiving a power-off request from a single node, the backplane notifies all nodes in the multi-node server. The specific steps of step S1 are as follows: S11. The user sends a power-off restart command to a single node in a multi-node server via out-of-band or in-band methods. S12. Upon receiving a power-off restart command, the single node sends a power-off request to the backplane. S13. After receiving the power failure request, the backplane sends a power failure notification to each node in the multi-node server. S2. After the backplane receives the power-off consent information from the service-related nodes of a single node, it allows that single node to be powered off independently from the multi-node server; the specific steps of step S2 are as follows: S21. After receiving the power outage notification sent by the backplane, each node in the multi-node server determines whether the single node requesting the power outage has any business dealings with its own node. If so, proceed to step S22; If not, return the agreed power-off information to the backplane and proceed to step S23; S22. Nodes that have business dealings with the single node requesting power outage initiate business handover, and after the business handover is completed, return the power outage consent information to the backplane. S23. The backplane allows a single node that requests a power outage to perform a power outage restart; S24. The single node requesting a power outage initiates a separate power outage within the multi-node server based on its own power supply mode.

[0041] The present invention provides a method for implementing single-node power failure in a multi-node server. By broadcasting a power failure notification through the backplane, the power failure node requests the other nodes of the multi-node server to complete the business handover before disconnecting its own power supply. This enables a rapid response to single-node power failure requests and meets the user's need for single-node power failure without affecting the overall business of the multi-node server.

[0042] Example 3: like Figure 3 As shown, the present invention provides a method for implementing single-node power failure in a multi-node server, comprising the following steps: S1. After receiving a power-off request from a single node, the backplane notifies all nodes in the multi-node server. The specific steps of step S1 are as follows: S11. The user sends a power-off restart command to a single node in a multi-node server via out-of-band or in-band methods. S12. Upon receiving a power-off restart command, the single node sends a power-off request to the backplane; the specific steps of step S12 are as follows: S121. A single node that receives a power failure restart command obtains its own node number; S122. Upon receiving the power failure restart command, the single node encapsulates its own node number into the power failure request and sends it to the backplane. S13. After receiving the power-off request, the backplane sends a power-off notification to each node in the multi-node server; the specific steps of step S13 are as follows: S131. After receiving a power-off request, the backplane in the middle parses out the node number of the single node requesting the power-off. S132. The backplane generates a power outage notification based on the node number of the single node requesting a power outage and sends it to other nodes of the multi-node server. S2. After the backplane receives the power-off consent information from the service-related nodes of a single node, it allows that single node to be powered off independently from the multi-node server; the specific steps of step S2 are as follows: S21. After receiving the power outage notification sent by the backplane, each node in the multi-node server determines whether the single node requesting the power outage has any business dealings with its own node. If so, proceed to step S22; If not, return the agreed power-off information to the backplane and proceed to step S23; The specific steps of step S21 are as follows: S211. After receiving the power failure notification sent by the backplane, each node of the multi-node server parses out the node number of the single node requesting the power failure. S212. Within a multi-node server, each node determines, based on its node number, whether the single node requesting a power outage has any business dealings with itself. If so, proceed to step S22; If not, encapsulate its own node number into the agreed power failure information and return it to the middle backplane, proceeding to step S23; S22. Nodes that have business dealings with the single node requesting power outage initiate business handover, and after the business handover is completed, return a power outage agreement message to the backplane; the specific steps of step S22 are as follows: S221. Initiate service handover with nodes that have business dealings with the single node requesting power outage, and suspend or transfer the services of that single node; S222. After the service handover is completed, the relevant nodes of the service will encapsulate their own node number in the power outage information and return it to the backplane. S23. The backplane allows a single node that requests a power outage to perform a power-off restart; the specific steps of S23 are as follows: S231. The backplane receives the agreed power-off information returned by each node in the multi-node server and parses out the node number; S232. The backplane determines whether the number of nodes returning the agreed power outage information is equal to the number of nodes in the multi-node server based on the node number; If so, proceed to step S235; If not, proceed to step S233; S233. The backplane positioning node did not return the power outage information and sent the power outage notification to the positioning node again; S234. The backplane determines whether the number of times it sends a power failure notification to the positioning node exceeds a set threshold. If so, issue a timeout alarm and proceed to step S235; If not, return to step S21; S235. The backplane sends a power-off permission command to the single node that requested the power-off; S24. The single node requesting a power outage initiates a separate power outage within the multi-node server based on its own power supply mode; the specific steps of step S24 are as follows: S241. A single node requesting a power outage obtains its own power supply mode; When the self-powered mode is external power supply mode, proceed to step S242; When its own power supply mode is the adjacent node power supply mode, proceed to step S243; S242. The single node requesting a power outage sends a power cut-off command to cut off the motherboard's power supply from the external power source and sends a command to prohibit the motherboard from drawing power from neighboring nodes, thereby achieving a power outage restart for the single node and ending the process. S243. The single node requesting power outage sends a power cut-off command to cut off the motherboard from drawing power from the neighboring node's motherboard, and sends a command to prohibit drawing power from the neighboring node's motherboard, thereby achieving power outage restart of the single node.

[0043] Example 4: like Figure 4 As shown, the present invention provides a device for implementing single-node power failure in a multi-node server, comprising: The power outage notification module is used to notify all nodes in the multi-node server after receiving a power outage request from a single node. The negotiated individual power-off module is used to allow a single node to independently power off from the multi-node server after receiving consent to power off from the business-related nodes of that single node.

[0044] The device for implementing single-node power failure in a multi-node server provided by this invention broadcasts a power failure notification through the backplane, requesting the power failure node to disconnect its own power supply after completing the business handover with each node of the multi-node server, thereby achieving a rapid response to single-node power failure requests and meeting the user's need for single-node power failure without affecting the overall business of the multi-node server.

[0045] Example 5: like Figure 4 As shown, the present invention provides a device for implementing single-node power failure in a multi-node server, comprising: The power outage notification module is used to notify all nodes in the multi-node server after receiving a power outage request from a single node. The power outage notification module includes: The power-off restart command sending unit is used to send a power-off restart command to a single node in a multi-node server via out-of-band or in-band methods. The power failure request issuing unit is used to send a power failure request to the backplane from a single node that receives a power failure restart command. The power failure notification sending unit is used to send power failure notifications to each node in the multi-node server after receiving a power failure request on the backplane. The negotiated individual power-off module is used to allow a single node to independently power off from the multi-node server after receiving consent for power-off from the relevant service nodes. The negotiated individual power-off module includes: The business interaction judgment unit is used to determine whether there is any business interaction between the single node requesting the power outage and its own node after each node in the multi-node server receives a power outage notification sent by the backplane. The power outage information return unit is used to initiate business handover when the single node requesting the power outage has no business dealings with its own node, or when the single node requesting the power outage has business dealings with its own node, and after the business handover is completed, to return the power outage information to the backplane. The power-off restart permission unit is used to perform a power-off restart on a single node that requests a power-off in the backplane. A separate power-off unit is used by a single node requesting a power-off to initiate a separate power-off within a multi-node server based on its own power supply mode.

[0046] The device for implementing single-node power failure in a multi-node server provided by this invention broadcasts a power failure notification through the backplane, requesting the power failure node to disconnect its own power supply after completing the business handover with each node of the multi-node server, thereby achieving a rapid response to single-node power failure requests and meeting the user's need for single-node power failure without affecting the overall business of the multi-node server.

[0047] Example 6: like Figure 4 As shown, the present invention provides a device for implementing single-node power failure in a multi-node server, comprising: The power outage notification module is used to notify all nodes in the multi-node server after receiving a power outage request from a single node. The power outage notification module includes: The power-off restart command sending unit is used to send a power-off restart command to a single node in a multi-node server via out-of-band or in-band methods. A power failure request issuing unit is used by a single node that receives a power failure restart command to issue a power failure request to the backplane; the power failure request issuing unit includes: The self-node number acquisition sub-unit is used for a single node to obtain its own node number upon receiving a power failure restart command. The node number encapsulation and power failure request sending subunit is used for a single node that receives a power failure restart command to encapsulate its own node number into a power failure request and send it to the middle backplane; A power outage notification sending unit is used to send a power outage notification to each node in the multi-node server after receiving a power outage request on the backplane. The power outage notification sending unit includes: The node number parsing subunit is used to parse out the node number of the single node requesting power failure after receiving a power failure request on the backplane. The power outage notification broadcast subunit is used to generate a power outage notification on the backplane based on the node number of the single node requesting a power outage, and send it to other nodes of the multi-node server. The negotiated individual power-off module is used to allow a single node to independently power off from the multi-node server after receiving consent for power-off from the relevant service nodes. The negotiated individual power-off module includes: The business communication determination unit is used by each node in a multi-node server to determine whether there is any business communication between the requesting node and itself after receiving a power outage notification from the backplane. The business communication determination unit includes: The request node number parsing subunit is used to parse out the node number of the single node requesting power outage after each node of the multi-node server receives the power outage notification sent by the backplane. The business transaction judgment subunit is used to determine, within a multi-node server, whether the single node requesting a power outage has any business transactions with itself, based on the node number. The no-traffic node consent return subunit is used when a single node requesting a power outage has no business dealings with itself. It encapsulates its own node number into the consent power outage information and returns it to the middle backplane. The power outage consent return unit is used when a single node requesting a power outage has no business communication with itself, or when a single node requesting a power outage has business communication with itself, initiates business handover, and returns a power outage consent information to the backplane after the business handover is completed; the power outage consent return unit includes: The service handover initiation subunit is used to initiate service handover with nodes that have business dealings with a single node requesting a power outage, thereby suspending or transferring the services of that single node. The incoming and outgoing nodes agree to return to the sub-unit, which is used to encapsulate the node number of the relevant nodes in the power-off information after the business handover is completed, and return it to the middle backplane; A power-off restart permission unit is used to perform a power-off restart on a single node that requests a power-off request in the backplane. The power-off restart permission unit includes: The node number parsing subunit is used to receive the agreed power-off information returned by each node in the multi-node server on the backplane and parse out the node number. The "Agree Node Count Determination Subunit" is used to determine, based on the node number, whether the number of nodes returning the "Agree Power Off Information" is equal to the number of nodes in the multi-node server. The power outage notification is resent to the sub-unit. When the number of nodes returning the power outage information is not equal to the number of nodes in the multi-node server, the node that did not return the power outage information is located in the backplane and the power outage notification is resent to the located node. The power failure notification sending count subunit is used to determine in the backplane whether the number of power failure notifications sent to the positioning node exceeds a set threshold. The timeout alarm sending subunit is used to send a timeout alarm when the number of times the power outage notification is sent to the positioning node exceeds a set threshold. The sub-unit that allows power outage instructions is used to send a power outage instruction to the single node requesting the power outage when the number of nodes returning the power outage agreement is equal to the number of nodes in the multi-node server. A separate power-off unit is used by a single node requesting a power-off to initiate a separate power-off within the multi-node server based on its own power supply mode; the separate power-off unit includes: The self-powered module obtains a sub-unit, which is used to obtain the self-powered mode of a single node requesting a power outage; The external power cut-off subunit is used when the power supply mode is external power supply mode. When a single node requests power cut-off, it sends a power cut-off command to cut off the motherboard from the external power supply and sends a command to prohibit the motherboard from drawing power from the adjacent node, thereby realizing the power cut-off restart of the single node. The neighboring node motherboard power cut-off subunit is used when the power supply mode is adjacent node power supply mode. When a single node requests power cut-off, it sends a power cut-off command to cut off the motherboard from the neighboring node motherboard and sends a command to prohibit power from the neighboring node motherboard, thereby realizing the power cut-off restart of the single node.

[0048] Example 7: This invention provides a device, including a processor and a memory; The memory is used to store computer programs, and the processor is used to call and run the computer programs from the memory, so that the device performs the methods described in Embodiment 1, Embodiment 2, or Embodiment 3.

[0049] Example 8: This invention provides a storage medium, The storage medium stores instructions that, when run on a computer, cause the computer to execute the methods described in Embodiment 1, Embodiment 2, or Embodiment 3.

[0050] Although the present invention has been described in detail with reference to the accompanying drawings and preferred embodiments, the invention is not limited thereto. Various equivalent modifications or substitutions can be made to the embodiments of the invention by those skilled in the art without departing from the spirit and essence of the invention, and such modifications or substitutions should all be within the scope of the invention. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the invention should also be covered within the protection scope of the invention. Therefore, the protection scope of the invention should be determined by the scope of the claims.

Claims

1. A method for implementing single-node power failure in a multi-node server, characterized in that, Includes the following steps: S1. After receiving a power failure request from a single node, the backplane notifies all nodes in the multi-node server. S2. After receiving the power-off consent information returned by the business-related nodes of a single node, the backplane allows that single node to perform a power-off independently from the multi-node server. The specific steps of step S2 are as follows: S21. After receiving the power outage notification sent by the backplane, each node in the multi-node server determines whether the single node requesting the power outage has any business dealings with its own node. If so, proceed to step S22; If not, return the agreed power-off information to the backplane and proceed to step S23; S22. Nodes that have business dealings with the single node requesting power outage initiate business handover, and after the business handover is completed, return the power outage consent information to the backplane. S23. The backplane allows a single node that requests a power outage to perform a power outage restart; S24. The single node requesting a power outage initiates a separate power outage within the multi-node server based on its own power supply mode.

2. The method for implementing single-node power failure in a multi-node server as described in claim 1, characterized in that, The specific steps of step S1 are as follows: S11. The user sends a power-off restart command to a single node in a multi-node server via out-of-band or in-band methods. S12. Upon receiving a power-off restart command, the single node sends a power-off request to the backplane. S13. After receiving the power outage request, the backplane sends a power outage notification to each node in the multi-node server.

3. The method for implementing single-node power failure in a multi-node server as described in claim 2, characterized in that, The specific steps of step S12 are as follows: S121. A single node that receives a power failure restart command obtains its own node number; S122. Upon receiving the power failure restart command, the single node encapsulates its own node number into the power failure request and sends it to the backplane.

4. The method for implementing single-node power failure in a multi-node server as described in claim 3, characterized in that, The specific steps of step S13 are as follows: S131. After receiving a power-off request, the backplane in the middle parses out the node number of the single node requesting the power-off. S132. The backplane generates a power outage notification based on the node number of the single node requesting the power outage and sends it to other nodes of the multi-node server.

5. The method for implementing single-node power failure in a multi-node server as described in claim 4, characterized in that, The specific steps of step S21 are as follows: S211. After receiving the power failure notification sent by the backplane, each node of the multi-node server parses out the node number of the single node requesting the power failure. S212. Within a multi-node server, each node determines, based on its node number, whether the single node requesting a power outage has any business dealings with itself. If so, proceed to step S22; If not, encapsulate its own node number into the agreed power outage information and return it to the backplane, proceeding to step S23.

6. The method for implementing single-node power failure in a multi-node server as described in claim 4, characterized in that, The specific steps of step S22 are as follows: S221. Initiate service handover with nodes that have business dealings with the single node requesting power outage, and suspend or transfer the services of that single node; S222. After the service handover is completed, the relevant nodes of the service will encapsulate their own node number in the power outage information and return it to the backplane.

7. The method for implementing single-node power failure in a multi-node server as described in claim 6, characterized in that, The specific steps of step S23 are as follows: S231. The backplane receives the agreed power-off information returned by each node in the multi-node server and parses out the node number; S232. The backplane determines whether the number of nodes returning the agreed power outage information is equal to the number of nodes in the multi-node server based on the node number; If so, proceed to step S235; If not, proceed to step S233; S233. The backplane positioning node did not return the power outage information and sent the power outage notification to the positioning node again; S234. The backplane determines whether the number of times it sends a power failure notification to the positioning node exceeds a set threshold. If so, issue a timeout alarm and proceed to step S235; If not, return to step S21; S235. The backplane sends a power-off permission command to the single node that requests a power-off.

8. The method for implementing single-node power failure in a multi-node server as described in claim 4, characterized in that, The specific steps of step S24 are as follows: S241. A single node requesting a power outage obtains its own power supply mode; When the self-powered mode is external power supply mode, proceed to step S242; When its own power supply mode is the adjacent node power supply mode, proceed to step S243; S242. The single node requesting a power outage sends a power cut-off command to cut off the motherboard's power supply from the external power source and sends a command to prohibit the motherboard from drawing power from neighboring nodes, thereby achieving a power outage restart for the single node and ending the process. S243. The single node requesting power outage sends a power cut-off command to cut off the motherboard from drawing power from the neighboring node's motherboard, and sends a command to prohibit drawing power from the neighboring node's motherboard, thereby achieving power outage restart of the single node.

9. A device for implementing single-node power failure in a multi-node server, characterized in that, include: The power outage notification module is used to notify all nodes in the multi-node server after receiving a power outage request from a single node. The negotiated individual power-off module is used to allow a single node to independently power off from the multi-node server after receiving consent to power off from the business-related nodes of the single node. The separately powered-off modules after negotiation include: The business interaction judgment unit is used to determine whether there is any business interaction between the single node requesting the power outage and its own node after each node in the multi-node server receives a power outage notification sent by the backplane. The power outage information return unit is used to initiate business handover when the single node requesting the power outage has no business dealings with its own node, or when the single node requesting the power outage has business dealings with its own node, and after the business handover is completed, to return the power outage information to the backplane. The power-off restart permission unit is used to perform a power-off restart on a single node that requests a power-off in the backplane. A separate power-off unit is used by a single node requesting a power-off to initiate a separate power-off within the multi-node server based on its own power supply mode.

Images