CN111200513A - Server starting method, server and cluster server system - Google Patents
Server starting method, server and cluster server system Download PDFInfo
- Publication number
- CN111200513A CN111200513A CN201811384397.3A CN201811384397A CN111200513A CN 111200513 A CN111200513 A CN 111200513A CN 201811384397 A CN201811384397 A CN 201811384397A CN 111200513 A CN111200513 A CN 111200513A
- Authority
- CN
- China
- Prior art keywords
- reset
- cluster
- cluster node
- address information
- receiving
- 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.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
- H04L41/0813—Configuration setting characterised by the conditions triggering a change of settings
- H04L41/082—Configuration setting characterised by the conditions triggering a change of settings the condition being updates or upgrades of network functionality
Abstract
The embodiment of the application provides a server starting method, a server and a cluster server system, after the cluster server system is powered on, a cluster master controller outputs a holding reset signal to each cluster node by default, and the cluster nodes keep a reset state when the holding reset signal is input; the cluster master control selects cluster nodes to be released from reset according to a preset sequence and outputs a reset releasing signal to the cluster nodes; after receiving the reset release signal, the cluster node starts to operate, completes configuration updating of the address information and returns configuration completion information to the cluster master control; and after receiving the returned configuration completion information, the cluster master control selects and sends a release reset signal to a next cluster node to be released and reset according to a preset sequence. The address allocation in the cluster starting process is realized by releasing reset and software interactive design one by one, and the expandability of the cluster server is enhanced without depending on a hardware design index input mode.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a server starting method, a server, and a cluster server system.
Background
In the existing network cluster server system, the interior of the system connects cluster nodes through two layers of network exchange, the exterior of the system realizes external service access in the form of a gateway, and the cluster nodes in the cluster server system perform data exchange with the exterior through the gateway, for example, the cluster nodes acquire data from the exterior and realize operation, return operation results, or process service requests distributed by the gateway. The overall operational capacity of the cluster server depends on the operational capacity and the concurrency number of the internal cluster nodes, and the cluster nodes and the gateways are managed through cluster master control, or the functions of the gateways are realized through cluster master control agents.
In the cluster server, to implement unified management, the operation resources of the cluster nodes, such as kernel system, rootfs, application, etc., are stored in the cluster master, and in the starting process of the cluster server, the cluster nodes are downloaded from the cluster master to the local through the network at the bootloader stage, so as to start the operation resources.
In an existing cluster server system, an index number of a cluster node is determined by a hardware design GPIO (General Purpose Input/Output) Input level value, an IP (Internet Protocol) address and an MAC (Media Access Control) address of the cluster node are set by the index number at a bootloader stage, and then the cluster node communicates with a cluster master controller to further obtain an operation resource and then operates.
However, in the above method, in order to distinguish the index numbers of different cluster nodes, a corresponding number of number pins need to be configured. With the increase of the number of cluster nodes, the number of required numbering pins is correspondingly increased, which causes poor expandability of the cluster server system.
Disclosure of Invention
An object of the embodiments of the present application is to provide a server starting method, a server, and a cluster server system, so as to increase scalability of the cluster server system. The specific technical scheme is as follows:
in a first aspect, an embodiment of the present application provides a server startup method, which is applied to a cluster server system, where the cluster server system includes a cluster master and a plurality of cluster nodes, and the method includes:
the cluster master controller sends a reset signal to each cluster node;
after receiving the reset signal, the cluster node keeps a reset state;
the cluster master control determines a target cluster node to be released and reset currently according to a preset release sequence and sends a release reset signal to the target cluster node;
after receiving the reset releasing signal, the target cluster node releases reset to start running and updates the configuration of address information; after completing the configuration and updating of the address information, the target cluster node returns a configuration completion message to the cluster master control;
and after receiving the configuration completion message of the target cluster node, the cluster master determines the cluster node to be released and reset next according to the preset release sequence, and sends a release reset signal to the cluster node to be released and reset next.
Optionally, in the server starting method in the embodiment of the present application, the cluster server system further includes: and the cluster master controller communicates with each cluster node and each cluster node through the internal Ethernet switching module.
Optionally, after receiving the reset release signal, the target cluster node releases reset and starts operation, and performs configuration update of address information, including:
after receiving the reset releasing signal, the target cluster node releases reset to start operation, performs network initialization, and sends a broadcast message by default address information;
after receiving a broadcast message sent by default address information, the cluster master configures new address information for the target cluster node and sends the new address information to the target cluster node;
and after receiving the new address information, the target cluster node configures the address of the target cluster node as the new address information to complete configuration and updating of the address information.
Optionally, after receiving the reset release signal, the target cluster node releases reset to start operation, performs network initialization, and sends a broadcast message with default address information, including:
and after receiving the reset release signal, the target cluster node releases the reset, loads a boot loader BootLoader, performs network initialization in the running process of the BootLoader, and sends an Address Resolution Protocol (ARP) broadcast message by default address information.
