Method and equipment for setting remote access address

Abstract

The invention provides a method and equipment for setting a remote access address. The method and the equipment are applied to a network cloud controller of a distributed decoupling machine frame cluster. The method comprises the following steps: setting a management IP address state flag bit array; wherein the first flag bit of the management IP address state flag bit array indicates the state of the cluster management IP address, and the second flag bit of the management IP address state flag bit array starts to be the state of the Ethernet management interface IP address of each network cloud controller arranged according to a set sequence of the distributed decoupling machine frame cluster; and all flag bits of the management IP address state flag bit array are set to be valid, so that the distributed decoupling frame cluster is remotely logged in through the cluster management IP address and the Ethernet management interface IP address of each network cloud controller.

Description

Technical Field

[0001] This application relates to the technology of distributed disaggregated chassis, and specifically to a method and device for setting a remote access address. Background Art

[0002] With the rapid development of big data, cloud computing, and artificial intelligence technologies, data centers have encountered challenges of surging traffic. How to quickly process this traffic has brought huge pressure to the core switches of data centers.

[0003] DDC (Distributed Disaggregated Chassis) technology is an innovative network architecture design that uses a distributed disaggregated method to improve the flexibility and scalability of data center networks and realizes the decentralized deployment of network functions.

[0004] In the DDC cloud cluster network, the management networks of NCC (Network Cloud Controller) and NCP (Network Cloud Packet) need to interact through NGT (Management Switch). The DDC cloud cluster network also sets up NCF (Network Cloud Fabric).

[0005] When an NCP in the cluster escapes, it triggers the overall escape of the NCPs in the cloud cluster. According to the algorithm, an NCP is elected as the new master device, and the whole breaks away from the NCC and runs independently. The management interface IP address of the newly elected master device in each group of devices after the escape time All remain is the management interface IP address of the old master device in the cluster before the split. In this way, when a user remotely accesses the management interface IP address of the old master device, this IP address is preempted by the new master devices of each group of devices after the split, resulting in the drift of the user's remote access among multiple groups of devices after the split and unable to stably remotely access the cluster. After a cluster split event occurs in DCC, there will also be the same problem of being unable to stably remotely access the cluster. Summary of the Invention

[0006] The purpose of this application is to provide a method and device for setting a remote access address to stably remotely access a distributed disaggregated chassis cluster.

[0007] To achieve the above object, the present application provides a method for setting a remote access address, which is applied to a network cloud controller of a distributed decoupled chassis cluster. The method includes: setting an array of management IP address status flag bits; wherein, the first flag bit of the array of management IP address status flag bits indicates the status of the cluster management IP address, and starting from the second bit of the array of management IP address status flag bits are the statuses of the Ethernet management interface IP addresses of each network cloud controller in the distributed decoupled chassis cluster arranged in a set order; setting all the flag bits of the array of management IP address status flag bits to valid, so as to remotely log in to the distributed decoupled chassis cluster through the cluster management IP address and the Ethernet management interface IP addresses of each network cloud controller.

[0008] To achieve the above object, the present application provides a device for setting a remote access address. The device, as a network cloud controller of a distributed decoupled chassis cluster, includes a processor and a memory; the memory is used to store machine-executable instructions; wherein, the processor executes the method for setting a remote access address implemented by the network cloud controller of the distributed decoupled chassis cluster by running the machine-executable instructions in the memory.

[0009] To achieve the above object, the present application provides a method for setting a remote access address, which is applied to a network cloud service board of a distributed decoupled chassis cluster. The method includes: setting an array of management IP address status flag bits; wherein, the first bit of the array of management IP address status flag bits indicates the status of the cluster management IP address, and the flag bits starting from the second bit of the array of management IP address status flag bits indicate the statuses of the Ethernet management interface IP addresses of each network cloud controller in the distributed decoupled chassis cluster arranged in a set order; setting all the flag bits of the array of management IP address status flag bits to valid, so as to remotely log in through the cluster management IP address and the Ethernet management interface IP addresses of each network cloud controller.

[0010] To achieve the above object, the present application provides a device for setting a remote access address. The device, which is applied to a network cloud service board of a distributed decoupled chassis cluster, includes a processor and a memory; wherein, the processor executes the method for setting a remote access address implemented by the network cloud service board of the distributed decoupled chassis cluster by running the machine-executable instructions in the memory.

[0011] The beneficial effect of the present application is that in the stable state of the distributed decoupled chassis cluster, the entire cluster can be remotely accessed through the IP address of any Ethernet management interface of the management IP address status flag bit array without address conflict. Description of the Drawings

[0012] Figure 1Flowchart of an embodiment of a method for setting a remote access address provided by this application;

[0013] Figure 2 Schematic diagram of a distributed decoupled chassis cluster;

[0014] Figures 3A - 3B Flowchart of an embodiment of setting a remote access address after collective escape of a container cluster provided by this application;

[0015] Figure 4 Flowchart of an embodiment of setting a remote access address after cluster splitting provided by this application;

[0016] Figure 5 Schematic diagram of an embodiment of a device for setting a remote access address provided by this application;

[0017] Figure 6 Schematic diagram of an embodiment of a device for setting a remote access address provided by this application. Detailed implementation manners

[0018] Multiple examples shown in multiple drawings will be described in detail. In the following detailed description, multiple specific details are used to provide a comprehensive understanding of this application. Known methods, steps, components, and circuits are not described in detail in the examples to avoid making these examples difficult to understand.

[0019] Among the terms used, the term "including" means including but not limited to; the term "containing" means including but not limited to; the terms "above", "within", and "below" include the number itself; the terms "greater than" and "less than" do not include the number itself. The term "based on" means at least based on a part of it.

[0020] Figure 1 Flowchart of an embodiment of a method for setting a remote access address provided by this application; this embodiment can be applied to a network cloud controller of a distributed decoupled chassis cluster, and the method of this embodiment includes the following steps:

[0021] Step 101, set an array of management IP address status flag bits; among them, the first flag bit of the array of management IP address status flag bits indicates the status of the cluster management IP address, and starting from the second bit of the array of management IP address status flag bits are the statuses of the Ethernet management interface IP addresses of each network cloud controller in the distributed decoupled chassis cluster arranged in a set order;

[0022] Step 102, set the first flag bit used to indicate the status of the cluster management IP address and other flag bits starting from the second bit of the Ethernet management interface IP addresses of each network cloud controller in the distributed decoupled chassis cluster arranged in a set order in the array of management IP address status flag bits to valid;

[0023] In this way, all flag bits of the array for managing the IP address status flags are set to valid, so as to remotely log in to the distributed decoupled chassis cluster through the cluster management IP address and the Ethernet management interface IP addresses of each network cloud controller.

[0024] Figure 1 Beneficial effects of the embodiment: In the stable state of the distributed decoupled chassis cluster, the entire cluster can be remotely accessed through any one of the IP addresses in the array for managing the IP address status flags.

[0025] Figure 2 It is a schematic diagram of a distributed decoupled chassis cluster. In the networking of this DDC cluster, it includes two NCC11 and NCC12, two NCF31 and 32, and four NCP41 - 44.

[0026] NCC11 and 12, NCF31 and 32, and NCP41 - 44 are all connected to MGT21 and 22. The DDC cluster is virtualized into a virtual chassis network device, and the resources on all physical devices are owned and uniformly managed by this virtual network device.

[0027] When NCC11 and NCC12, two NCF31 and 32, and four NCP41 - 44 are started, a Route container will be run to build the management network of the DDC cloud cluster.

[0028] In the networking of a DDC cluster, only the NCC participates in the physical cluster management, that is, as the Manager of the DDC cluster. Figure 2 In the DDC cluster networking, there are two NCCs, and the NCC with the smaller MemberID is selected as the Leader.

[0029] In a DDC cluster (physical cluster), a cluster member number mapping file containing the serial number, MAC address, and Slot ID member number is stored on NCC21 and 22. When NCF31 and 32, and NCP41 - 44 register with NCC21 and 22 carrying their own serial numbers and MAC addresses, the primary NCC11 selected as the leader will search for the physical cluster member number mapping file according to the respective serial numbers and MAC addresses of NCF31 and 32, and NCP41 - 44, and assign the found SlotID to NCF31 and 32, and NCP41 - 44. NCF31 and 32, and NCP41 - 44 each generate a MemberID based on this SlotID. Relying on the management network of the DDC cluster, the physical cluster completes the protocol message interaction, and the physical cluster will be automatically established.

[0030] When the connection between NCC11 and NCC12 is disconnected and NCC12 does not receive the protocol message from the primary NCC11, the standby NCC12 will consider the primary NCC11 to be faulty. When NCC11 and NCC12 detect a physical cluster split, NCC12 and the NCF31, 32, and NCP41 - 44 attached to it will form a new DDC physical cluster, and the original physical DDC cluster will split into two physical clusters. The newly split physical cluster will elect NCC12 as the new leader according to the leader election rule (the one with a smaller member number has priority).

[0031] When the faulty links between NCC11 and NCC12 are repaired, the two split physical clusters will automatically merge into one NCC physical cluster. Only one NCC will be retained as the leader of the re - merged physical cluster, and the physical cluster that fails in the election will re - join the merged physical cluster as a follower. The election rules for the two physical clusters are as follows: the physical cluster with a larger total number of NCPs wins; if the total number of NCPs is the same, the physical cluster with a smaller member number wins. NCC12 is elected as the leader of the merged NCC physical cluster, and the failed NCC11 re - joins the merged physical cluster.

[0032] In addition, when NCC11 and 12, NCF31 and 32, and NCP41 - 44 in the physical cluster start up, they will automatically run operating system containers to provide message forwarding services. To improve the reliability and maintainability of the services, these operating system containers will form a container cluster. When the physical cluster is working properly, the roles of the primary container and the secondary container in the container cluster are determined by the leader in the physical cluster; when the physical cluster is not working properly, both the primary device and the secondary device are elected. There can only be one primary container in a container cluster at the same time. To prevent all the operating system containers of NCC from failing and the DDC cloud cluster from being unable to provide data forwarding services, the DDC cloud cluster provides a container cluster escape function. An additional operating system container is automatically run on each of the two NCPs that are first registered to the NCC for container cluster management. This additional operating system container is called an escape container. When there is an NCC in the network, the operating system containers on the NCPs all run as secondary containers in the standby role; when all the NCCs in the network fail, the DDC cloud cluster will elect one from the escape containers as the Master container. The rule for electing the Master from the escape containers in the secondary container cluster is: the escape container running on the NCP with the smallest member number is elected as the Master container. For example: Figure 2In the embodiment, NCP41 is registered to NCC11 and 12 first. When both NCC11 and 12 in the network are faulty, NCC11, 12, NCP41 - 44 detect the container cluster escape event. The escape containers running on NCP41 and 42 elect the escape container of NCP41 as the Master container, and NCP41 is the master device of the escape container cluster.

[0033] The above are respectively the DDC physical cluster split and the collective escape mechanism of the container cluster.

[0034] In the embodiment of the present application, both the NCC device and the NCP device of the DDC cluster (physical cluster) are set with a management IP address status flag bit array, as follows:

[0035] [Flag bit 0, Flag bit 1, Flag bit 2,... Flag bit N];

[0036] This array has N + 1 management IP addresses. Among them, the first flag bit 0 is used to indicate the status of the cluster management IP address; flag bits 1 - N are used to indicate the management IP addresses of the Ethernet management interfaces of N NCCs in the DDC cluster arranged in the specified order.

[0037] In the embodiment of the present application, the management IP addresses of the Ethernet management interfaces of N NCCs in this array are arranged in the order of SlotID.

[0038] Figure 2 In the shown DDC cluster, the management IP address status flag bit arrays set for NCC11, NCC12, NCP41 - 44 are all: [Flag bit 0, Flag bit 1, Flag bit 2]; among them, flag bit 0 is the cluster management IP address, and flag bits 1 and 2 are the management IP addresses of the Ethernet management interfaces of NCC11 and NCC12 arranged in ascending order according to the slot ID.

[0039] In Figure 2 In the stable state of the shown DCC cluster, this array is set to [1, 1, 1] so that users can remotely log in to the DDC cloud cluster through the cluster management IP address, the Ethernet management interface IP addresses of NCC11 and NCC12, in ways such as Telnet, SSH, SNMP, etc. through the login terminal.

[0040] Figures 3A - 3B This is the flowchart of the embodiment for setting the remote access address after the collective escape of the container cluster of the distributed decoupled chassis cluster provided by the present application; Figure 3A The shown example is applied to the NCC of the DCC cluster, Figure 3B Applied to the NCP of the DDC cluster.

[0041] Figure 3AThe flowchart shown includes the following steps:

[0042] Step 301, detecting a container cluster collective escape event;

[0043] Based on the container cluster collective escape detection mechanism of the DDC cluster, NCC11 and NCC12 detect the collective escape of the container clusters of NCP41 - 44.

[0044] Step 302, detecting other network cloud controllers of the same container cluster;

[0045] NCC11 and NCC12 detect the existence of other NCC devices in the container cluster with each other.

[0046] Step 303, in the management IP address status flag bit array, set the flag bits of the Ethernet management interface IP address of this device and the Ethernet management interface IP addresses of the other detected network cloud controllers to valid, and set the remaining flag bits of the management IP address status flag bit array to invalid.

[0047] Based on the slotID sorting respectively, NCC11 and NCC12 set the flag bits of the Ethernet management interface IP addresses of NCC11 and NCC12 in the local array to valid, and set other flag bits to invalid, for example: [0, 1, 1]; in this way, through the IP addresses of the Ethernet management interfaces of NCC11 and NCC12, remotely log in to the container cluster composed of NCC11 and 12.

[0048] Step 304, detecting a container cluster merge and recovery event;

[0049] Based on the container cluster merge and recovery mechanism of the DDC cluster, NCC11 and NCC12 detect the collective merge of the container clusters.

[0050] Step 305, set all flag bits of the management IP address status flag bit array to valid.

[0051] NCC11 and NCC12 respectively set all flag bits of the local setting array to valid, for example: [1, 1, 1]; in this way, through the cluster management IP address and the IP addresses of the Ethernet management interfaces of NCC11 and NCC12, all can remotely log in.

[0052] Figure 3B The flowchart shown includes the following steps:

[0053] Step 311, detecting a container cluster collective escape event;

[0054] NCP41, 42, 43, and 44 respectively, according to the container cluster collective escape detection mechanism of the DDC cluster, detect that the container clusters of NCP41, 42, 43, and 44Figure 2 Escape from the container cluster with the shown architecture.

[0055] Step 312: Elect the master device of the current container cluster.

[0056] Based on the master device election mechanism of the escape container cluster, NCP41 and 42 select NCP41 as the master device of the escape container cluster.

[0057] Step 313: Determine that this device is elected as the master device, set the first flag bit of the management IP address status flag bit array to valid, and set other flag bits to invalid.

[0058] The NCP sets the local management IP address status flag bit array to [1, 0, 0] to enable remote login to the current container cluster through the cluster management IP address.

[0059] Step 314: Detect a container cluster merge and recovery event.

[0060] Based on the container cluster merge and recovery mechanism of the DDC cluster, NCP41 detects the collective merge of the container clusters.

[0061] Step 315: Set all flag bits of the management IP address status flag bit array to valid.

[0062] The NCP sets the local management IP address status flag bit array to [1, 1, 1], so that remote login can be performed through the cluster management IP address, the IP addresses of the Ethernet management interfaces of NCC11 and 12.

[0063] NCP42 is not elected as the master device, but can perform the same configuration as NCP41. In the example shown in Figure 3, after container cluster escape occurs in the DDC cluster, the two container clusters can be remotely logged in through different management IP addresses, and there will be no problem of remote access drifting between multiple groups of devices after splitting.

[0064] Figure 4 This is the flowchart of the embodiment for setting the remote access address after the distributed decoupled chassis cluster provided by this application splits; this embodiment includes the following steps:

[0065] Step 401: Detect a cluster split event.

[0066] According to the cluster split detection mechanism of the DDC cluster, NCC11 and NCC12 detect that the DDC cluster splits into two new physical clusters. Among them, NCC12 and NCP41 - 42 split into one new physical cluster, and NCC11 splits into one new physical cluster.

[0067] Step 402: Detect other network cloud controllers of the new physical cluster.

[0068] NCC11 and NCC12 detect other NCCs in the new physical cluster after splitting according to the NCC detection mechanism of the DDC cluster.

[0069] Although, Figure 2 The DDC cluster shown contains two NCCs; those skilled in the art can apply this embodiment to a DDC cluster containing multiple NCCs, and one or both of the physical clusters after splitting have more than two NCCs.

[0070] Step 403: Determine whether more than one network cloud service board of the new physical cluster is detected. If so, execute Step 404; if not, execute Step 405.

[0071] NCC11 and NCC12 detect other NCPs in the new physical cluster after splitting according to the NCP detection mechanism of the DDC cluster.

[0072] Step 404: Set the first flag bit, the IP address of the Ethernet management interface of this device, and the flag bits corresponding to the IP addresses of the Ethernet management interfaces of other detected network cloud controllers in the management IP address status flag bit array to valid, and set the remaining flag bits of the management IP address status flag bit array to invalid.

[0073] NCC12 detects that the new physical cluster after splitting includes NCP41 - 44, and the local management IP address status flag bit array is set to [1, 0, 1]. In this way, both the cluster management IP address and the IP address of the Ethernet management interface of NCC12 can be used for remote login.

[0074] Although there is only one NCC in the new physical cluster where NCC12 is located after splitting; those skilled in the art can apply this embodiment to the new physical cluster containing multiple NCCs after splitting. Except that the first identification bit is set to 1 to indicate valid, the flag bits corresponding to the NCCs in the same new physical cluster are set to 1 according to the slot ID.

[0075] Step 405: Set the IP address of the Ethernet management interface of this device and the IP addresses of the Ethernet management interfaces of other detected network cloud controllers in the management IP address status flag bit array to valid, and set the remaining flag bits of the management IP address status flag bit array to invalid, so as to enable remote login through the IP addresses of the Ethernet ports of each network cloud controller in the new physical cluster.

[0076] NCC11 detects that the new physical cluster after splitting does not include NCP, and the local management IP address status flag bit array is set to [0, 1, 0]. In this way, the IP address of the Ethernet management interface of NCC11 can be used for remote login.

[0077] Those skilled in the art can apply this embodiment to the newly split physical cluster that contains multiple NCCs and does not contain NCP, and set the flag bits corresponding to the NCCs within the same newly split physical cluster to the valid value 1 according to the slot ID.

[0078] Figure 4 As shown in the illustrated example, after the DDC cluster undergoes cluster splitting, the two newly split physical clusters can be remotely logged in through different management IP addresses, and there will be no problem of remote access drifting among the multiple groups of devices after splitting.

[0079] Step 406: Determine whether a physical cluster merge and recovery event is detected; if so, execute step 407; if not, return to step 406.

[0080] Step 407: Set all the flag bits of the management IP address status flag bit array to valid.

[0081] NCC11 and 12 detect the merge and recovery of the newly split physical cluster, and their respective local management IP address status flag bit arrays are set to [1, 1, 1]. In this way, the DDC cluster after merge and recovery can be remotely logged in through the cluster management IP address and the IP addresses of the Ethernet management interfaces of NCC11 and NCC12.

[0082] Figure 5 It is a schematic diagram of an embodiment of a device for setting a remote access address provided by this application; this device 50 is applied to the network cloud controller of a distributed decoupled chassis cluster, and this device 50 includes a processor 51, a machine-readable storage medium 52, a switching chip 53, and a network interface 54.

[0083] The processor 51 executes the following processing by executing the machine-executable instructions recorded in the machine-readable storage medium 52: setting the management IP address status flag bit array; where the first flag bit of the management IP address status flag bit array indicates the status of the cluster management IP address, and starting from the second bit of the management IP address status flag bit array are the statuses of the Ethernet management interface IP addresses of each network cloud controller of the distributed decoupled chassis cluster arranged in a set order; setting all the flag bits of the management IP address status flag bit array to valid so as to remotely log in to the distributed decoupled chassis cluster through the cluster management IP address and the Ethernet management interface IP addresses of each network cloud controller.

[0084] The processor 51 also performs the following processes by executing machine-executable instructions recorded in the machine-readable storage medium 52: detecting a container cluster collective escape event; detecting other network cloud controllers of the same container cluster; setting the flag bits of the Ethernet management interface IP address of this device and the Ethernet management interface IP addresses of the detected other network cloud controllers in the management IP address status flag bit array to valid, and setting the remaining flag bits of the management IP address status flag bit array to invalid.

[0085] The processor 51 also performs the following processes by executing machine-executable instructions recorded in the machine-readable storage medium 52: detecting a cluster split event; detecting other network cloud controllers and one or more network cloud service boards of a new physical cluster; setting the first flag bit, the Ethernet management interface IP address of this device, and the flag bits corresponding to the Ethernet management interface IP addresses of the detected other network cloud controllers in the management IP address status flag bit array to valid, and setting the remaining flag bits of the management IP address status flag bit array to invalid, so as to remotely log in to the Ethernet management interface through the cluster management IP address and the Ethernet management interface IP addresses of the network cloud controllers of the new cluster.

[0086] The processor 51 also performs the following processes by executing machine-executable instructions recorded in the machine-readable storage medium 52: detecting a cluster split event; detecting other network cloud controllers of a new physical cluster; setting the Ethernet management interface IP address of this device and the Ethernet management interface IP addresses of the detected other network cloud controllers in the management IP address status flag bit array to valid, and setting the remaining flag bits of the management IP address status flag bit array to invalid, so as to remotely log in through the Ethernet port IP addresses of the network cloud controllers of the new physical cluster.

[0087] The processor 51 also performs the following processes by executing machine-executable instructions recorded in the machine-readable storage medium 52: detecting a physical cluster merge recovery event; setting all the flag bits of the management IP address status flag bit array to valid; or, detecting a container cluster merge recovery event; setting all the flag bits of the management IP address status flag bit array to valid.

[0088] Figure 6 It is a schematic diagram of an embodiment of a device for setting a remote access address provided for this application; this device 60 is applied to a network cloud service board of a distributed decoupled chassis cluster; this device 60 includes a processor 61, a machine-readable storage medium 62, a switching chip 63, and a network interface 64.

[0089] The processor 61 executes the following processes by executing machine-executable instructions recorded in the machine-readable storage medium 62: setting an array of management IP address status flag bits; wherein, the first bit of the array of management IP address status flag bits indicates the status of the cluster management IP address, and the flag bits starting from the second bit of the array of management IP address status flag bits indicate the status of the Ethernet management interface IP addresses of the respective network cloud controllers of the distributed decoupled chassis cluster arranged in a set order; setting all the flag bits of the array of management IP address status flag bits to valid so as to enable remote login via the cluster management IP address and the Ethernet management interface IP addresses of the respective network cloud controllers.

[0090] The processor 61 also executes the following processes by executing machine-executable instructions recorded in the machine-readable storage medium 62: detecting a container cluster collective escape event; electing a master device of the current container cluster; determining that the device itself is elected as the master device, setting the first flag bit of the array of management IP address status flag bits to valid, and setting the other flag bits to invalid so as to enable remote login to the current container cluster via the cluster management IP address.

[0091] The processor 61 also executes the following processes by executing machine-executable instructions recorded in the machine-readable storage medium 62: detecting a container cluster merge and recovery event; setting all the flag bits of the array of management IP address status flag bits to valid so as to enable remote login to the distributed decoupled chassis cluster via the cluster management IP address and the Ethernet management interface IP addresses of the respective network cloud controllers.

[0092] The foregoing are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the scope of protection of the present application.

Claims

1. A method for setting a remote access address, which is applied to a network cloud controller of a distributed decoupled chassis cluster, and is characterized in that, The method includes: Setting an array of management IP address status flags; wherein, the first flag bit of the array of management IP address status flags indicates the status of the cluster management IP address, and starting from the second bit of the array of management IP address status flags are the statuses of the Ethernet management interface IP addresses of each network cloud controller in the distributed decoupled chassis cluster arranged in a set order; Setting all the flag bits of the array of management IP address status flags to valid, so as to remotely log in to the distributed decoupled chassis cluster through the cluster management IP address and the Ethernet management interface IP addresses of each network cloud controller.

2. The method according to claim 1, wherein The method includes, Detecting a container cluster collective escape event; Probing other network cloud controllers of the same container cluster; Setting the flag bits of the Ethernet management interface IP address of this device and the Ethernet management interface IP addresses of the other network cloud controllers detected in the array of management IP address status flags to valid, and setting the remaining flag bits of the array of management IP address status flags to invalid.

3. The method according to claim 1, wherein The method includes, Detecting a cluster split event; Probing other network cloud controllers and more than one network cloud service board of the new physical cluster; Setting the first flag bit, the flag bit corresponding to the Ethernet management interface IP address of this device, and the flag bits corresponding to the Ethernet management interface IP addresses of the other network cloud controllers detected in the array of management IP address status flags to valid, and setting the remaining flag bits of the array of management IP address status flags to invalid, so as to remotely log in through the cluster management IP address and the Ethernet management interface IP addresses of each network cloud controller of the new cluster.

4. The method according to claim 1, wherein The method includes, Detecting a cluster split event; Probing other network cloud controllers of the new physical cluster; Setting the Ethernet management interface IP address of this device and the Ethernet management interface IP addresses of the other network cloud controllers detected in the array of management IP address status flags to valid, and setting the remaining flag bits of the array of management IP address status flags to invalid, so as to remotely log in through the Ethernet interface IP addresses of each network cloud controller of the new physical cluster.

5. The method according to claim 2 or 3 or 4, characterized in that, The method further includes, Detecting a physical cluster merge recovery event; Setting all the flag bits of the array of management IP address status flags to valid; Or, Detecting a container cluster merge recovery event; Setting all the flag bits of the array of management IP address status flags to valid.

6. A device for setting a remote access address, characterized in that, The device, as a network cloud controller of a distributed decoupled chassis cluster, includes a processor and a memory; the memory is used for storing machine-executable instructions; wherein, the processor executes any one of the methods in claims 1-6 by running the machine-executable instructions in the memory.

7. A method for setting a remote access address, which is applied to a network cloud service board of a distributed decoupled chassis cluster, is characterized in that The method includes: Set up an array of management IP address status flags; wherein, the first bit of the management IP address status flag array indicates the status of the cluster management IP address, and the flag bits starting from the second bit of the management IP address status flag array indicate the status of the Ethernet management interface IP addresses of each network cloud controller in the distributed decoupled chassis cluster arranged in a set order; Set all the flag bits of the management IP address status flag array to valid so as to enable remote login through the cluster management IP address and the Ethernet management interface IP addresses of each network cloud controller.

8. The method according to claim 7, characterized in that The method further includes, Detect a container cluster collective escape event; Elect a master device of the current container cluster; Determine that the device is elected as the master device, set the first flag bit of the management IP address status flag array to valid, and set the other flag bits to invalid so as to enable remote login to the current container cluster through the cluster management IP address.

9. The method according to claim 8, characterized in that, The method further includes, Detect a container cluster merge and recovery event; Set all the flag bits of the management IP address status flag array to valid so as to enable remote login to the distributed decoupled chassis cluster through the cluster management IP address and the Ethernet management interface IP addresses of each network cloud controller.

10. A device for setting a remote access address, characterized in that, The device is applied to a network cloud service board of a distributed decoupled chassis cluster, which includes a processor and a memory; wherein, the processor executes the method of any one of claims 7-9 by running machine-executable instructions in the memory.

Images