CN116755533A - Method and device for realizing single-node power failure of multi-node server - Google Patents
Method and device for realizing single-node power failure of multi-node server Download PDFInfo
- Publication number
- CN116755533A CN116755533A CN202310504722.XA CN202310504722A CN116755533A CN 116755533 A CN116755533 A CN 116755533A CN 202310504722 A CN202310504722 A CN 202310504722A CN 116755533 A CN116755533 A CN 116755533A
- Authority
- CN
- China
- Prior art keywords
- node
- power
- server
- middle backboard
- requesting
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000004891 communication Methods 0.000 claims description 5
- 230000004044 response Effects 0.000 abstract description 6
- 238000004806 packaging method and process Methods 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 4
- 238000004590 computer program Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000725 suspension Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/26—Power supply means, e.g. regulation thereof
- G06F1/266—Arrangements to supply power to external peripherals either directly from the computer or under computer control, e.g. supply of power through the communication port, computer controlled power-strips
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention provides a method and a device for realizing single-node power failure of a multi-node server, belonging to the technical field of server power control, wherein the method comprises the following steps: s1, after receiving a power-off restarting request of a single node, a middle backboard informs each node in a multi-node server; s2, after receiving the power-off agreement information returned by the service related node of the single node, the middle back plate allows the single node to independently power off from the multi-node server. According to the method and the device for realizing single-node power failure of the multi-node server, the power supply of each node of the power failure node and the multi-node server is disconnected after service handover is completed through the middle backboard broadcasting power failure notification, so that the quick response of the single-node power failure request is realized, and the requirement of a user on the single-node power failure is met under the condition that the whole service of the multi-node server is not influenced.
Description
Technical Field
The invention belongs to the technical field of server power supply control, and particularly relates to a method and a device for realizing single-node power failure of a multi-node server.
Background
In the current power-on process of the server system, power distribution and conversion are performed by a host, PCIe daughter cards, back boards and the like are powered on from a main board through cables or connectors, and after corresponding power lines on the main board are established, all the daughter cards can complete the power-on process at the same time.
Because the operation and business of each server as a single point server are performed separately, the current server power control strategy can completely meet the power supply requirement of the single-node server. However, for the node server, when any node needs to be powered off by configuring a middle backboard, a power-off restarting instruction needs to be transmitted to the middle backboard, after the middle backboard receives the instruction, the priority of the instruction is judged, and when the priority of the instruction is higher and no other node occupies resources, the power of the whole multi-node server can be powered off, and if the other node occupies resources and the priority of the node occupying the resources is higher, the power of the whole server is not allowed, so that the requirement of the power-off node cannot be met. If the whole server is powered off to meet the requirement of the node needing to be powered off, the service of other nodes is suspended or interrupted.
In summary, the current single-node power-off requirement of the single-node service requires power-off of the whole server, and the problem of service conflict with other nodes exists.
In view of the above drawbacks, it is very necessary to provide a method and apparatus for implementing single-node power outage of a multi-node server.
Disclosure of Invention
Aiming at the defect that the power-off requirement of a single node of the current single-node service needs to power off the whole server and has business conflict with other nodes, the invention provides a method and a device for realizing single-node power-off of a multi-node server, so as to solve the technical problems.
In a first aspect, the present invention provides a method for implementing single-node power outage by using a multi-node server, including the following steps:
s1, after receiving a power-off restarting request of a single node, a middle backboard informs each node in a multi-node server;
and S2, after receiving the power-off agreement information returned by the service related node of the single node, the middle back plate allows the single node to independently power off from the multi-node server. And the single node informs each node in the multi-node server of the self power-off restarting request through the middle backboard, and independently powers off the self node after receiving the power-off permission of each node, so as to prevent the business influence on other nodes.
Further, the specific steps of step S1 are as follows:
s11, a user sends a power-off restarting instruction to a single node in the multi-node server in an out-of-band or in-band mode;
s12, a single node which receives the power-off restarting instruction sends a power-off request to the middle backboard;
and S13, after the middle backboard receives the power-off request, sending a power-off notification to each node in the multi-node server. The single node can receive the in-band power-off restarting instruction, and also can receive the power-off restarting instruction returned by the BMC from the out-of-band, and the middle backboard realizes the broadcasting of the power-off notification.
Further, the specific steps of step S12 are as follows:
s121, a single node which receives a power-off restarting instruction acquires a self node number;
s122, the single node which receives the power-off restarting instruction encapsulates the node number of the single node into a power-off request and sends the power-off request to the middle backboard. When a single node sends a power-off request, the node numbers of the single node are simultaneously sent together.
Further, the specific steps of step S13 are as follows:
s131, after the middle backboard receives the power-off request, analyzing the node number of a single node requesting power-off;
and S132, the middle backboard generates a power-off notification according to the node number of the single node requesting power-off and sends the power-off notification to other nodes of the multi-node server. The middle backboard sends the node number of the single node requesting the power off and the power off notification together so as to inform other nodes in the server which single node requests the power off.
Further, the specific steps of step S2 are as follows:
s21, after each node in the multi-node server receives a power-off request sent by the middle backboard, judging whether a single node requesting power-off and a self node have business to and from each other;
if yes, go to step S22;
if not, returning the power-off agreement information to the middle backboard, and entering step S23;
s22, a node having service communication with a single node requesting power failure starts service handover, and after the service handover is completed, power failure agreement information is returned to the middle backboard;
s23, allowing a single node requesting power failure to perform power failure restarting by the middle backboard;
s24, the single node requesting power failure starts independent power failure in the multi-node server according to the self power supply mode. The node in the multi-node server, which has service to and from a single node requesting power off, needs to start service handover so as to avoid the influence of service when the single node is powered off, ensure that the single node is independent in the server node, is not related to other nodes, and provides for subsequent independent power off.
Further, the specific steps of step S21 are as follows:
s211, after each node of the multi-node server receives a power-off request sent by the middle backboard, analyzing the node number of a single node requesting power-off;
s212, each node in the multi-node server judges whether a single node requesting power failure and a self node have business to and from according to the node number;
if yes, go to step S22;
if not, the self node number is packaged into the power-off agreement information, and returns to the middle backboard, and the step S23 is entered. The service communication among nodes of the multi-node server needs to be carried out corresponding data transmission through node numbers so as to ensure the accuracy of the transmission.
Further, the specific steps of step S22 are as follows:
s221, a node having service to and from a single node requesting power off starts service handover, and the service of the single node is stopped or transferred;
s222, after the service handover is completed, the service related node encapsulates the node number of the service related node into the outage information and returns the outage information to the middle backboard. After the service handover is completed, the power-off information needs to be timely returned to be agreed so that the single node can timely start the power-off process.
Further, the specific steps of step S23 are as follows:
s231, the middle backboard receives the power-off agreeing information returned by each node in the multi-node server, and analyzes the node number;
s232, judging whether the number of the nodes returning the power-off agreement information is equal to the number of the nodes in the multi-node server or not by the middle back plate according to the node numbers;
if yes, go to step S235;
if not, go to step S233;
s233, the middle backboard locates the node which does not return the power-off information, and sends a power-off notification to the locating node again;
s234, the middle backboard judges whether the number of times of power failure notification sent to the positioning node exceeds a set threshold value or not;
if yes, sending out a timeout alarm, and entering step S235;
if not, returning to the step S21;
s235, the middle backboard sends a power-off permission instruction to a single node requesting power off. The middle backboard confirms the number of the nodes replying to the power failure, judges whether other nodes except the single node have the power failure, if the nodes do not reply, positions the node, and sends the power failure notification again, so that the node service is ensured to wait when the service handover time is too long, the failure of sending the power failure notification to the node is avoided, and the limitation of the threshold value of the power failure notification times is to eliminate the situation that the power failure information can not be returned in time due to the failure of a link for sending the power failure notification, so that the power failure request of the single node can be executed.
Further, the specific steps of step S24 are as follows:
s241, a single node requesting power failure acquires a self power supply mode;
when the self power supply mode is the external power supply mode, the step S242 is entered;
when the self power supply mode is the adjacent node power supply mode, entering step S243;
s242, a single node requesting power off sends a power off instruction, cuts off the power of the main board from an external power supply, sends a power off prohibition to the main board of a neighboring node, and achieves the power off restarting of the single node and ends;
s243, the single node requesting power off sends a power off instruction, the power off main board takes power from the main board of the adjacent node, and sends a power off prohibition to take power from the main board of the adjacent node, so that the power off restarting of the single node is realized. The power supply in the multi-node server is powered by the external power supply from the main board of each node, or the power supply of the main boards of other nodes in the multi-node server is powered by two modes, no matter which mode is powered, the main board power supply is required to be disconnected, and because of the power supply stability requirement of the multi-node server, when a certain power supply of the main board fails, the power is autonomously taken from other nodes, so that the main and standby power supply of the nodes is realized, the power supply of the main boards of other nodes is forbidden, the power-off restarting of a single node can be truly realized, the situation that one power supply is disconnected and the power is taken again from the main boards of other nodes is avoided.
In a second aspect, the present invention provides an apparatus for implementing single-node power outage of a multi-node server, including:
the power-off notification module is used for notifying each node in the multi-node server after receiving a power-off restarting request of a certain single node;
and the independent power-off module after negotiation is used for allowing the single node to independently power off from the multi-node server after receiving the power-off agreement information returned by the service related node of the single node.
Further, the power outage notification module includes:
the power-off restarting instruction sending unit is used for sending a power-off restarting instruction to a certain single node in the multi-node server in an out-of-band or in-band mode;
the power-off request sending unit is used for sending a power-off request to the middle backboard by the single node which receives the power-off restarting instruction;
and the power-off notification sending unit is used for sending power-off notification to each node in the multi-node server after the middle backboard receives the power-off request.
Further, the power-off request issuing unit includes:
the self node number acquisition subunit is used for acquiring the self node number by a single node which receives the power-off restarting instruction;
and the node number packaging and power-off request sending subunit is used for packaging the self node number into the power-off request by the single node which receives the power-off restarting instruction and sending the power-off request to the middle backboard.
Further, the power outage notification transmission unit includes:
the node number analysis subunit is used for analyzing the node number of the single node requesting power off after the middle backboard receives the power off request;
and the power-off notification broadcasting subunit is used for generating a power-off notification on the middle backboard according to the node number of the single node requesting power-off and sending the power-off notification to other nodes of the multi-node server.
Further, the individual power-off module after negotiation includes:
the service business transaction judging unit is used for judging whether a single node requesting power failure and a self node have business transaction after each node in the multi-node server receives the power failure request sent by the middle backboard;
the power-off agreement information return unit is used for enabling the single node requesting power off to have no business with the self node or enabling the single node requesting power off to have business with the self node, starting business handover, and returning power-off agreement information to the middle backboard after the business handover is completed;
the power-off restarting allowing unit is used for allowing a single node requesting power off to perform power-off restarting on the middle backboard;
and the independent power-off unit is used for enabling the single node requesting power off to start independent power off in the multi-node server according to the self power supply mode.
Further, the service traffic judgment unit includes:
the request node number analysis subunit is used for analyzing the node number of a single node requesting power failure after each node of the multi-node server receives the power failure request sent by the middle backboard;
the service business trip judging subunit is used for judging whether a single node requesting power off and a self node have business trips or not according to the node number in the multi-node server by each node;
and the non-current node agrees with the returning subunit, and when the single node requesting the power-off does not have business with the self node, the self node number is packaged into the information agreeing to power-off, and the information is returned to the middle backboard.
Further, the consent power outage information return unit includes:
a service handover promoter unit for starting service handover with a node having service to/from a single node requesting power off, and stopping or transferring the service of the single node;
and the sending and receiving node agrees to return the subunit, which is used for packaging the node number of the sending and receiving node number into the outage agreeing information at the service phase node after the service handover is completed, and returning the outage agreeing information to the middle backboard.
Further, the power-off restart permission unit includes:
the consent node number analysis subunit is used for receiving consent power-off information returned by each node in the multi-node server at the middle backboard and analyzing out the node number;
the consent node quantity judging subunit is used for judging whether the quantity of the nodes returning consent power-off information is equal to the quantity of the nodes in the multi-node server or not according to the node number in the middle back plate;
the power-off notification resending subunit is used for locating the node which does not return the power-off information in the middle backboard and resending the power-off notification to the locating node when the number of the nodes which return the power-off information is unequal to the number of the nodes in the multi-node server;
the power-off notification sending frequency subunit is used for judging whether the power-off notification sending frequency exceeds a set threshold value or not to the positioning node in the middle backboard;
the timeout alarm sending subunit is used for sending timeout alarms when the power-off notification times to the positioning node exceed a set threshold value;
and the power-off permission instruction sending subunit is used for sending a power-off permission instruction to a single node requesting power off in the middle backboard when the number of the nodes which return the power-off permission information is equal to the number of the nodes in the multi-node server.
Further, the individual power-off unit includes:
the self power supply module acquisition subunit is used for acquiring a self power supply mode at a single node requesting power failure;
the external power supply cut-off subunit is used for sending a power supply cut-off instruction to a single node which requests power off when the self power supply mode is the external power supply mode, cutting off the power supply of the main board from the external power supply, and sending a power supply prohibition instruction to prohibit the power supply of the main board of the adjacent node, so that the power off restarting of the single node is realized;
and the adjacent node main board power supply cutting-off subunit is used for sending a power supply cutting-off instruction to a single node requesting power off when the self power supply mode is the adjacent node power supply taking mode, cutting off the power supply of the main board from the adjacent node main board, and sending a power supply prohibiting power supply from the main board of the adjacent node to realize the power off restarting of the single node.
In a third aspect, the invention provides an apparatus comprising a processor and a memory;
wherein the memory is for storing a computer program and the processor is for calling and running the computer program from the memory for causing the device to perform the method according to the first aspect.
In a fourth aspect, the present invention provides a storage medium,
the storage medium has stored therein instructions which, when run on a computer, cause the computer to perform the method of the first aspect described above.
The invention has the beneficial effects that:
according to the method and the device for realizing single-node power failure of the multi-node server, the power supply of each node of the power failure node and the multi-node server is disconnected after service handover is completed through the middle backboard broadcasting power failure notification, so that the quick response of the single-node power failure request is realized, and the requirement of a user on the single-node power failure is met under the condition that the whole service of the multi-node server is not influenced.
In addition, the invention has reliable design principle, simple structure and very wide application prospect.
It can be seen that the present invention has outstanding substantial features and significant advances over the prior art, as well as the benefits of its implementation.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the description of the embodiments or the prior art will be briefly described below, and it will be obvious to those skilled in the art that other drawings can be obtained from these drawings without inventive effort.
FIG. 1 is a flow chart of an embodiment 1 of a method for implementing single-node power-off by a multi-node server according to the present invention.
FIG. 2 is a flow chart of embodiment 2 of a method for implementing single-node power-off by a multi-node server according to the present invention.
FIG. 3 is a flow chart of embodiment 3 of a method for implementing single-node power-off by a multi-node server according to the present invention.
FIG. 4 is a schematic diagram of an apparatus for implementing single-node power down for a multi-node server of the present invention.
Detailed Description
In order to make the technical solution of the present invention better understood by those skilled in the art, the technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
For a multi-node server, each node can be independently powered off to be important for the whole server, the node can be independently powered off to ensure that normal operation of other node services cannot be influenced, a single node can also be ensured to respond to urgent power-off demands of clients in time, a power-off restarting instruction is transmitted to a middle backboard, the middle backboard judges the priority of the instruction after receiving the instruction, the server can be powered off and restarted on the premise that no other node occupies resources, all the server is powered off, a plurality of other nodes occupy resources, and under the condition that the priority of the instruction is higher, a machine cannot be powered off, the power-off demands of a user on the single node cannot be timely responded, the normal operation of other node services cannot be influenced if the power-off of the whole server is forced, the service suspension or interruption of other nodes is caused, and the problem is solved through the following embodiment.
Example 1:
as shown in fig. 1, the present invention provides a method for implementing single-node power-off by a multi-node server, comprising the following steps:
s1, after receiving a power-off restarting request of a single node, a middle backboard informs each node in a multi-node server;
and S2, after receiving the power-off agreement information returned by the service related node of the single node, the middle back plate allows the single node to independently power off from the multi-node server.
For a multi-node server, each node can represent an independent server, but other nodes of the whole server are related through the design of a plurality of hardware cables, so that the nodes are independent and are in an interconnection relationship, when a single node is required to be powered off and restarted due to service requirements or emergency requirements, the single node is required to control the power on and power off of the single node, after receiving a power off restarting instruction, the single node waits for a middle backboard to send a message informing that the other node is required to be powered off, and then the single node can independently complete power off and restarting after the service cutting of the traffic is completed in a reasonable time period.
According to the method for realizing single-node power failure of the multi-node server, provided by the invention, after the service handover between the power failure node and each node of the multi-node server is requested through the broadcasting power failure notification of the middle backboard, the power supply of the power failure node is disconnected, so that the quick response of the single-node power failure request is realized, and the requirement of a user on the single-node power failure is met under the condition that the whole service of the multi-node server is not influenced.
Example 2:
as shown in fig. 2, the present invention provides a method for implementing single-node power-off by a multi-node server, comprising the following steps:
s1, after receiving a power-off restarting request of a single node, a middle backboard informs each node in a multi-node server; the specific steps of the step S1 are as follows:
s11, a user sends a power-off restarting instruction to a single node in the multi-node server in an out-of-band or in-band mode;
s12, a single node which receives the power-off restarting instruction sends a power-off request to the middle backboard;
s13, after receiving the power-off request, the middle backboard sends a power-off notification to each node in the multi-node server;
s2, after receiving the power-off agreement information returned by the service related node of the single node, the middle back plate allows the single node to independently power off from the multi-node server; the specific steps of the step S2 are as follows:
s21, after each node in the multi-node server receives a power-off request sent by the middle backboard, judging whether a single node requesting power-off and a self node have business to and from each other;
if yes, go to step S22;
if not, returning the power-off agreement information to the middle backboard, and entering step S23;
s22, a node having service communication with a single node requesting power failure starts service handover, and after the service handover is completed, power failure agreement information is returned to the middle backboard;
s23, allowing a single node requesting power failure to perform power failure restarting by the middle backboard;
s24, the single node requesting power failure starts independent power failure in the multi-node server according to the self power supply mode.
According to the method for realizing single-node power failure of the multi-node server, provided by the invention, after the service handover between the power failure node and each node of the multi-node server is requested through the broadcasting power failure notification of the middle backboard, the power supply of the power failure node is disconnected, so that the quick response of the single-node power failure request is realized, and the requirement of a user on the single-node power failure is met under the condition that the whole service of the multi-node server is not influenced.
Example 3:
as shown in fig. 3, the present invention provides a method for implementing single-node power-off by a multi-node server, comprising the following steps:
s1, after receiving a power-off restarting request of a single node, a middle backboard informs each node in a multi-node server; the specific steps of the step S1 are as follows:
s11, a user sends a power-off restarting instruction to a single node in the multi-node server in an out-of-band or in-band mode;
s12, a single node which receives the power-off restarting instruction sends a power-off request to the middle backboard; the specific steps of step S12 are as follows:
s121, a single node which receives a power-off restarting instruction acquires a self node number;
s122, the single node which receives the power-off restarting instruction encapsulates the node number of the single node into a power-off request and sends the power-off request to the middle backboard;
s13, after receiving the power-off request, the middle backboard sends a power-off notification to each node in the multi-node server; the specific steps of step S13 are as follows:
s131, after the middle backboard receives the power-off request, analyzing the node number of a single node requesting power-off;
s132, the middle backboard generates a power-off notification according to the node number of the single node requesting power-off and sends the power-off notification to other nodes of the multi-node server;
s2, after receiving the power-off agreement information returned by the service related node of the single node, the middle back plate allows the single node to independently power off from the multi-node server; the specific steps of the step S2 are as follows:
s21, after each node in the multi-node server receives a power-off request sent by the middle backboard, judging whether a single node requesting power-off and a self node have business to and from each other;
if yes, go to step S22;
if not, returning the power-off agreement information to the middle backboard, and entering step S23;
the specific steps of step S21 are as follows:
s211, after each node of the multi-node server receives a power-off request sent by the middle backboard, analyzing the node number of a single node requesting power-off;
s212, each node in the multi-node server judges whether a single node requesting power failure and a self node have business to and from according to the node number;
if yes, go to step S22;
if not, packaging the self node number into the power-off agreement information, returning to the middle backboard, and entering step S23;
s22, a node having service communication with a single node requesting power failure starts service handover, and after the service handover is completed, power failure agreement information is returned to the middle backboard; the specific steps of step S22 are as follows:
s221, a node having service to and from a single node requesting power off starts service handover, and the service of the single node is stopped or transferred;
s222, after the service handover is completed, the service related node encapsulates the node number of the service related node into the outage information and returns the outage information to the middle backboard;
s23, allowing a single node requesting power failure to perform power failure restarting by the middle backboard; the specific steps of step S23 are as follows:
s231, the middle backboard receives the power-off agreeing information returned by each node in the multi-node server, and analyzes the node number;
s232, judging whether the number of the nodes returning the power-off agreement information is equal to the number of the nodes in the multi-node server or not by the middle back plate according to the node numbers;
if yes, go to step S235;
if not, go to step S233;
s233, the middle backboard locates the node which does not return the power-off information, and sends a power-off notification to the locating node again;
s234, the middle backboard judges whether the number of times of power failure notification sent to the positioning node exceeds a set threshold value or not;
if yes, sending out a timeout alarm, and entering step S235;
if not, returning to the step S21;
s235, the middle backboard sends a power-off permission instruction to a single node requesting power off;
s24, starting independent power failure in the multi-node server by the single node requesting power failure according to the self power supply mode; the specific steps of step S24 are as follows:
s241, a single node requesting power failure acquires a self power supply mode;
when the self power supply mode is the external power supply mode, the step S242 is entered;
when the self power supply mode is the adjacent node power supply mode, entering step S243;
s242, a single node requesting power off sends a power off instruction, cuts off the power of the main board from an external power supply, sends a power off prohibition to the main board of a neighboring node, and achieves the power off restarting of the single node and ends;
s243, the single node requesting power off sends a power off instruction, the power off main board takes power from the main board of the adjacent node, and sends a power off prohibition to take power from the main board of the adjacent node, so that the power off restarting of the single node is realized.
Example 4:
as shown in fig. 4, the present invention provides an apparatus for implementing single-node power outage of a multi-node server, including:
the power-off notification module is used for notifying each node in the multi-node server after receiving a power-off restarting request of a certain single node;
and the independent power-off module after negotiation is used for allowing the single node to independently power off from the multi-node server after receiving the power-off agreement information returned by the service related node of the single node.
According to the device for realizing single-node power failure of the multi-node server, provided by the invention, after the service handover between the power failure node and each node of the multi-node server is requested through the broadcasting power failure notification of the middle backboard, the power supply of the device is disconnected, so that the quick response of the single-node power failure request is realized, and the requirement of a user on the single-node power failure is met under the condition that the whole service of the multi-node server is not influenced.
Example 5:
as shown in fig. 4, the present invention provides an apparatus for implementing single-node power outage of a multi-node server, including:
the power-off notification module is used for notifying each node in the multi-node server after receiving a power-off restarting request of a certain single node; the power-off notification module includes:
the power-off restarting instruction sending unit is used for sending a power-off restarting instruction to a certain single node in the multi-node server in an out-of-band or in-band mode;
the power-off request sending unit is used for sending a power-off request to the middle backboard by the single node which receives the power-off restarting instruction;
the power-off notification sending unit is used for sending power-off notification to each node in the multi-node server after the middle backboard receives the power-off request;
the single-node power-off module is used for allowing the single node to independently power off from the multi-node server after receiving power-off agreement information returned by the service related node of the single node; the individual power-off module after negotiation comprises:
the service business transaction judging unit is used for judging whether a single node requesting power failure and a self node have business transaction after each node in the multi-node server receives the power failure request sent by the middle backboard;
the power-off agreement information return unit is used for enabling the single node requesting power off to have no business with the self node or enabling the single node requesting power off to have business with the self node, starting business handover, and returning power-off agreement information to the middle backboard after the business handover is completed;
the power-off restarting allowing unit is used for allowing a single node requesting power off to perform power-off restarting on the middle backboard;
and the independent power-off unit is used for enabling the single node requesting power off to start independent power off in the multi-node server according to the self power supply mode.
According to the device for realizing single-node power failure of the multi-node server, provided by the invention, after the service handover between the power failure node and each node of the multi-node server is requested through the broadcasting power failure notification of the middle backboard, the power supply of the device is disconnected, so that the quick response of the single-node power failure request is realized, and the requirement of a user on the single-node power failure is met under the condition that the whole service of the multi-node server is not influenced.
Example 6:
as shown in fig. 4, the present invention provides an apparatus for implementing single-node power outage of a multi-node server, including:
the power-off notification module is used for notifying each node in the multi-node server after receiving a power-off restarting request of a certain single node; the power-off notification module includes:
the power-off restarting instruction sending unit is used for sending a power-off restarting instruction to a certain single node in the multi-node server in an out-of-band or in-band mode;
the power-off request sending unit is used for sending a power-off request to the middle backboard by the single node which receives the power-off restarting instruction; the power-off request issuing unit includes:
the self node number acquisition subunit is used for acquiring the self node number by a single node which receives the power-off restarting instruction;
the node number packaging and power-off request sending subunit is used for packaging the self node number into the power-off request by a single node which receives the power-off restarting instruction and sending the power-off request to the middle backboard;
the power-off notification sending unit is used for sending power-off notification to each node in the multi-node server after the middle backboard receives the power-off request; the power outage notification transmission unit includes:
the node number analysis subunit is used for analyzing the node number of the single node requesting power off after the middle backboard receives the power off request;
the power-off notification broadcasting subunit is used for generating a power-off notification on the middle backboard according to the node number of the single node requesting power-off and sending the power-off notification to other nodes of the multi-node server;
the single-node power-off module is used for allowing the single node to independently power off from the multi-node server after receiving power-off agreement information returned by the service related node of the single node; the individual power-off module after negotiation comprises:
the service business transaction judging unit is used for judging whether a single node requesting power failure and a self node have business transaction after each node in the multi-node server receives the power failure request sent by the middle backboard; the business transaction judging unit comprises:
the request node number analysis subunit is used for analyzing the node number of a single node requesting power failure after each node of the multi-node server receives the power failure request sent by the middle backboard;
the service business trip judging subunit is used for judging whether a single node requesting power off and a self node have business trips or not according to the node number in the multi-node server by each node;
the non-past node agrees to return the subunit, is used for requesting the single node of the outage to have business to come and go with self node, pack self node number into agreeing to cut off the power information, and return to the middle backboard;
the power-off agreement information return unit is used for enabling the single node requesting power off to have no business with the self node or enabling the single node requesting power off to have business with the self node, starting business handover, and returning power-off agreement information to the middle backboard after the business handover is completed; the agreeing power-off information return unit includes:
a service handover promoter unit for starting service handover with a node having service to/from a single node requesting power off, and stopping or transferring the service of the single node;
the business node agrees to return the subunit, is used for after the business handing-over is finished, the node number of the business looks node encapsulates and agrees to cut off the power information, and returns to the middle backboard;
the power-off restarting allowing unit is used for allowing a single node requesting power off to perform power-off restarting on the middle backboard; the power-off restart allowing unit includes:
the consent node number analysis subunit is used for receiving consent power-off information returned by each node in the multi-node server at the middle backboard and analyzing out the node number;
the consent node quantity judging subunit is used for judging whether the quantity of the nodes returning consent power-off information is equal to the quantity of the nodes in the multi-node server or not according to the node number in the middle back plate;
the power-off notification resending subunit is used for locating the node which does not return the power-off information in the middle backboard and resending the power-off notification to the locating node when the number of the nodes which return the power-off information is unequal to the number of the nodes in the multi-node server;
the power-off notification sending frequency subunit is used for judging whether the power-off notification sending frequency exceeds a set threshold value or not to the positioning node in the middle backboard;
the timeout alarm sending subunit is used for sending timeout alarms when the power-off notification times to the positioning node exceed a set threshold value;
the power-off permission instruction sending subunit is used for sending a power-off permission instruction to a single node requesting power off at the middle backboard when the number of nodes for returning the power-off permission information is equal to the number of nodes in the multi-node server;
the independent power-off unit is used for starting independent power-off in the multi-node server according to the self power-on mode by the single node requesting power-off; the individual power-off unit includes:
the self power supply module acquisition subunit is used for acquiring a self power supply mode at a single node requesting power failure;
the external power supply cut-off subunit is used for sending a power supply cut-off instruction to a single node which requests power off when the self power supply mode is the external power supply mode, cutting off the power supply of the main board from the external power supply, and sending a power supply prohibition instruction to prohibit the power supply of the main board of the adjacent node, so that the power off restarting of the single node is realized;
and the adjacent node main board power supply cutting-off subunit is used for sending a power supply cutting-off instruction to a single node requesting power off when the self power supply mode is the adjacent node power supply taking mode, cutting off the power supply of the main board from the adjacent node main board, and sending a power supply prohibiting power supply from the main board of the adjacent node to realize the power off restarting of the single node.
Example 7:
the invention provides a device comprising a processor and a memory;
wherein the memory is used for storing a computer program and the processor is used for calling and running the computer program from the memory, so that the device executes the method described in embodiment 1, embodiment 2 or embodiment 3.
Example 8:
the present invention provides a storage medium that is configured to store,
the storage medium has instructions stored therein that, when executed on a computer, cause the computer to perform the method described in embodiment 1, embodiment 2 or embodiment 3 above.
Although the present invention has been described in detail by way of preferred embodiments with reference to the accompanying drawings, the present invention is not limited thereto. Various equivalent modifications and substitutions may be made in the embodiments of the present invention by those skilled in the art without departing from the spirit and scope of the present invention, and it is intended that all such modifications and substitutions be within the scope of the present invention/be within the scope of the present invention as defined by the appended claims. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A method for realizing single-node power-off of a multi-node server is characterized by comprising the following steps:
s1, after receiving a power-off restarting request of a single node, a middle backboard informs each node in a multi-node server;
s2, after receiving the power-off agreement information returned by the service related node of the single node, the middle back plate allows the single node to independently power off from the multi-node server.
2. The method for implementing single-node power-off by the multi-node server according to claim 1, wherein the step S1 comprises the following specific steps:
s11, a user sends a power-off restarting instruction to a single node in the multi-node server in an out-of-band or in-band mode;
s12, a single node which receives a power-off restarting instruction sends a power-off request to the middle backboard;
s13, after receiving the power-off request, the middle backboard sends power-off notification to each node in the multi-node server.
3. The method for implementing single-node power-off by the multi-node server according to claim 2, wherein the step S12 comprises the following specific steps:
s121, a single node which receives a power-off restarting instruction acquires a self node number;
s122, the single node which receives the power-off restarting instruction encapsulates the node number of the single node into a power-off request and sends the power-off request to the middle backboard.
4. The method for implementing single-node power-off by the multi-node server as claimed in claim 3, wherein the step S13 comprises the following specific steps:
s131, after the middle backboard receives the power-off request, analyzing the node number of a single node requesting power-off;
s132, the middle backboard generates a power-off notification according to the node number of the single node requesting power-off, and sends the power-off notification to other nodes of the multi-node server.
5. The method for implementing single-node power-off of multi-node server according to claim 4, wherein step S2 comprises the following specific steps:
s21, after each node in the multi-node server receives a power-off request sent by the middle backboard, judging whether a single node requesting power-off and a self node have business to and from each other;
if yes, go to step S22;
if not, returning the power-off agreement information to the middle backboard, and entering step S23;
s22, a node having service communication with a single node requesting power off starts service handover, and after the service handover is completed, power off agreement information is returned to the middle backboard;
s23, allowing a single node requesting power failure to perform power failure restarting by the middle backboard;
s24, the single node requesting power failure starts independent power failure in the multi-node server according to the self power supply mode.
6. The method for implementing single-node power-off of multi-node server according to claim 5, wherein step S21 comprises the following specific steps:
s211, after each node of the multi-node server receives a power-off request sent by the middle backboard, analyzing the node number of a single node requesting power-off;
s212, each node in the multi-node server judges whether a single node requesting power failure and a self node have business to and from according to the node number;
if yes, go to step S22;
if not, the self node number is packaged into the power-off agreement information, and returns to the middle backboard, and the step S23 is entered.
7. The method for implementing single-node power-off of multi-node server according to claim 5, wherein step S22 comprises the following specific steps:
s221, a node with service to/from a single node requesting power off starts service handover, and service of the single node is stopped or transferred;
s222, after the service handover is completed, the service related node encapsulates the node number of the service related node into the outage information and returns the outage information to the middle backboard.
8. The method for implementing single-node power-off of multi-node server according to claim 7, wherein step S23 comprises the following specific steps:
s231, the middle backboard receives the power-off agreeing information returned by each node in the multi-node server, and analyzes the node number;
s232, judging whether the number of the nodes returning the power-off agreement information is equal to the number of the nodes in the multi-node server or not by the middle back plate according to the node numbers;
if yes, go to step S235;
if not, go to step S233;
s233, positioning a node which does not return the power-off agreement information by the middle backboard, and sending a power-off notification to the positioning node again;
s234, the middle backboard judges whether the number of times of power-off notification sent to the positioning node exceeds a set threshold value or not;
if yes, sending out a timeout alarm, and entering step S235;
if not, returning to the step S21;
s235, the middle backboard sends a power-off permission instruction to a single node requesting power off.
9. The method for implementing single-node power-off of multi-node server according to claim 5, wherein step S24 comprises the following specific steps:
s241, a single node requesting power failure obtains a self power supply mode;
when the self power supply mode is the external power supply mode, the step S242 is entered;
when the self power supply mode is the adjacent node power supply mode, entering step S243;
s242, a single node requesting power off sends a power-off instruction, a main board is cut off to take power from an external power supply, and the main board of a neighboring node is sent to be forbidden to take power, so that the power off of the single node is restarted and ended;
s243, sending a power-off instruction by the single node requesting power off, cutting off the power of the main board from the main board of the adjacent node, and sending a power-off prohibition instruction to prohibit the power of the main board of the adjacent node, so as to realize the power-off restarting of the single node.
10. An apparatus for implementing single-node power outage by a multi-node server, comprising:
the power-off notification module is used for notifying each node in the multi-node server after receiving a power-off restarting request of a certain single node;
and the independent power-off module after negotiation is used for allowing the single node to independently power off from the multi-node server after receiving the power-off agreement information returned by the service related node of the single node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310504722.XA CN116755533A (en) | 2023-05-06 | 2023-05-06 | Method and device for realizing single-node power failure of multi-node server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310504722.XA CN116755533A (en) | 2023-05-06 | 2023-05-06 | Method and device for realizing single-node power failure of multi-node server |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116755533A true CN116755533A (en) | 2023-09-15 |
Family
ID=87946738
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310504722.XA Pending CN116755533A (en) | 2023-05-06 | 2023-05-06 | Method and device for realizing single-node power failure of multi-node server |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116755533A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117666746A (en) * | 2024-01-30 | 2024-03-08 | 苏州元脑智能科技有限公司 | Multi-node server, method, device and medium applied to multi-node server |
-
2023
- 2023-05-06 CN CN202310504722.XA patent/CN116755533A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117666746A (en) * | 2024-01-30 | 2024-03-08 | 苏州元脑智能科技有限公司 | Multi-node server, method, device and medium applied to multi-node server |
| CN117666746B (en) * | 2024-01-30 | 2024-04-05 | 苏州元脑智能科技有限公司 | Multi-node server, method, device and medium applied to multi-node server |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103338473B (en) | The treatment facility of cell outage | |
| CN103199972A (en) | Double machine warm backup switching method and warm backup system achieved based on SOA and RS485 bus | |
| CN116755533A (en) | Method and device for realizing single-node power failure of multi-node server | |
| CN106170948A (en) | A kind of referee method for dual-active data center, Apparatus and system | |
| CN110704250B (en) | Hot backup device of distributed system | |
| CN113067782A (en) | High-reliability electric energy acquisition and transmission system based on redundancy system | |
| CN110958151B (en) | Keep-alive detection method, keep-alive detection device, node, storage medium and communication system | |
| CN109327383A (en) | A kind of fault handling method and equipment | |
| CN103262470A (en) | FCoE network linking management method, device and system | |
| US20120284553A1 (en) | Method, apparatus and system for reducing power consumption of service system | |
| CN111262745A (en) | Information processing platform redundancy system design | |
| CN116155718A (en) | Side data interaction method oriented to distributed resource regulation | |
| JPH09247194A (en) | Lan control system | |
| WO2021254466A1 (en) | Method, apparatus and system for configuring edge side device | |
| US20160119871A1 (en) | Method and system for waking sleeping trunking terminal up | |
| CN111224803B (en) | Multi-master detection method in stacking system and stacking system | |
| CN114598593A (en) | Message processing method, system, computing device and computer storage medium | |
| JP2003018165A (en) | Network information informing method | |
| CN110087333B (en) | Method, device, equipment and storage medium for preventing service interruption | |
| JP3966118B2 (en) | Redundant crossbar switch control method with credit-based flow control and recording medium thereof | |
| Kitamura | Configuration of a Power-saving High-availability Server System Incorporating a Hybrid Operation Method | |
| CN114258155B (en) | Network connection control method and device, main base station and auxiliary base station | |
| CN112910547B (en) | Space-based network data security processing center device | |
| CN103888986A (en) | Method and device for processing cell failure | |
| CN110875822B (en) | Power supply system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |