WO2016062037A1 - Method, apparatus and system for information transmission and controller fault handling through interface cards - Google Patents

Abstract

Disclosed is a storage system for solving the technical problems of service interruption occurs in a faulty interface card of a controller and the use of the apparatus connected with the interface card cannot be continued. The system comprises: a number M of controllers (201) for controlling the system, wherein the number M of controllers (201) comprise a master controller (201), and a number M-1 of slave controllers (201) as redundancy, and M is a positive integer; a number N of interface cards (202), wherein each interface card (202) is connected to at least two controllers (201), so as to relay signals transmitted by the controllers (201) or signals transmitting to the controllers (201), or to process signals from the controllers (201); and N is an integer less than or equal to M. Also disclosed are a method and corresponding apparatus for information transmission and fault handling for the controllers (201) through the interface cards (202).

Description

Information transmission through interface card, controller failure processing method, device and system Technical field

The present invention relates to the field of communications, and in particular, to a method for transmitting information through an interface card, a controller fault handling method, an apparatus, and a system.

Background technique

At present, most of the storage systems in the industry use more than two controllers for redundant design, which is designed for high reliability requirements of storage. In this context, the controllers are designed with redundancy. If one controller fails, it will not affect the system business. All services are taken over by redundant controllers. This design is very important for storage reliability. of.

The interface card for controller control and management belongs to each controller. As shown in Figure 1, interface card 1 is connected to controller A, and interface card 2 is connected to controller B. Each interface card only serves one control. Device. For example, if controller A fails, the service on the link is stopped and the interface card 1 cannot be used any more. Among them, one controller can also be connected with multiple interface cards. In Figure 1, an interface card is taken as an example, but each interface card can only serve one controller.

It can be seen that once a controller fails, all interface cards belonging to the controller are also unavailable. Generally, the interface card is connected to other devices such as disks. According to the working mode in the prior art, after the interface card is stopped, the service in the interface card is interrupted, and the disk connected to the interface card may also be caused. The device can no longer receive or send information, which is equivalent to the device connected to the interface card can no longer be used.

Summary of the invention

Embodiments of the present invention provide a method, a device, and a system for processing information through a interface card, which are used to solve a service interruption in a controller faulty interface card and a device that cannot be used continuously with the interface card. problem.

In a first aspect of the invention, a storage system is provided, comprising:

M controllers for controlling the system; the M controllers include a main controller and M-1 slave controllers as redundant, and M is a positive integer;

N interface cards, wherein each interface card is coupled to at least two controllers for relaying signals transmitted by the controller, or signals transmitted to the controller, or for processing from the controller Signal; N is an integer less than or equal to M.

In conjunction with the first aspect, in a first possible implementation of the first aspect, the interface card is connected to the controller through a PCIE bus.

In conjunction with the first possible implementation of the first aspect, in a second possible implementation of the first aspect, a serial control bus and/or parallel control is further connected between the interface card and the controller Bus for transmitting control signals.

In conjunction with the first aspect or the first possible implementation or the second possible implementation of the first aspect, in a third possible implementation of the first aspect, the system further includes at least one storage device, wherein Each storage device is coupled to the at least one interface card to enable the controller to interact with the storage device via a corresponding interface card.

A second aspect of the present invention provides a method for transmitting information through an interface card, including:

When the first controller as the primary controller of the M controllers fails, the second controller of the M controllers competes for the new primary controller;

The second controller performs information relaying by using at least one of the N interface cards; wherein the first interface card is respectively connected to the first controller and the second controller.

With reference to the second aspect, in a first possible implementation manner of the second aspect, before the second controller of the M controllers competes for the new primary controller, the method further includes:

The second controller receives a first fault notification message sent by the first controller, where the first fault notification message is used to notify the second controller that the first controller has a fault.

With reference to the second aspect or the first possible implementation manner of the second aspect, in a second possible implementation manner of the second aspect, the second controller in the M controllers competes for a new master control After the device, it also includes:

The second controller receives a second fault notification message sent by a third controller of the M controllers, where the second fault notification message is used to notify the second controller, the third controller A failure has occurred;

The second controller removes information of the third controller from a list of redundant controllers; wherein the redundant controller list is used to record information that can be used as redundant controllers.

A third aspect of the present invention provides a controller fault processing method, including:

When the first controller as the primary controller of the M controllers fails, the first interface card connected to the first controller and serving the first controller of the N interface cards receives the a second fault notification message sent by the first controller, where the second fault notification message is used to notify the first interface card, where the first controller has a fault;

The first interface card controls the port connected to the first controller to enter an inactive state according to the second failure notification message to stop communication with the first controller.

With reference to the third aspect, in a first possible implementation manner of the third aspect, after the first interface card receives the second fault notification message that is sent by the first controller, the method further includes:

The first interface card receives a master control notification message sent by a second controller of the M controllers, where the master control notification message is used to notify the first interface card, and the second controller has been contending For the new primary controller;

The first interface card controls, according to the master control notification message, that a port connected to the second controller enters an active state to communicate with the second controller through a port connected to the second controller. .

A fourth aspect of the present invention provides a method for transmitting information through an interface card, including:

When the first controller that is the primary controller of the M controllers included in the storage system fails, the first controller sends a second failure notification message to the first interface card, where the second failure notification message is used. The first interface card is configured to notify the first interface card that the first controller is faulty; wherein the first interface card is connected to the first controller in the N interface cards included in the storage system, and is An interface card served by the first controller;

The first interface card controls, according to the second failure notification message, that a port connected to the first controller enters an inactive state to stop communication with the first controller;

When the second controller of the M controllers contends to be the new master controller, the second controller sends a master control notification message to the first interface card, where the master control notification message is used to notify The first interface card, the second controller has competed for a new primary controller; wherein the first interface card is connected to the second controller;

The first interface card controls, according to the master control notification message, that a port connected to the second controller enters an active state to communicate with the second controller through a port connected to the second controller. .

According to a fifth aspect of the invention, a controller is provided, comprising:

An operation module, when the first controller as the main controller of the M controllers fails, causing the controller to compete for a new main controller;

And a communication module, configured to perform information relaying by using at least one of the N interface cards; wherein the first interface card is respectively connected to the first controller and the controller.

In conjunction with the fifth aspect, in a first possible implementation manner of the fifth aspect, the controller further includes: a receiving module, configured to: before the operating module causes the controller to compete for a new primary controller, Receiving a first fault notification message sent by the first controller, where the first fault notification message is used to notify the controller that the first controller has a fault.

With reference to the fifth aspect, or the first possible implementation manner of the fifth aspect, in a second possible implementation manner of the fifth aspect, the controller further includes a receiving module and a removing module;

The receiving module is configured to: after the operating module causes the controller to compete for a new primary controller, receive a second fault notification message sent by a third controller of the M controllers, where the second fault notification is sent a message is used to notify the controller that the third controller has failed;

The removal module is configured to remove information of the third controller from a list of redundant controllers; wherein the redundant controller list is used to record information that can be used as redundant controllers.

According to a sixth aspect of the invention, an interface card is provided, including:

a receiving module, configured to receive a second fault notification message sent by the first controller when a first controller that is the master controller of the M controllers fails, where the second fault notification message is used to notify the In the interface card, the first controller is faulty; wherein the interface card is an interface card of the N interface cards that is connected to the first controller and serves the first controller;

And a control module, configured to control, according to the second fault notification message, a port connected to the first controller to enter an inactive state to stop communication with the first controller.

With reference to the sixth aspect, in a first possible implementation manner of the sixth aspect, the receiving module is further configured to: after receiving the second fault notification message sent by the first controller, receive the M controls a master control notification message sent by the second controller in the device, the master control notification message is used to notify the interface card, and the second controller has competed for a new master controller;

The control module is further configured to: according to the master control notification message, control a port connected to the second controller to enter an activation state, to connect a port connected to the second controller with the second controller Communicate.

According to a seventh aspect of the present invention, a storage system is provided, including:

a first controller, configured to send a second fault message to the first interface card when the first controller that is the master controller in the M controllers included in the storage system fails, the second The fault notification message is used to notify the first interface card that the first controller has a fault; wherein the first interface card is among the N interface cards included in the storage system and the first control An interface card that is connected to and serves the first controller;

The first interface card is configured to control, according to the second fault notification message, a port connected to the first controller to enter an inactive state to stop communication with the first controller;

a second controller, configured to send a master control notification message to the first interface card when the second controller of the M controllers contends to be a new master controller, the master control notification message For notifying the first interface card, the second controller has competed for a new primary controller; wherein the first interface card is connected to the second controller;

The first interface card is further configured to control, according to the master control notification message, with the second controller The connected port enters an active state to communicate with the second controller through a port connected to the second controller.

In the embodiment of the present invention, an interface card is connected to at least two controllers. If one of the controllers connected to an interface card fails, the interface card can stop serving the controller, and the interface card is also Connected to other controllers, it can continue to serve other controllers. In this way, even if the controller fails, the interface card can continue to serve other controllers as long as there is no fault, and can continue to be used. Compared with the prior art, the services in the interface card are not interrupted, and the interface card is connected. Other hardware devices can continue to transmit information through the interface card, which ensures the reliability of the system.

Moreover, the interface card and the device connected to the interface card can continue to be used, which also saves hardware resources to a certain extent and improves the utilization of the interface card. Moreover, with the technical solution in the embodiment of the present invention, the number of interface cards can be reduced to a certain extent, the system structure tends to be simple, and the volume of the system is reduced.

DRAWINGS

1 is a structural diagram of a storage system in the prior art;

2 is a schematic structural diagram of a storage system according to an embodiment of the present invention;

3 is a detailed structural diagram of an implementation manner of a storage system according to an embodiment of the present invention;

4 is a schematic structural diagram of another implementation manner of a storage system according to an embodiment of the present invention;

FIG. 5 is a main flowchart of a method for transmitting information by using an interface card according to an embodiment of the present invention;

6 is a main flowchart of a controller fault processing method according to an embodiment of the present invention;

FIG. 7 is a main flowchart of another method for transmitting information by using an interface card according to an embodiment of the present invention;

8 is a main structural block diagram of a controller in an embodiment of the present invention;

FIG. 9 is a main structural block diagram of an interface card according to an embodiment of the present invention.

detailed description

An embodiment of the present invention provides a storage system, including: M controllers for controlling the system; the M controllers include one main controller and M-1 as redundant slave controllers, M Positive An integer number; N interface cards, wherein each interface card is coupled to at least two controllers for relaying signals transmitted by the controller or transmitted to the controller, or for processing from the control The signal of the device; N is an integer less than or equal to M.

In the embodiment of the present invention, an interface card is connected to at least two controllers. If one of the controllers connected to an interface card fails, the interface card can stop serving the controller, and the interface card is also Connected to other controllers, it can continue to serve other controllers. In this way, even if the controller fails, the interface card can continue to serve other controllers as long as there is no fault, and can continue to be used. Compared with the prior art, the services in the interface card are not interrupted, and the interface card is connected. Other hardware devices can continue to transmit information through the interface card, which ensures the reliability of the system.

Moreover, the interface card and the device connected to the interface card can continue to be used, which also saves hardware resources to a certain extent and improves the utilization of the interface card. Moreover, with the technical solution in the embodiment of the present invention, the number of interface cards can be reduced to a certain extent, the system structure tends to be simple, and the volume of the system is reduced.

The technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. It is a partial embodiment of the invention, and not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

Additionally, the terms "system" and "network" are used interchangeably herein. The term "and/or" in this context is merely an association describing the associated object, indicating that there may be three relationships, for example, A and / or B, which may indicate that A exists separately, and both A and B exist, respectively. B these three situations. In addition, the character "/" in this article, unless otherwise specified, generally indicates that the contextual object is an "or" relationship.

The embodiments of the present invention are further described in detail below with reference to the accompanying drawings.

Referring to FIG. 2, an embodiment of the present invention provides a storage system, which may include M controllers 201 and N interface cards 202. FIG. 2 takes M=2 and N=1 as an example.

M controllers 201 are used to control the system. Among them, the M controllers 201 include one main controller 201 and M-1 as redundant slave controllers 201, and M is a positive integer. That is, the M-1 slave controllers 201 are backups of the master controller 201, and when the master controller 201 fails, one of the slave controllers 201 can continue to operate as the master controller 201, allowing the system operation to continue.

N interface cards 202, each of which is connected to at least two controllers 201 of the M controllers 201 for transmitting or transmitting the controller 201 connected to the interface card 202. The signal from the controller 201 connected to the interface card 202 is relayed, or the interface card 202 is also used to process signals from the controller 201 connected to the interface card 202.

N is an integer less than or equal to M, that is, in the system, the number of interface cards 202 is less than or equal to the number of controllers 201. In the embodiment of the present invention, one interface card 202 corresponds to multiple controllers 201 (ie, connected to at least two controllers 201), one controller 201 may correspond to only one interface card 202, or one controller 201 may correspond to multiple Interface card 202.

The interface card 202 can have multiple types of ports that can be connected to different hardware modules. Then the signal sent by the controller 201 can be relayed through the interface card 202. For example, the format of the signal sent by the controller 201 to the interface card 202 is format 1. If the controller 201 wants to send the signal to the hardware module 1, and the corresponding signal format of the hardware module 1 is format 2, the interface card 202 can The format of the received signal is converted from format 1 to format 2 and then sent to the hardware module 1. Similarly, the signal sent by the hardware module 1 to the controller 201 is also relayed through the interface card 202.

The controller 201 can also send signals to the interface card 202, such as signals for controlling the interface card 202, or signals to the interface card 202 to inform the status of the controller 201, etc., which the interface card 202 can process. Similarly, the interface card 202 can also send a signal to the controller 201, such as a signal for notifying the controller 201 of the status of the interface card 202, and the like.

Optionally, referring to FIG. 3, the interface card 202 and the controller 201 can be connected through a PCIE (Peripheral Component Interconnect Express) bus. FIG. 3 takes M=2 and N=1 as an example.

Specifically, for an interface card 202 and a controller 201 connected thereto, There may be two PCIE buses connected, namely PCIE TX and PCIE RX, respectively called PCIE transmit bus and PCIE receive bus, where the transmission and reception are both for the controller 201. Through the PCIE TX, the controller 201 can transmit signals at high speed, and the controller 201 can receive signals through the PCIE RX. In Figure 3 there are two controllers 201, a for PCIE TX and b for PCIE RX.

In the embodiment of the present invention, the PCIE TX and the PCIE RX can be designed according to actual service requirements, and can be bandwidths of different standards such as X4, X8, and X16, and are the main service data channels between the controller 201 and the interface card 202.

Optionally, still referring to FIG. 3, between the interface card 202 and the controller 201, a serial control bus and/or a parallel control bus may be connected for transmitting control signals. In FIG. 3, a serial control bus and a parallel control bus are connected at the same time. In FIG. 3, c denotes a serial control bus, and d denotes a parallel control bus.

The serial control bus transmits a low-speed signal, and is mainly used to perform some mutual handshake between the controller 201 and the interface card 202, for example, the controller 201 can read the interface card 202 type, status, alarm information, etc., and The interface card 202 acquires status information of the controller 201, such as master-slave status information of the controller 201, that is, whether the controller 201 is the master controller 201 or the slave controller 201, and the like.

The parallel control bus transmits high-speed signals, and is mainly used for the controller 201 to control the state of the interface card 202 and upgrade the firmware, version, etc. of the interface card 202, including general basic control signals, such as in-position signals, power-on enable, and interrupt. Signal, reset signal, etc.

In general, in addition to the PCIE is an international standard protocol, the serial control bus and the parallel control bus are all buses that can be customized according to the required functions, and are not limited to a fixed form.

Optionally, referring to FIG. 3, each of the N interface cards 202 can be connected to the server, so that information exchange can be implemented between the server and the system through the interface card 202.

Optionally, still referring to FIG. 3, the storage system may further include at least one storage device, wherein each storage device is connected to at least one interface card 202, so that the controller 201 and the storage device can pass the corresponding interface card. 202 performs information interaction. The storage device may for example be a hard disk or may be other types of devices for storing information. Figure 3 is an interface card 202, and the connection The port card 202 is connected to two storage devices as an example.

In the embodiment of the present invention, each interface card 202 can have a control module, and the function of the interface card 202 is implemented by the control module.

The control module in the interface card 202 is a module that can process, control, convert or switch signals between the controller 201 and the interface card 202, and may be one or more chips or external onboard lines, according to the state of the controller 201. Or command to implement the state change and switching of the port connected to the controller 201 on the interface card 202, and also convert the signal from the PCIE protocol format to other various protocol formats, such as converting from the PCIE protocol format to the FC (Fibre Channel, Fibre Channel protocol format, from PCIE protocol format to GE (Gigabit Ethernet, Gigabit Ethernet interface) protocol format, from PCIE protocol format to SAS (Serial Attached SCSI) protocol format, and so on. The control module can also implement a mirroring (NT) function to back up data in real time with the controller 201.

In addition, it can be seen in FIG. 3 that there is a connection line between the two controllers 201, for example between the two controllers 201, there may be a connection line for transmitting mirrored data, wherein the mirrored data may be Refers to backup data, which may have a connection line for transmitting heartbeat information, and may also have a serial control bus and/or a parallel control bus.

In the embodiment of the present invention, if M>2, each of the two controllers 201 may have a connection line, for example, both may have a connection line for transmitting mirror data and a connection line for transmitting heartbeat information, and It may have a serial control bus and/or a parallel control bus.

The two controllers 201 perform state information transmission through mirroring and heartbeat, and monitor each other's status and service characteristics in real time.

Both of the above-described Figures 2 and 3 are architectural designs that are more redundant with a more commonly used dual-control memory controller and an interface card 202. When the service is running normally, the service flow is transmitted by the main controller 201 through the PCIE bus and the interface card 202. Generally, the switch card 202 converts the PCIE format message into a message of another protocol bus format, such as a message. The FC card, the GE protocol, the SAS protocol, and the like, the interface card 202 can be connected to a server port of the front end or a disk of the back end. When the main controller 201 fails, it will pass the heartbeat signal or power down between the controllers 201. The interrupt signal is notified to the slave controller 201, and the slave controller 201 can compete for the master controller 201. Because the interface card 202 is connected to both controllers 201, the interface card 202 controls the port state to perform corresponding switching, and the interface card 202 can continue to work to serve the new primary controller 201, so that the service of the interface card 202 can continue. The service is maintained with a server at the front end of the interface card 202 or a cascading box at the back end.

Referring to FIG. 4, another possible schematic diagram of the storage system is provided. In Fig. 4, M = 2 and N = 2 are taken as an example. As can be seen from Figure 4, each controller 201 can connect two interface cards 202.

For example, the controller 201 on the left is first used as the main controller 201, and the main controller 201 uses the interface card 202 on the left. When the left controller 201 fails, the controller 201 notifies the right controller 201. Of course, the controller 201 also notifies each interface card 202 connected thereto so that each interface card changes the corresponding port state. The controller 201 on the right competes for the main controller 201. At this time, the controller 201 on the right side may continue to select to use the interface card 202 on the left side, or may choose to use the interface card 202 on the right side, or may also select to use the interface card 202 on the left side and the interface card 202 on the right side. Depending on the different requirements of the different controllers 201, they may be according to pre-set rules, or may be randomly selected, and the like.

4 is a redundant design architecture of a plurality of controllers 201 and a plurality of interface cards 202. The interface card 202 and the plurality of controllers 201 implement a redundant design in an intersecting manner, and the redundant manner of the cross-connects makes the entire interface card 202 The reliability of the service is better. When the controller 201 and the interface card 202 are both faulty, as long as there are controllers 201 and interface cards 202 that have not failed, the services at the front and rear ends of the interface card 202 can be continued to avoid service interruption. The system reliability has been greatly improved.

In summary, an interface card 202 can be connected to a plurality of controllers 201 to be able to serve a plurality of controllers 201. For example, in accordance with the example of FIG. 2, first the controller 201 on the left is the master controller 201, and the interface card 202 is the controller 201 on the left. If the main controller 201 fails, the right slave controller 201 can compete for the slave controller 201, and the interface card 202 can continue to serve the controller 201 on the right side, without being associated with the controller 201. The interface card 202 connected to the controller 201 cannot be used as much as possible to ensure the continuity of the service in the interface card 202, so as to be connected with the interface card. Other devices that are connected can continue to be used, improve the reliability of the system, save hardware resources as much as possible, and improve the utilization of the interface card 202.

Referring to FIG. 5, based on the same inventive concept, an embodiment of the present invention provides a method for transmitting information through an interface card, which may be applied to the storage system shown in FIG. 2, FIG. 3, and FIG. The main flow of the method is described below.

Step 501: When the first controller as the main controller of the M controllers fails, the second controller of the M controllers competes for the new main controller. The M controllers 201 include a main controller 201 and M-1 as redundant slave controllers 201. The M controllers 201 belong to a storage system, and the storage system further includes N An interface card 202, wherein each interface card 202 is coupled to at least two controllers 201 for relaying signals transmitted by the controller 201 or transmitted to the controller 201, or for processing from the The signal of the controller 201; N is an integer less than or equal to M.

Take the system architecture in Figure 2 and Figure 3 as an example. For example, the first controller is the controller 201 on the left side of the figure, and the second controller is the controller 201 on the right side of the figure.

When the storage system starts to work, it first needs to be powered on. After the power is turned on, the system performs initialization. After the initialization is completed, the main controller 201 detects the in-position signal of the interface card 202, that is, determines whether the interface card 202 has been inserted into the correct insertion. groove. If the interface card 202 is not in place, the interface card 202 needs to be inserted. If it can be detected that the interface card is pulled low (because it is normally active low), the binary value of the type of the interface card 202 is determined by the driver. That is, it is judged whether the type of the interface card 202 is a type supported by the system. If the type of the interface card 202 is of a non-system supported type, the alarm message is sent by the driver, the red light on the interface card 202 is illuminated, and the user can reinsert the interface card 202 supported by the system.

After the interface card 202 is identified (ie, the interface card 202 has been inserted into the correct slot, the interface card 202 is not faulty, and the type of the interface card 202 is of a type supported by the system), the system sends power-on enable to the interface card 202. The signal, clock signal and reset signal cause the interface card 202 to begin operation. After the sending is completed, the interface card 202 and the main controller 201 automatically negotiate the port, and the so-called negotiation port refers to the interface card 202. In consultation with the main controller 201, information such as a transmission bandwidth, a transmission rate, and the like of a transmission channel between the interface card 202 and the main controller 201 is determined. After the negotiation is completed, the interface card 202 and the main controller 201 can communicate. The main controller 201 at this time is, for example, the controller 201 on the left side in FIG.

Optionally, in the embodiment of the present invention, before the second controller of the M controllers 201 competes for the new main controller 201, the method may further include:

The second controller receives a first fault notification message sent by the first controller, where the first fault notification message is used to notify the second controller that the first controller has a fault.

When the main controller 201 fails, the main controller 201 notifies the slave controller 201 of the failure information of the main controller 201 by the power-down interrupt signal or the heartbeat signal, that is, transmits the first failure notification message to the slave controller 201, and then The controller 201 competes for the main controller 201, and the new main controller 201 at this time refers to, for example, the controller 201 on the right side of FIG. At the same time, the original main controller 201 also notifies the interface card 202 of the failure information, or the interface card 202 can automatically detect the status of each controller 201 periodically, periodically or randomly.

At this time, the controller 201 on the right side notifies the interface card 202 through the serial control bus or the parallel control bus between the interface card 202, and the controller 201 has already competed for the master controller 201, that is, the master-slave has occurred between the controllers 201. The switching event, the working state of the port connected between the interface card 202 and the controller 201 also needs to be switched, that is, the port connected between the interface card 202 and the controller 201 on the left side can be stopped, that is, the interface card 202 and the left side are The port connected between the controller 201 enters an inactive state, or enters a mirrored state, and the port connected between the interface card 202 and the controller 201 on the right can start working, that is, the interface between the interface card 202 and the controller 201 on the right. The port enters the active state. The port on the control interface card 202 enters an inactive state, a mirrored state, or an activated state, which may be specifically performed by a control module in the interface card 202.

When the control port enters the mirroring state, it can control the port to transfer mirror data, that is, back up data. At this time, the port that enters the mirroring state is equivalent to the backup port of another port.

Further, in the embodiment of the present invention, after the second controller of the M controllers 201 competes for the new main controller 201, the method may further include:

The second controller receives the second reason sent by the third controller of the M controllers 201 The second notification message is used to notify the second controller that the third controller has a fault;

The second controller removes information of the third controller from a list of redundant controllers; wherein the redundant controller list is used to record information of each controller 201 that can be redundant.

For example, the system includes three controllers 201, which are a controller 1, a controller 2 and a controller 3, respectively. Initially, the controller 1 is a master controller 201, and the controller 2 and the controller 3 are slave controllers. 201.

When the controller 1 fails, the controller 2 competes for the main controller 201, and then the controller 3 is still the slave controller 201.

When the controller 2 is working, if the controller 3 also fails, the controller 3 notifies the controller 2 by the power-down interrupt signal or the heartbeat signal, that is, sends the second fault notification message to the controller 2, and the controller 2 The information of the controller 3 can be removed from the list of redundant controllers. At the same time, the controller 3 also notifies the interface card 202 of the fault information, or the interface card 202 can automatically detect the state of each controller 201 periodically, periodically or randomly, and the interface card 202 controls the port entry with the controller 3. Inactive or mirrored state.

In the embodiment of the present invention, when the slave controller 201 fails, the service between the master controller 201 and the interface card 202 is not affected. If there are other slave controllers 201 in the system, for example, there is also a fourth controller. Then, when the controller 2 as the main controller 201 fails, the controller 2 does not send the fault information to the faulty controller 3, that is, when the master-slave switch is required, the faulty slave controller 201 is not selected. Instead, the slave controller 201 is selected to be faultless. Of course, if there is no other slave controller 201 in the system, if the controller 2 as the master controller 201 also fails, the system may stop running.

When the fault of the controller 3 is restored, the controller 3 can notify the controller 2 by means of a heartbeat signal, etc., and the controller 2 will re-introducing the controller 3 into a selection range in which master-slave switching can be performed, that is, re-controlling the controller 3. The information is added to the list of redundant controllers. Of course, the slave controller 201 can also notify the interface card 202 when recovering from the failure of the controller 201.

In the embodiment of the present invention, the controller 201 and the interface card 202 can exchange information. For example, the controller 201 can send a notification message to the interface card 202 in real time, timing, or upon stateful transition to inform the interface card 202 of the current state of the controller 201, or the interface card 202 can also be in real time, timed, or randomly. A probe message is sent to the controller 201 to ascertain the current state of the controller 201. Of course, the interface card 202 can send a notification message to the controller 201 in real time, timing, or when there is a state transition to inform the controller 201 of the current state of the interface card 202, or the controller 201 can also be real time, timed, or randomly. A probe message is sent to the interface card 202 to ascertain the current state of the interface card 202.

Step 502: The second controller performs information relaying by using at least one of the N interface cards, where the first interface card is respectively associated with the first controller and the second controller. connection.

When the first controller fails, the second controller competes for the main controller 201. In FIG. 2 and FIG. 3, the first controller and the second controller are connected to the same interface card 202. Although the first controller fails, the interface card 202 can continue to be used. The second controller can continue to use the interface card 202, which is referred to herein as the first interface card.

Referring to FIG. 6, based on the same inventive concept, an embodiment of the present invention provides a controller fault processing method, which may be applied to the storage system shown in FIG. 2, FIG. 3, and FIG. The main flow of the method is described below.

Step 601: When the first controller that is the primary controller of the M controllers fails, the first interface card that is connected to the first controller and serves the first controller among the N interface cards Receiving a second fault notification message sent by the first controller, where the second fault notification message is used to notify the first interface card, and the first controller has a fault. The M controllers 201 include a main controller 201 and M-1 as redundant slave controllers 201, and the M controllers 201 and the N interface cards 202 belong to a storage system, wherein Each interface card 202 is coupled to at least two controllers 201 for relaying signals transmitted by the controller 201 or transmitted to the controller 201, or for processing from the controller 201 Signal; N is an integer less than or equal to M.

Taking FIG. 2 and FIG. 3 as an example, for example, the controller 201 on the left side is the main controller 201. When it fails, the first failure notification message may be sent to the slave controller 201, and at the same time, the interface card 202 may also be sent to the interface card 202. The second failure notification message is described. The specific process has been described in the introduction of FIG. 2 to FIG. 5, and will not be described here.

Optionally, in the embodiment of the present invention, after the first interface card receives the second fault notification message sent by the first controller, the method may further include:

The first interface card receives a master control notification message sent by a second controller of the M controllers 201, where the master control notification message is used to notify the first interface card, and the second controller has Competing for a new master controller 201;

The first interface card controls, according to the master control notification message, that a port connected to the second controller enters an active state to communicate with the second controller through a port connected to the second controller. .

Taking FIG. 2 and FIG. 3 as an example, for example, the controller 201 on the left side is the main controller 201. When it fails, the first failure notification message can be sent to the slave controller 201, and the controller 201 competes for the new master. The controller 201 sends a new control notification message to the first interface card, and after receiving the main control notification message, the first interface card activates and controls the new main control. The port between the devices 201 communicates with the new main controller 201. The specific process has been described in the description of FIG. 2 to FIG. 5, and details are not described herein.

Step 602: The first interface card controls, according to the second failure notification message, that a port connected to the first controller enters an inactive state to stop communication with the first controller.

Upon receiving the second failure notification message, the interface card 202 can control the port connected to the controller 201 on the left to enter an inactive state, so that communication with the controller 201 on the left can be stopped.

Of course, after the controller 201 on the right competes with the main controller 201, the master control notification message may also be sent to the interface card 202, and the interface card 202 may control the port connected to the controller 201 on the right to enter an active state, thereby The controller 201 performs communication.

The specific implementation process has been described in the introduction of FIG. 2 to FIG. 5, and will not be described here.

Referring to FIG. 7, based on the same inventive concept, an embodiment of the present invention provides another method for transmitting information through an interface card, which may be applied to the storage system shown in FIG. 2, FIG. 3, and FIG. The main flow of the method is described below.

Step 701: When a first controller that is the primary controller of the M controllers included in the storage system fails, the first controller sends a second fault notification message to the first interface card, where the second fault occurs. The notification message is used to notify the first interface card that the first controller is faulty; wherein the first interface card is connected to the first controller in the storage system including N interface cards And an interface card serving the first controller.

Taking FIG. 2 and FIG. 3 as an example, for example, the controller 201 on the left side is the main controller 201. When it fails, the first failure notification message may be sent to the slave controller 201, and at the same time, the interface card 202 may also be sent to the interface card 202. The second failure notification message is described. The specific process has been described in the introduction of FIG. 2 to FIG. 5, and will not be described here.

Step 702: The first interface card controls, according to the second failure notification message, that a port connected to the first controller enters an inactive state to stop communication with the first controller.

Upon receiving the second failure notification message, the interface card 202 can control the port connected to the controller 201 on the left to enter an inactive state, so that communication with the controller 201 on the left can be stopped.

Step 703: When the second controller of the M controllers contends to be the new master controller, the second controller sends a master control notification message to the first interface card, where the master control notification message is sent. And the second controller is used to notify the first interface card that the second controller has competed for a new primary controller; wherein the first interface card is connected to the second controller.

Step 704: The first interface card controls, according to the master control notification message, that a port connected to the second controller enters an active state to pass a port connected to the second controller and the second control. Communicate.

Continuing with FIG. 2 and FIG. 3 as an example, for example, the controller 201 on the left is the main controller 201, when it is out When the fault occurs, the first fault notification message may be sent to the slave controller 201, and then the slave controller 201 competes for the new master controller 201, and the new master controller 201 sends the new master controller 201 to the first interface card. The master control notification message, after the first interface card receives the master control notification message, activates a port with the new master controller 201 to communicate with the new master controller 201, the specific process It has been described in the introduction of Figures 2 to 5, and will not be described here.

Referring to FIG. 8, based on the same inventive concept, an embodiment of the present invention provides a controller, which may be the controller 201 in the storage system shown in FIG. 2 to FIG. 4, that is, the flow of FIG. 5-7. The controller 201 described above, in particular, the controller 201 may be the second controller described in the flow of Figures 5-7. The controller 201 can include an operation module 801 and a communication module 802.

The operation module 801 is configured to cause the controller 201 to compete for a new main controller 201 when a failure occurs in the M controllers 201 as the first controller of the main controller 201. The M controllers 201 include a main controller 201 and M-1 as redundant slave controllers 201, and the M controllers 201 belong to the storage system, and the storage system further includes N interface cards 202, wherein each interface card 202 is coupled to at least two controllers 201 for relaying signals transmitted by the controller 201 or transmitted to the controller 201, or for processing from The signal of the controller 201; N is an integer less than or equal to M.

The communication module 802 is configured to perform information relaying by using at least one of the N interface cards 202; wherein the first interface card is respectively connected to the first controller and the controller.

Optionally, in the embodiment of the present invention, the controller 201 may further include a receiving module, configured to: before the operating module 801 causes the controller 201 to compete for the new primary controller 201, receive the first controller to send The first failure notification message is used to notify the controller 201 that the first controller has a failure.

Optionally, in the embodiment of the present invention, the controller 201 may further include the receiving module and the removing module.

The receiving module is configured to receive a second fault notification message sent by a third controller of the M controllers 201 after the operating module 801 causes the controller 201 to compete for the new master controller 201. The second failure notification message is used to notify the controller 201 that the third controller has a fault;

The removal module is configured to remove information of the third controller from a list of redundant controllers; wherein the redundant controller list is used to record information of each controller 201 that can be redundant.

Referring to FIG. 9, based on the same inventive concept, an embodiment of the present invention provides an interface card, which may be an interface card 202 in the storage system shown in FIG. 2 to FIG. 4, that is, the flow of FIG. 5-7. The interface card 202 described above, in particular, the interface card 202 can be the first interface card described in the flow of Figures 5-7. The interface card 202 can include a receiving module 901 and a control module 902.

The receiving module 901 is configured to receive a second fault notification message sent by the first controller when a first controller of the M controllers 201 is faulty as the primary controller 201, where the second fault notification message is used. Informing the interface card 202 that the first controller has a fault; wherein the interface card 202 is connected to the first controller and served by the first controller in the N interface cards 202. Interface card 202;

The control module 902 is configured to control, according to the second failure notification message, a port connected to the first controller to enter an inactive state to stop communication with the first controller.

Optionally, in the embodiment of the present invention, the receiving module 901 is further configured to: after receiving the second fault notification message sent by the first controller, receive the second controller sent by the M controllers 201 The master control notification message is used to notify the interface card 202 that the second controller has contend for the new master controller 201; the control module 902 is further configured to: according to the master control notification message And controlling a port connected to the second controller to enter an active state to communicate with the second controller through a port connected to the second controller.

Based on the same inventive concept, an embodiment of the present invention further provides a storage system, which may be the storage system shown in FIG. 2 to FIG. 4, that is, the storage system described in the flowcharts of FIG. 5-7. The storage system can include a first controller, a first interface card, and a second controller. In the embodiment of the present invention, the storage system may include multiple controllers 201 and multiple interface cards 202, here only two controllers 201 (ie, the first controller and the second controller) and An interface card 202 (ie, the first interface card) is taken as an example.

The first controller is configured to send a second fault to the first interface card when the first controller that is the main controller 201 in the M controllers 201 included in the storage system fails The second failure notification message is used to notify the first interface card that the first controller has a fault; wherein the first interface card is the N interface cards 202 included in the storage system. An interface card connected to the first controller and serving the first controller;

The first interface card is configured to control, according to the second fault notification message, a port connected to the first controller to enter an inactive state to stop communication with the first controller;

The second controller is configured to send a master control notification message to the first interface card when the second controller of the M controllers 201 competes for a new master controller 201, The master control notification message is used to notify the first interface card, and the second controller has competed for the new primary controller 201; wherein the first interface card is connected to the second controller;

The first interface card is further configured to control, according to the master control notification message, a port connected to the second controller to enter an active state, by using a port connected to the second controller, and the second control Communicate.

An embodiment of the present invention provides a storage system, including: M controllers 201 for controlling the system; the M controllers 201 include a main controller 201 and M-1 as redundant slave controls. The device 201, M is a positive integer; N interface cards 202, wherein each interface card 202 is connected to at least two controllers 201 for transmitting signals transmitted by the controller 201 or transmitted to the controller 201 Performing a relay, or for processing a signal from the controller 201; N is an integer less than or equal to M.

In the embodiment of the present invention, an interface card 202 is connected to at least two controllers 201. If one of the controllers 201 connected to an interface card 202 fails, the interface card 202 can stop serving the controller 201. At the same time, the interface card 202 is also connected to other controllers 201, and can continue to serve other controllers 201. In this way, even if the controller 201 fails, the interface card 202 can continue to serve other controllers 201 as long as there is no fault, and can continue to be used. Compared with the prior art, the services in the interface card 202 are not interrupted, and Other hardware connected to the interface card 202 The device can also continue to transmit information through the interface card 202, thereby ensuring the reliability of the system.

Moreover, the interface card 202 and the device connected to the interface card 202 can continue to be used, which also saves hardware resources to some extent and improves the utilization of the interface card 202. Moreover, with the technical solution in the embodiment of the present invention, the number of the interface cards 202 can be reduced to some extent, the system structure tends to be simple, and the volume of the system is reduced.

It will be clearly understood by those skilled in the art that for the convenience and brevity of the description, only the division of each functional module described above is exemplified. In practical applications, the above function assignment can be completed by different functional modules as needed. The internal structure of the device is divided into different functional modules to perform all or part of the functions described above. For the specific working process of the system, the device and the unit described above, reference may be made to the corresponding process in the foregoing method embodiments, and details are not described herein again.

In the several embodiments provided by the present application, it should be understood that the disclosed system, apparatus, and method may be implemented in other manners. For example, the device embodiments described above are merely illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there may be another division manner, for example, multiple units or components may be used. Combinations can be integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.

The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

The integrated unit is implemented in the form of a software functional unit and sold as a standalone product Or when used, it can be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application, in essence or the contribution to the prior art, or all or part of the technical solution may be embodied in the form of a software product stored in a storage medium. A number of instructions are included to cause a computer device (which may be a personal computer, server, or network device, etc.) or a processor to perform all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like. .

The above embodiments are only used to describe the technical solutions of the present application in detail, but the description of the above embodiments is only for helping to understand the method and the core idea of the present invention, and should not be construed as limiting the present invention. Those skilled in the art will be able to devise variations or alternatives within the scope of the present invention within the scope of the present invention.

Claims (16)

A storage system, comprising: M controllers for controlling the system; the M controllers include a main controller and M-1 slave controllers as redundant, and M is a positive integer; N interface cards, wherein each interface card is coupled to at least two controllers for relaying signals transmitted by the controller, or signals transmitted to the controller, or for processing from the controller Signal; N is an integer less than or equal to M. The method of claim 1 wherein said interface card is coupled to said controller via a fast peripheral interconnect PCIE bus. The method of claim 2, wherein a serial control bus and/or a parallel control bus are further coupled between the interface card and the controller for transmitting control signals. A method according to any one of claims 1 to 3, wherein the system further comprises at least one storage device, wherein each storage device is coupled to at least one interface card to cause the controller and the storage device Information exchange through the corresponding interface card. A method for transmitting information through an interface card, comprising: When the first controller as the primary controller of the M controllers fails, the second controller of the M controllers competes for the new primary controller; The second controller performs information relaying by using at least one of the N interface cards; wherein the first interface card is respectively connected to the first controller and the second controller. The method of claim 5, wherein before the second one of the M controllers competes for the new primary controller, the method further comprises: The second controller receives a first fault notification message sent by the first controller, where the first fault notification message is used to notify the second controller that the first controller has a fault. The method according to claim 5 or 6, wherein after the second controller of the M controllers competes for the new master controller, the method further includes: The second controller receives a second fault notification message sent by a third controller of the M controllers, where the second fault notification message is used to notify the second controller, the third controller A failure has occurred; The second controller removes information of the third controller from a list of redundant controllers; wherein the redundant controller list is used to record information that can be used as redundant controllers. A controller fault processing method, comprising: When the first controller as the primary controller of the M controllers fails, the first interface card connected to the first controller and serving the first controller of the N interface cards receives the a second fault notification message sent by the first controller, where the second fault notification message is used to notify the first interface card, where the first controller has a fault; The first interface card controls the port connected to the first controller to enter an inactive state according to the second failure notification message to stop communication with the first controller. The method of claim 8, wherein after the first interface card receives the second failure notification message sent by the first controller, the method further includes: The first interface card receives a master control notification message sent by a second controller of the M controllers, where the master control notification message is used to notify the first interface card, and the second controller has been contending For the new primary controller; The first interface card controls, according to the master control notification message, that a port connected to the second controller enters an active state to communicate with the second controller through a port connected to the second controller. . A method for transmitting information through an interface card, comprising: When the first controller that is the primary controller of the M controllers included in the storage system fails, the first controller sends a second failure notification message to the first interface card, where the second failure notification message is used. The first interface card is configured to notify the first interface card that the first controller is faulty; wherein the first interface card is connected to the first controller in the N interface cards included in the storage system, and is An interface card served by the first controller; The first interface card controls, according to the second failure notification message, that a port connected to the first controller enters an inactive state to stop communication with the first controller; When the second controller of the M controllers contends to be the new master controller, the second controller sends a master control notification message to the first interface card, where the master control notification message is used to notify The first interface card, the second controller has competed for a new primary controller; wherein the first interface card is connected to the second controller; The first interface card controls, according to the master control notification message, that a port connected to the second controller enters an active state to communicate with the second controller through a port connected to the second controller. . A controller, comprising: An operation module, when the first controller as the main controller of the M controllers fails, causing the controller to compete for a new main controller; And a communication module, configured to perform information relaying by using at least one of the N interface cards; wherein the first interface card is respectively connected to the first controller and the controller. The controller according to claim 11, wherein said controller further comprises a receiving module, configured to: receive said first before said operating module causes said controller to compete for a new primary controller And a first fault notification message sent by the controller, where the first fault notification message is used to notify the controller that the first controller has a fault. The controller according to claim 11 or 12, wherein the controller further comprises a receiving module and a removing module; The receiving module is configured to: after the operating module causes the controller to compete for a new primary controller, receive a second fault notification message sent by a third controller of the M controllers, where the second fault notification is sent a message is used to notify the controller that the third controller has failed; The removal module is configured to remove information of the third controller from a list of redundant controllers; wherein the redundant controller list is used to record information that can be used as redundant controllers. An interface card, comprising: a receiving module, configured to receive a second fault notification message sent by the first controller when a first controller that is the master controller of the M controllers fails, where the second fault notification message is used to notify the In the interface card, the first controller is faulty; wherein the interface card is an interface card of the N interface cards that is connected to the first controller and serves the first controller; And a control module, configured to control, according to the second fault notification message, a port connected to the first controller to enter an inactive state to stop communication with the first controller. The interface card according to claim 14, wherein the receiving module is further configured to: after receiving the second failure notification message sent by the first controller, receive a second one of the M controllers a master control notification message sent by the controller, the master control notification message is used to notify the interface card, and the second controller has competed for a new master controller; The control module is further configured to: according to the master control notification message, control a port connected to the second controller to enter an activation state, to connect a port connected to the second controller with the second controller Communicate. A storage system, comprising: a first controller, configured to send a second fault message to the first interface card when the first controller that is the master controller in the M controllers included in the storage system fails, the second The fault notification message is used to notify the first interface card that the first controller has a fault; wherein the first interface card is among the N interface cards included in the storage system and the first control An interface card that is connected to and serves the first controller; The first interface card is configured to control, according to the second fault notification message, a port connected to the first controller to enter an inactive state to stop communication with the first controller; a second controller, configured to send a master control notification message to the first interface card when the second controller of the M controllers contends to be a new master controller, the master control notification message For notifying the first interface card, the second controller has competed for a new primary controller; wherein the first interface card is connected to the second controller; The first interface card is further configured to control, according to the master control notification message, with the second controller The connected port enters an active state to communicate with the second controller through a port connected to the second controller.

Images