Optionally, after the target cluster node receives the new address information, the address of the target cluster node is configured as the new address information, so as to complete configuration update of address information, and the method further includes:
after the target cluster node completes the configuration updating of the address information, the target cluster node sends a broadcast message by the updated address information;
after receiving the broadcast message sent by the updated address information, the cluster master sends a unicast reply message to the target cluster node;
and after the target cluster node receives the unicast reply message, finishing firmware loading through a simple file transfer protocol (TFTP).
In a second aspect, an embodiment of the present application provides a cluster server system, where the system includes:
the system comprises a cluster master controller and a plurality of cluster nodes, wherein the cluster master controller is in communication connection with each cluster node;
the cluster master control: for sending a hold reset signal to each of said cluster nodes; determining a target cluster node to be released and reset currently according to a preset release sequence, and sending a release reset signal to the target cluster node; after configuration completion information sent by the target cluster node is received, determining a cluster node to be reset next to be released according to the preset release sequence, and sending a release reset signal to the cluster node to be reset next to be released;
the cluster node: the reset circuit is used for keeping a reset state after receiving the reset keeping signal; and after receiving the reset releasing signal, releasing reset starting operation, updating the configuration of the address information, and after finishing the updating of the configuration of the address information, returning configuration finishing information to the cluster master control.
Optionally, the cluster server system further includes: and the internal Ethernet switching module is used for communicating between the cluster master controller and each cluster node and between each cluster node.
Optionally, after receiving the reset release signal, releasing reset to start operation, and performing configuration update of address information, includes:
after receiving the reset releasing signal, releasing reset to start operation, performing network initialization, and sending a broadcast message by default address information; after receiving new address information returned by the cluster master controller, configuring an address into the new address information to complete configuration and updating of the address information;
the cluster master control: and the system is further configured to determine new address information for the target cluster node after receiving a broadcast message sent by the target cluster node with default address information, and send the new address information to the target cluster node.
Optionally, after receiving the reset release signal, the target cluster node releases reset to start operation, performs network initialization, and sends a broadcast message with default address information, including:
and after receiving the reset release signal, the target cluster node releases the reset, loads a boot loader BootLoader, performs network initialization in the running process of the BootLoader, and sends an Address Resolution Protocol (ARP) broadcast message by default address information.
Optionally, the cluster node: the system is also used for sending the broadcast message by the updated address information after the address information configuration is updated; after receiving the unicast reply message returned by the cluster master control, completing firmware loading through a simple file transfer protocol (TFTP);
the cluster master control: and the unicast reply message is also sent to the target cluster node after the broadcast message sent by the target cluster node with the updated address information is received.
In a third aspect, an embodiment of the present application provides a server, where the server includes:
the system comprises a cluster master controller and a plurality of cluster nodes, wherein the cluster master controller is in communication connection with each cluster node;
the cluster master control: for sending a hold reset signal to each of said cluster nodes; determining a target cluster node to be released and reset currently according to a preset release sequence, and sending a release reset signal to the target cluster node; after configuration completion information sent by the target cluster node is received, determining a cluster node to be reset next to be released according to the preset release sequence, and sending a release reset signal to the cluster node to be reset next to be released;
the cluster node: the reset circuit is used for keeping a reset state after receiving the reset keeping signal; and after receiving the reset releasing signal, releasing reset starting operation, updating the configuration of the address information, and after finishing the updating of the configuration of the address information, returning configuration finishing information to the cluster master control.
Optionally, the server further includes: and the internal Ethernet exchange module is used for communicating between the cluster master control and each cluster node and between each cluster node through the internal Ethernet interaction module.
Optionally, after receiving the reset release signal, the target cluster node releases reset to start operation, and performs configuration update of address information, including:
after receiving the reset releasing signal, releasing reset to start operation, performing network initialization, and sending a broadcast message by default address information; after receiving new address information returned by the cluster master controller, configuring an address into the new address information to complete configuration and updating of the address information;
the cluster master control: and the system is further configured to determine new address information for the target cluster node after receiving a broadcast message sent by the target cluster node with default address information, and send the new address information to the target cluster node.
Optionally, after receiving the reset release signal, the target cluster node releases reset to start operation, performs network initialization, and sends a broadcast message with default address information, including:
and after receiving the reset releasing signal, the target cluster node releases reset and starts to operate, loads a boot loader BootLoader, performs network initialization in the boot loader operation process, and sends an Address Resolution Protocol (ARP) broadcast message by default address information.
Optionally, the cluster node: the system is also used for sending the broadcast message by the updated address information after the address information configuration is updated; after receiving the unicast reply message returned by the cluster master control, completing firmware loading through a simple file transfer protocol (TFTP);
the cluster master control: and the unicast reply message is also sent to the target cluster node after the broadcast message sent by the target cluster node with the updated address information is received.
According to the server starting method, the server and the cluster server system, the cluster master control sends the reset signal to each cluster node; after receiving the reset signal, the cluster node keeps a reset state; the cluster master control determines a target cluster node to be released and reset currently according to a preset release sequence and sends a release reset signal to the target cluster node; after receiving the reset releasing signal, the target cluster node releases reset to start running and updates the configuration of the address information; after completing the configuration and updating of the address information, the target cluster node returns a configuration completion message to the cluster master control; after receiving the configuration completion message of the target cluster node, the cluster master determines the next cluster node to be released and reset according to a preset release sequence, and sends a release reset signal to the next cluster node to be released and reset. And releasing and resetting each cluster node one by one according to a preset release sequence without using a hardware GPIO pin input level value as a configuration reference of address information of the cluster node, thereby enhancing the expandability of the cluster server system. The address information of each cluster node is configured one by one, and the problem of conflict of IP addresses and MAC addresses in the starting process of the cluster nodes is solved. Of course, not all advantages described above need to be achieved at the same time in the practice of any one product or method of the present application.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a first flowchart illustrating a server startup method according to an embodiment of the present application;
fig. 2 is a first schematic diagram of a cluster server system according to an embodiment of the present application;
fig. 3 is a second flowchart of a server startup method according to an embodiment of the present application;
fig. 4 is a third flowchart illustrating a server startup method according to an embodiment of the present application;
fig. 5 is a second schematic diagram of a cluster server system according to an embodiment of the present application;
fig. 6 is a third schematic diagram of a cluster server system according to an embodiment of the present application;
fig. 7 is a schematic diagram of a server according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In the existing cluster server system, in order to distinguish index numbers of different cluster nodes, a corresponding number of number pins need to be configured, and taking a cluster server system of 16 cluster nodes as an example, GPIO input levels of 4 number pins are needed to represent index values of 16 cluster nodes. With the increase of the number of cluster nodes, the number of required numbering pins is correspondingly increased, which causes poor expandability of the cluster server system.
In view of this, an embodiment of the present application provides a server starting method, referring to fig. 1, which is applied to a server side, where the server side includes a cluster master and a plurality of cluster nodes, and the method includes:
s101, the cluster master sends a hold reset signal to each cluster node.
The server side can be an independent server or a cluster server system consisting of a plurality of servers. After the server terminal is powered on, the cluster master sends a hold reset signal to each cluster node, for example, the cluster master sends a GPIO high level to each cluster node through a reset pin. The cluster master control can be an independent server or a module in the server, and meanwhile, the cluster master control can also be an integrated gateway function.
And S102, after receiving the reset maintaining signal, the cluster node maintains a reset state.
The cluster node may be an independent server or a module in the server. The cluster node maintains the reset state after receiving the hold reset signal, for example, after the reset pin of the cluster node receives the GPIO high level, the cluster node maintains the reset state. Wherein the cluster node in the reset-holding state has not entered the run state.
S103, the cluster master control determines a current target cluster node to be released and reset according to a preset release sequence, and sends a release reset signal to the target cluster node.
The preset release order may be set according to actual conditions, for example, according to the index number of each cluster node. The cluster master determines a cluster node of the service to be currently released, namely a target cluster node, according to a preset release sequence, and sends a release reset signal to the target cluster node, for example, the cluster master sends a GPIO low level to the cluster node of the service to be currently released through a reset pin.
S104, after receiving the reset releasing signal, the target cluster node releases reset to start running and updates the configuration of the address information; and after the address information configuration updating is completed, the target cluster node returns a completion message to the cluster master control.
After the target cluster node receives the reset release signal, for example, after the reset pin of the target cluster node receives the GPIO low level, the reset is released, and the target cluster node starts to operate into an operating state, and updates its own address information, including the MAC address information and the IP address information, according to the preset flow configuration. For example, default address information is configured for each cluster node in advance, after a target cluster node is started to operate, the target cluster node interacts with a cluster master according to the default address information configured for the target cluster node, new address information is distributed and returned by the cluster master, and the cluster node receives feedback of the cluster master. And finishing the configuration updating of the address information, and returning the finishing information of the new address configuration to the cluster master control by the target cluster node.
Optionally, after receiving the reset release signal, the target cluster node releases reset to start operation, and performs configuration update of address information, including:
step one, after receiving the reset releasing signal, the target cluster node releases reset to start operation, performs network initialization, and sends a broadcast message by default address information.
Optionally, after receiving the reset release signal, the target cluster node releases reset to start operation, performs network initialization, and sends a broadcast message with default address information, where the method includes:
after receiving the reset release signal, the target cluster node releases the reset, loads a BootLoader (boot loader), performs network initialization during the operation of the BootLoader, and sends an Address Resolution Protocol (ARP) broadcast message by using default Address information.
And step two, after receiving the broadcast message sent by the target cluster node by the default address information, the cluster master configures new address information for the target cluster node and sends the new address information to the target cluster node.
After receiving the broadcast message sent by the target cluster node by the default address information, the cluster master configures new address information for the target cluster node, for example, the cluster master acquires an unused intranet address from the address resource pool as new address information of the target cluster node, and sends the new address information to the target cluster node.
And step three, after the target cluster node receives the new address information, configuring the address of the target cluster node as the new address information so as to complete the configuration and updating of the address information.
And after receiving the new address information sent by the cluster master control, the target cluster node changes the address of the target cluster node from the default address information to the new address information, thereby completing the configuration update of the address information.
In the embodiment, the cluster master controller configures the address information for each cluster node, so that the unified management and utilization of address resources are facilitated.
And S105, after receiving the configuration completion message sent by the target cluster node, the cluster master determines the next cluster node to be released and reset according to the preset release sequence, and sends a release reset signal to the next cluster node to be released and reset.
After receiving configuration completion information sent by a target cluster node which has released reset currently, the cluster master determines a cluster node to be released and reset next according to a preset release sequence, and sends a release reset signal to the cluster node to be released and reset next. And after receiving the configuration completion information sent by the cluster node which is started to operate next, the cluster master control continues to determine the cluster node which is to be released and reset next according to a preset release sequence until the completion of the address information configuration update of all the cluster nodes.
Each cluster node is provided with a reset signal input from a cluster master controller, after a cluster server is powered on, the cluster master controller outputs a holding reset signal to each cluster node in a default mode, and the cluster nodes are kept in a reset state when the holding reset signal is input; the cluster master control selects cluster nodes to be released from reset according to a preset sequence and outputs a reset releasing signal to the selected cluster nodes; the selected cluster node starts to operate after receiving the reset releasing signal, completes the configuration updating of the address information through the interaction of the network and the master control, and returns the address information after the configuration updating to the cluster master control after the configuration updating of the address information is completed; after receiving the address information, the cluster master controller selects the next cluster node to be released and reset according to a preset sequence, and sends a release reset signal to the next cluster node to be released and reset.
In the embodiment of the application, the cluster nodes are released and reset one by one according to the preset release sequence, and the input level value of the pin of the GPIO does not need to be used as the configuration reference of the address information of the cluster nodes, so that the expandability of a cluster server system/server is enhanced. The address information of each cluster node is configured one by one, and the problem of conflict of IP addresses and MAC addresses in the starting process of the cluster nodes is solved.
Optionally, the server further includes: and the internal Ethernet switching module is used for communicating between the cluster master controller and each cluster node and between each cluster node through the internal Ethernet switching module.
The server side further comprises an internal Ethernet switching module, the internal Ethernet switching module can be constructed through a two-layer network switching chip, the cluster main control is communicated with all cluster nodes through the internal Ethernet switching module, communication can be conducted among all the cluster nodes through the internal Ethernet switching module, and the expandability of the server side is improved.
As shown in fig. 2, 8 cluster nodes are taken as an example for illustration. The cluster master control and the cluster nodes are interconnected through two-layer network exchange, an uplink port of the exchange is connected with the cluster master control, an internal cluster node actively communicates with an external network through a software gateway function of the cluster master control to obtain data to be analyzed and processed and returns a processing result, the cluster master control can also passively send a network request from the outside (through a software gateway) to a specific cluster node for processing and returns a processing result, the internal two-layer network communication is realized by an internal Ethernet exchange module, and the cluster master control realizes the functions of management and gateway routing.
After the server terminal is powered on, the cluster master control sends a reset maintaining signal to maintain 8 cluster nodes in a reset state, for example, a cluster master control CPU uses 8 GPIOs as reset inputs to the 8 cluster nodes to maintain the GPIO outputs at a high level, the 8 cluster nodes do not enter an operating state in the reset state, and after the reset is released, that is, the GPIO outputs a low level, each cluster node receiving the low level starts to operate.
The use of the pipeline mode for releasing the reset by each cluster node is shown in fig. 3. The method comprises the steps that a cluster master control is in a state of keeping cluster nodes reset at the beginning of electrification, firstly, reset is released for the cluster node #1, the cluster node #1 starts to operate, after the cluster node #1 operates to a bootloader stage, the cluster node #1 enters an IP and MAC redistribution process after completing initialization of a network, the cluster node #1 updates and configures network parameters of the cluster node #1 after acquiring new IP and MAC addresses, after configuration is completed, the cluster node #1 communicates with the cluster master control through the new IP and MAC, firmware resources required by operation, such as Kernel, Rootfs and Application, are downloaded, and then starting operation of firmware is completed. Once the configuration of the IP and MAC of the cluster node #1 is completed, the cluster node #2 is released and reset one by one, then the configuration of the IP and MAC of the cluster node #2, the loading of the firmware network, the starting and the running of the firmware are completed, similarly, after the configuration of the IP and MAC of the cluster node #2 is completed, the cluster node #3 is released and reset, and so on until the starting and the running of the cluster node #8 are completed.
Optionally, after the target cluster node receives the new address information, the address of the target cluster node is configured as the new address information, so as to complete configuration and update of address information, where the method further includes:
step one, after the target cluster node completes the address information configuration updating, the target cluster node sends a broadcast message by the updated address information.
And the running cluster nodes are started, and the ARP broadcast message is sent by the updated IP address and the MAC address.
And step two, after receiving the broadcast message sent by the updated address information, the cluster master controller sends a unicast reply message to the target cluster node.
After receiving the ARP broadcast message sent by the target cluster node which is released and reset before and the updated IP address and MAC address, the cluster master returns a reply message to the current cluster node to be released and reset.
And step three, after the target cluster node receives the unicast reply message, finishing firmware loading through a TFTP protocol.
After the cluster node which has released the reset currently receives the unicast reply message, the cluster node which is to be released the reset currently completes the firmware loading through a TFTP (simple File Transfer Protocol) Protocol.
In the embodiment of the application, the cluster node sends the broadcast message with the updated address information, the cluster master controller returns the reply message according to the broadcast message, and the cluster node finishes firmware loading through the TFTP protocol after receiving the unicast reply message.
The server starting method according to the embodiment of the present application may also be as shown in fig. 4. After the server is powered on, the cluster master control keeps each cluster node in a reset state, prepares the firmware resource of each cluster node in the starting process of the cluster master control, starts the TFTP service, and then sequentially releases the reset of each cluster node according to a preset release sequence according to a pipeline type reset mode shown in FIG. 3. The reset is first released to the first cluster node. Once the cluster node is released and reset, the CPU of the cluster node starts to operate a bootloader, further performs initialization of corresponding network and other peripherals, further enters a network loading process of firmware resources, utilizes the characteristics of an ARP protocol in a firmware loading process, acquires new IP and MAC addresses allocated to the cluster node by a cluster master controller after ARP interaction, immediately updates the effective setting of the IP and MAC addresses of the cluster node, and then performs ARP interaction again by using the new IP and MAC addresses for informing that the IP and MAC of the cluster master controller are configured completely so that the cluster master controller performs release and reset of the next cluster node. And the subsequent cluster node carries out subsequent firmware loading, starting and external network communication by using the new IP and MAC address, and the next cluster node interacts with the cluster master control according to the same flow to complete the setting of IP and MAC and the firmware loading operation until all the cluster nodes of the whole server end are started.
In the embodiment of the application, the problem of IP and MAC address allocation in the starting of cluster nodes is solved by a pipeline type reset design of each cluster node in a cluster server system/server and combining the configuration process of IP and MAC, the original hardware level solidification mode is changed into the interactive distribution of cluster master control, the original starting process of the cluster nodes is integrated in a software layer, the flexibility is higher, the dependency on hardware design is reduced, the pipeline type reset isolation fundamentally avoids the IP and MAC conflict problem in the starting process of the cluster nodes, the most basic ARP protocol of a two-layer network is utilized for improvement, the design is simple and reliable, the influence on other upper-layer protocols is reduced, and the application is higher.
An embodiment of the present application provides a cluster server system, referring to fig. 5, where the system includes:
a cluster master 501 and a plurality of cluster nodes 502, wherein the cluster master 501 is in communication connection with each cluster node 502;
the above cluster master 501: the cluster node is used for sending a holding reset signal to each cluster node; determining a target cluster node to be released and reset currently according to a preset release sequence, and sending a release reset signal to the target cluster node; after configuration completion information sent by the target cluster node is received, determining a cluster node to be released and reset next according to the preset release sequence, and sending a release reset signal to the cluster node to be released and reset next;
the above cluster node 502: the reset circuit is used for keeping a reset state after receiving the reset keeping signal; and after the address information configuration updating is completed, returning configuration completion information to the cluster master control.
In this embodiment, the cluster master 501 and the cluster nodes 502 may be independent servers respectively, and release and reset the cluster nodes one by one according to a preset release sequence, without inputting a certain number of GPIO pins through hardware to index the cluster nodes, thereby enhancing the scalability of the cluster server system. The address information of each cluster node is configured one by one, and the problem of conflict of IP addresses and MAC addresses in the starting process of the cluster nodes is solved.
Optionally, referring to fig. 6, in the cluster server system according to the embodiment of the present application, the cluster server system further includes: an internal ethernet switch module 503, through which the cluster master 501 communicates with each cluster node 502 and each cluster node 502 via the ethernet switch module 503. In this embodiment, since the cluster master 501 and the cluster node 502 are independent servers, the internal ethernet switch module 503 may be an independent ethernet switch.
Optionally, in the cluster server system in this embodiment of the present application, after receiving the release reset signal, the releasing reset starts operation, and performs configuration update of address information, where the releasing reset signal includes:
after receiving the reset releasing signal, releasing reset to start operation, carrying out network initialization, and sending a broadcast message by default address information; after receiving new address information returned by the cluster master, configuring the address into the new address information to complete configuration and updating of the address information;
the above cluster master 501: and the system is further configured to determine new address information for the target cluster node after receiving the broadcast message sent by the target cluster node as default address information, and send the new address information to the target cluster node.
Optionally, in the cluster server system in this embodiment of the present application, after receiving the reset release signal, the target cluster node releases reset to start operation, performs network initialization, and sends a broadcast message using default address information, where the method includes:
and after receiving the reset releasing signal, the target cluster node releases the reset, loads a BootLoader, performs network initialization in the operating process of the BootLoader, and sends an ARP broadcast message by default address information.
Optionally, in the cluster server system in the embodiment of the present application, the cluster node 502: and is also used for: the system is also used for sending the broadcast message by the updated address information after the address information configuration is updated; after receiving the unicast reply message returned by the cluster master control, finishing firmware loading through a simple file transfer protocol (TFTP);
the above cluster master 501: and the unicast reply message is sent to the target cluster node after the broadcast message sent by the target cluster node with the updated address information is received.
An embodiment of the present application further provides a server, referring to fig. 7, where the server includes:
the system comprises a cluster master controller and a plurality of cluster nodes, wherein the cluster master controller is in communication connection with each cluster node;
the above cluster master control: the cluster node is used for sending a holding reset signal to each cluster node; determining a target cluster node to be released and reset currently according to a preset release sequence, and sending a release reset signal to the target cluster node; after configuration completion information sent by the target cluster node is received, determining a cluster node to be released and reset next according to the preset release sequence, and sending a release reset signal to the cluster node to be released and reset next;
the cluster node comprises: the reset circuit is used for keeping a reset state after receiving the reset keeping signal; and after the address information configuration updating is completed, returning configuration completion information to the cluster master control.
In this embodiment, the cluster master 501 and the cluster node 502 may be multiple modules in the same server. And releasing and resetting all cluster nodes one by one according to a preset release sequence without inputting a certain number of GPIO pins through hardware to index the cluster nodes, thereby enhancing the expandability of the cluster server system. The address information of each cluster node is configured one by one, and the problem of conflict of IP addresses and MAC addresses in the starting process of the cluster nodes is solved.
Optionally, the server further includes: and the internal Ethernet exchange module is used for communicating between the cluster master control and each cluster node and between each cluster node through the internal Ethernet exchange module. In this embodiment, since the cluster master and the cluster nodes are respectively a plurality of modules in the same server, the internal ethernet switching module is also one module in the server, and may be constructed by a two-layer network switching chip.
Optionally, after receiving the reset release signal, the target cluster node releases reset to start operation, and performs configuration update of address information, including:
after receiving the reset releasing signal, releasing reset to start operation, carrying out network initialization, and sending a broadcast message by default address information; after receiving new address information returned by the cluster master, configuring the address into the new address information to complete configuration and updating of the address information;
the above cluster master control: and the system is further configured to determine new address information for the target cluster node after receiving the broadcast message sent by the target cluster node as default address information, and send the new address information to the target cluster node.
Optionally, after receiving the reset release signal, the target cluster node releases reset to start operation, performs network initialization, and sends a broadcast message with default address information, where the method includes:
and after receiving the reset releasing signal, the target cluster node releases the reset to start running, loads a boot loader BootLoader, performs network initialization in the running process of the BootLoader, and sends an address resolution protocol ARP broadcast message by default address information.
Optionally, the cluster node: the system is also used for sending the broadcast message by the updated address information after the address information configuration is updated; after receiving the unicast reply message returned by the cluster master control, completing firmware loading through a simple file transfer protocol (TFTP);
the above cluster master control: and the unicast reply message is sent to the target cluster node after the broadcast message sent by the target cluster node with the updated address information is received.
The cluster master control may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; the cluster node may have a special function arithmetic unit.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the embodiments of the apparatus, the electronic device, and the storage medium, since they are substantially similar to the method embodiments, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiments.
The above description is only for the preferred embodiment of the present application, and is not intended to limit the scope of the present application. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application are included in the protection scope of the present application.
Claims (11)
1. A server starting method is applied to a server side, the server side comprises a cluster main control and a plurality of cluster nodes, and the method comprises the following steps:
the cluster master controller sends a reset signal to each cluster node;
after receiving the reset signal, the cluster node keeps a reset state;
the cluster master control determines a target cluster node to be released and reset currently according to a preset release sequence and sends a release reset signal to the target cluster node;
after receiving the reset releasing signal, the target cluster node releases reset to start running and updates the configuration of address information; after completing the configuration and updating of the address information, the target cluster node returns a configuration completion message to the cluster master control;
and after receiving the configuration completion message of the target cluster node, the cluster master determines the cluster node to be released and reset next according to the preset release sequence, and sends a release reset signal to the cluster node to be released and reset next.
2. The method of claim 1, wherein the server-side further comprises: and the internal Ethernet switching module is used for communicating between the cluster master controller and each cluster node and between each cluster node.
3. The method of claim 1, wherein the target cluster node releases reset and starts operation after receiving the release reset signal, and performs configuration update of address information, and the method comprises:
after receiving the reset releasing signal, the target cluster node releases reset to start operation, performs network initialization, and sends a broadcast message by default address information;
after receiving a broadcast message sent by default address information, the cluster master configures new address information for the target cluster node and sends the new address information to the target cluster node;
and after receiving the new address information, the target cluster node configures the address of the target cluster node as the new address information to complete configuration and updating of the address information.
4. The method of claim 3, wherein the target cluster node releases a reset start operation after receiving the release reset signal, performs network initialization, and sends a broadcast message with default address information, and the method comprises:
and after receiving the reset release signal, the target cluster node releases the reset, loads a boot loader BootLoader, performs network initialization in the running process of the BootLoader, and sends an Address Resolution Protocol (ARP) broadcast message by default address information.
5. The method of claim 3, wherein after the target cluster node configures the address of the target cluster node as the new address information after receiving the new address information, so as to complete the configuration update of address information, the method further comprises:
after the target cluster node completes the configuration updating of the address information, the target cluster node sends a broadcast message by the updated address information;
after receiving the broadcast message sent by the updated address information, the cluster master sends a unicast reply message to the target cluster node;
and after the target cluster node receives the unicast reply message, finishing firmware loading through a simple file transfer protocol (TFTP).
6. A server, characterized in that the server comprises:
the system comprises a cluster master controller and a plurality of cluster nodes, wherein the cluster master controller is in communication connection with each cluster node;
the cluster master control: for sending a hold reset signal to each of said cluster nodes; determining a target cluster node to be released and reset currently according to a preset release sequence, and sending a release reset signal to the target cluster node; after configuration completion information sent by the target cluster node is received, determining a cluster node to be reset next to be released according to the preset release sequence, and sending a release reset signal to the cluster node to be reset next to be released;
the cluster node: the reset circuit is used for keeping a reset state after receiving the reset keeping signal; and after receiving the reset releasing signal, releasing reset starting operation, updating the configuration of the address information, and after finishing the updating of the configuration of the address information, returning configuration finishing information to the cluster master control.
7. The server of claim 6, further comprising: and the internal Ethernet exchange module is used for communicating between the cluster master control and each cluster node and between each cluster node through the internal Ethernet interaction module.
8. The server according to claim 6, wherein the target cluster node, after receiving the release reset signal, releases a reset start operation, and performs configuration update of address information, including:
after receiving the reset releasing signal, releasing reset to start operation, performing network initialization, and sending a broadcast message by default address information; after receiving new address information returned by the cluster master controller, configuring an address into the new address information to complete configuration and updating of the address information;
the cluster master control: and the system is further configured to determine new address information for the target cluster node after receiving a broadcast message sent by the target cluster node with default address information, and send the new address information to the target cluster node.
9. The server according to claim 8, wherein the target cluster node, after receiving the release reset signal, releases a reset start operation, performs network initialization, and sends a broadcast message with default address information, and the method includes:
and after receiving the reset releasing signal, the target cluster node releases reset and starts to operate, loads a boot loader BootLoader, performs network initialization in the boot loader operation process, and sends an Address Resolution Protocol (ARP) broadcast message by default address information.
10. The server of claim 8, wherein the cluster node: the system is also used for sending the broadcast message by the updated address information after the address information configuration is updated; after receiving the unicast reply message returned by the cluster master control, completing firmware loading through a simple file transfer protocol (TFTP);
the cluster master control: and the unicast reply message is also sent to the target cluster node after the broadcast message sent by the target cluster node with the updated address information is received.
11. A cluster server system, the system comprising:
the system comprises a cluster master controller and a plurality of cluster nodes, wherein the cluster master controller is in communication connection with each cluster node;
the cluster master control: for sending a hold reset signal to each of said cluster nodes; determining a target cluster node to be released and reset currently according to a preset release sequence, and sending a release reset signal to the target cluster node; after configuration completion information sent by the target cluster node is received, determining a cluster node to be reset next to be released according to the preset release sequence, and sending a release reset signal to the cluster node to be reset next to be released;
the cluster node: the reset circuit is used for keeping a reset state after receiving the reset keeping signal; and after receiving the reset releasing signal, releasing reset starting operation, updating the configuration of the address information, and after finishing the updating of the configuration of the address information, returning configuration finishing information to the cluster master control.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811384397.3A CN111200513B (en) | 2018-11-20 | 2018-11-20 | Server starting method, server and cluster server system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811384397.3A CN111200513B (en) | 2018-11-20 | 2018-11-20 | Server starting method, server and cluster server system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111200513A true CN111200513A (en) | 2020-05-26 |
| CN111200513B CN111200513B (en) | 2022-03-04 |
Family
ID=70746438
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811384397.3A Active CN111200513B (en) | 2018-11-20 | 2018-11-20 | Server starting method, server and cluster server system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111200513B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020186662A1 (en) * | 2001-05-04 | 2002-12-12 | Tomassetti Stephen Robert | Initialization method for an entertainment and communications network |
| CN1449164A (en) * | 2002-03-28 | 2003-10-15 | 东芝开利株式会社 | Network address determining method and node thereof |
| CN101447894A (en) * | 2008-12-22 | 2009-06-03 | 三一重工股份有限公司 | Identification of physical address of partial node, network configuration and maintaining method, network system |
| CN102231233A (en) * | 2011-06-29 | 2011-11-02 | 南京航空航天大学 | Automatic guiding vehicle distributed autonomous cooperation control system and control method |
| CN104219099A (en) * | 2014-09-25 | 2014-12-17 | 浪潮(北京)电子信息产业有限公司 | Automatic server cluster deploying method and device |
| CN104410547A (en) * | 2014-12-03 | 2015-03-11 | 合肥工大高科信息科技股份有限公司 | Address conflict detection method for distributed control system |
-
2018
- 2018-11-20 CN CN201811384397.3A patent/CN111200513B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020186662A1 (en) * | 2001-05-04 | 2002-12-12 | Tomassetti Stephen Robert | Initialization method for an entertainment and communications network |
| CN1449164A (en) * | 2002-03-28 | 2003-10-15 | 东芝开利株式会社 | Network address determining method and node thereof |
| CN101447894A (en) * | 2008-12-22 | 2009-06-03 | 三一重工股份有限公司 | Identification of physical address of partial node, network configuration and maintaining method, network system |
| CN102231233A (en) * | 2011-06-29 | 2011-11-02 | 南京航空航天大学 | Automatic guiding vehicle distributed autonomous cooperation control system and control method |
| CN104219099A (en) * | 2014-09-25 | 2014-12-17 | 浪潮(北京)电子信息产业有限公司 | Automatic server cluster deploying method and device |
| CN104410547A (en) * | 2014-12-03 | 2015-03-11 | 合肥工大高科信息科技股份有限公司 | Address conflict detection method for distributed control system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111200513B (en) | 2022-03-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11388228B2 (en) | Methods, systems and computer readable media for self-replicating cluster appliances | |
| EP3291499B1 (en) | Method and apparatus for network service capacity expansion | |
| AU2003257885B2 (en) | Method and System for Network Switch Configuration | |
| US10826768B2 (en) | Controlled node configuration | |
| CN111404753B (en) | Flat network configuration method, computer equipment and storage medium | |
| CN113746887A (en) | Cross-cluster data request processing method, device and storage medium | |
| JP2018156645A (en) | Storage system and operation method thereof | |
| US20050262218A1 (en) | System and method for DHCP-based assignment of IP addresses to servers based on geographic identifiers | |
| CN105993161B (en) | Element, method, system and computer readable storage device for resolving an address | |
| CN110213391B (en) | Method and device for configuring network protocol address | |
| CN107172214B (en) | Service node discovery method and device with load balancing function | |
| US20040143654A1 (en) | Node location management in a distributed computer system | |
| CN110391940A (en) | Response method, device, system, equipment and the storage medium of address of service | |
| WO2016115896A1 (en) | Method and device for management of virtual network function (vnf) | |
| US7039682B2 (en) | Extension of the BOOTP protocol towards automatic reconfiguration | |
| US9749288B2 (en) | Management of network address pools | |
| US20110238959A1 (en) | Distributed controller, distributed processing system, and distributed processing method | |
| WO2018072450A1 (en) | Method for elastic geographical database replication | |
| CN116233217B (en) | Page jump method and device based on route, electronic equipment and storage medium | |
| CN111200513B (en) | Server starting method, server and cluster server system | |
| CN112667293B (en) | Method, device and storage medium for deploying operating system | |
| CN110636149A (en) | Remote access method, device, router and storage medium | |
| CN112511440B (en) | Message forwarding method, system, storage medium and electronic equipment | |
| CN115268951A (en) | Installation method, device, equipment and storage medium of operating system | |
| US20190052347A1 (en) | Communication system, control device, relay device, control method, and storage medium in which program is stored |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |