CN103814352B - A kind of virtual unit method for reconstructing and device - Google Patents

A kind of virtual unit method for reconstructing and device Download PDF

Info

Publication number
CN103814352B
CN103814352B CN201280005835.5A CN201280005835A CN103814352B CN 103814352 B CN103814352 B CN 103814352B CN 201280005835 A CN201280005835 A CN 201280005835A CN 103814352 B CN103814352 B CN 103814352B
Authority
CN
China
Prior art keywords
described
vtl
virtual unit
address
node
Prior art date
Application number
CN201280005835.5A
Other languages
Chinese (zh)
Other versions
CN103814352A (en
Inventor
汪野
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to PCT/CN2012/088096 priority Critical patent/WO2014101211A1/en
Publication of CN103814352A publication Critical patent/CN103814352A/en
Application granted granted Critical
Publication of CN103814352B publication Critical patent/CN103814352B/en

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00Application independent communication protocol aspects or techniques in packet data networks
    • H04L69/40Techniques for recovering from a failure of a protocol instance or entity, e.g. failover routines, service redundancy protocols, protocol state redundancy or protocol service redirection in case of a failure or disaster recovery
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network-specific arrangements or communication protocols supporting networked applications
    • H04L67/10Network-specific arrangements or communication protocols supporting networked applications in which an application is distributed across nodes in the network
    • H04L67/1002Network-specific arrangements or communication protocols supporting networked applications in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers, e.g. load balancing
    • H04L67/1034Reaction to server failures by a load balancer

Abstract

The invention provides a kind of virtual unit reconstruction technique, in virtual unit method for reconstructing, second VTL node takes over the entry address of described first VTL node, described second VTL node obtains the configuration information of described virtual unit, device address, entry address and backup server exit address, second VTL node uses described configuration information to generate and rebuilds virtual unit, distribute to described device address rebuild virtual unit, set up the mapping relations rebuild between virtual unit and VTL equipment entry address, backup server exit address.The virtual unit that the application present invention rebuilds, can smooth and take over the business that former virtual unit is backing up.

Description

A kind of virtual unit method for reconstructing and device

Technical field

The present invention relates to field of storage, particularly with regard to a kind of virtual magnetic tape library VTL technology.

Background technology

Along with the high speed development of computer technology, the data volume of the data center of each mechanism, enterprise etc. is blast The growth of formula, correspondingly, the time needed for backing up the data of each data center is also significantly increased.The most each machine The data center of structure, enterprise etc. introduces the storage of backup performance higher Disk Backup, such as virtual magnetic magnetic one after another Tape pool (VTL, Virtual Tape Library) substitutes traditional physics magnetic tape library backup storage, with contracting Short data BACKUP TIME, reduces the backup traffic impact on the business of production.

VTL product is generally made up of server, VTL equipment and storage array, and VTL equipment can be run VTL program.The data needing backup are sent to VTL equipment by server, and VTL equipment backs up data to In storage array.VTL cluster is a kind of cluster product, comprises multiple stage VTL equipment in VTL cluster, every VTL equipment is properly termed as again a node, and multiple nodes can run VTL application software simultaneously.And single-unit Point form (only comprising a station server) VTL product is compared, and each VTL equipment can work simultaneously, because of This VTL collection group energy significantly shortens the Backup and Restore time, reduces backup traffic to a greater degree to the business of production Impact.

VTL cluster, while properties of product are substantially improved, also brings the availability issue of group system. In cluster, any one one malfunctions all will cause the fault of whole group system, the joint in cluster VTL Point is the most, and the probability of cluster VTL interruption of work because of fault is the biggest.

In prior art, after VTL cluster has nodes break down.Other nodes can be taken over The business of this malfunctioning node, during adapter, those that malfunctioning node is currently running Backup traffic can run failure.

Summary of the invention

The present invention provides a kind of virtual unit method for reconstructing and device, can smooth virtual unit before adapter is rebuild The business being currently running.

First aspect, the embodiment of the present invention provides a kind of virtual unit method for reconstructing, virtual magnetic tape library VTL Cluster includes the first VTL node and the second VTL node, backup server and described first VTL node And described second VTL node sets up communication connection, storage device and described first VTL node and described Second VTL node sets up communication connection, and the first VTL node creates at least one virtual unit, and first The device address of the virtual unit in VTL node is unique in VTL cluster, and the method includes: described second VTL node takes over the entry address of described first VTL node, and described second VTL node obtains described void Propose standby configuration information, device address, entry address and backup server exit address;Wherein, institute Before stating the entry address that the second VTL node takes over described first VTL node, farther include: described the One VTL node searches the device address not used by arbitrary VTL equipment in described VTL cluster;Described One VTL node uses virtual unit described in the device address information creating found;Second VTL node uses Described configuration information generates rebuilds virtual unit;Distribute to described device address rebuild virtual unit, set up Mapping relations between described reconstruction virtual unit and VTL equipment entry address, backup server exit address.

Second aspect, the embodiment of the present invention provides a kind of virtual unit reconstructing device, is positioned at the second VTL node In, virtual magnetic tape library VTL cluster includes the first VTL node and described second VTL node, backup Server and described first VTL node and described second VTL node set up communication connection, storage device with Described first VTL node and described second VTL node set up communication connection, and the first VTL node creates Have at least one virtual unit, the device address of the virtual unit in the first VTL node in VTL cluster only One, described virtual unit reconstructing device includes: creation module, is not appointed in described VTL cluster for searching The device address that one VTL equipment uses, creates described virtual unit for described second VTL equipment;Obtain Delivery block, for taking over the entry address of described first VTL node, and obtains joining of described virtual unit Confidence breath, device address, entry address and backup server exit address;Rebuild module, for second VTL node uses described configuration information to generate and rebuilds virtual unit;Enable module, for by described equipment ground Virtual unit is distributed to rebuild in location, and sets up described reconstruction virtual unit and VTL equipment entry address, standby Mapping relations between part server exit address.

The third aspect, the embodiment of the present invention provides a kind of second VTL node, virtual magnetic tape library VTL cluster Including the first VTL node and the second VTL node, backup server and described first VTL node and Described second VTL node sets up communication connection, storage device and described first VTL node and described second VTL node sets up communication connection, and the first VTL node creates at least one virtual unit, a VTL The device address of the virtual unit on node is unique in VTL cluster, and described second VTL node includes CPU, the first communication interface communicated with CPU, the second communication interface communicated with CPU, and and CPU The internal memory of communication: described first communication interface, for communicating with described backup server;Described second communication Interface, is used for and described memory device communication;Described internal memory, is used for storing computer instruction;Described CPU, Couple with described internal memory, for performing following operation by running described computer instruction: take over described the The entry address of one VTL node, obtains the configuration information of described virtual unit, device address, entry address And backup server exit address;Use described configuration information to generate and rebuild virtual unit;By described equipment Address is distributed to rebuild virtual unit, sets up described reconstruction virtual unit and VTL equipment entry address, backup Mapping relations between server exit address.Described CPU is further used for performing: make described reconstruction virtual Equipment, by the entry address of described reconstruction virtual unit, receives described backup server and uses described backup clothes The backup information that business device exit address sends;Make described reconstruction virtual unit that Backup Data to mail to described storage Device backup.

After the application embodiment of the present invention, rebuild the configuration information of virtual unit, device address, enter Port address and backup server exit address are all identical with rebuilding front virtual unit, therefore rebuild The backup traffic that front virtual unit is processing can be switched to virtual unit after reconstruction smoothly, no Loss of data can occur.

Accompanying drawing explanation

In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to enforcement In example or description of the prior art, the required accompanying drawing used is briefly described, it should be apparent that, describe below In accompanying drawing be only some embodiments of the present invention, for those of ordinary skill in the art, do not paying On the premise of going out creative work, it is also possible to obtain other accompanying drawing according to these accompanying drawings.

Fig. 1 is embodiment of the present invention application scenarios topological diagram;

Fig. 2 is virtual unit method for reconstructing embodiment flow chart;

Fig. 3 is virtual unit reconstructing device embodiment schematic diagram;

Fig. 4 is VTL node embodiment schematic diagram.

Detailed description of the invention

Below in conjunction with the accompanying drawing in the embodiment of the present invention, technical scheme is carried out clear, complete Ground describes, it is clear that described embodiment is only a part of embodiment of the present invention rather than whole realities Execute example.Based on the embodiment in the present invention, those of ordinary skill in the art are not before making creative work Put the every other embodiment obtained, broadly fall into the scope of protection of the invention.

VTL technology can be with emulated physics tape, and VTL allows client to configure virtual tape drive, virtual magnetic Tape drum and appointment cartridge capacity.The equipment realizing VTL technology realizes the main equipment of VTL technology in other words, We are referred to as VTL product or VTL equipment, the VTL product of different vendor, support that virtual tape drives The maximum number of device is the most different.Different from physics magnetic tape library, physics magnetic tape library needs buy and install volume Outer tape drive, but can increase virtual by change software configuration (configuration) for VTL Tape drive, and this need not spend any extra hardware cost.VTL equipment can be referred to as node, In VTL technology, the situation of only one of which VTL equipment, we are properly termed as VTL single node form; The situation of multiple VTL cooperation equipment, we are referred to as VTL multinode form, or address is VTL Cluster.In VTL cluster, protection between VTL node, can be provided, when certain VTL node fault, By the business on the VTL node taking over fault node not broken down.

As it is shown in figure 1,3 backup servers 11 are by switch 12 and VTL equipment 131 and VTL Equipment 132 communicates to connect, and VTL equipment 131, VTL equipment 132 all communicate to connect with storage array 14, VTL equipment 131 and VTL equipment 132 form VTL cluster 13, and VTL system 1 includes VTL cluster 13 and storage array 14.Communication connection between equipment can pass through optical-fibre channel (FC, Fiber Channel) Or the technology such as Ethernet (Ethernet) realize, and subsequent embodiment of the present invention is situated between as a example by FC technology Continue.When using FC, described switch 12 is FC switch, described backup server 11 and VTL Equipment 131, VTL equipment 132 are by World Wide Port title (WWPN, World Wide Port Name) Address realizes the communication with FC switch 12.The feature of VTL cluster 13 is VTL equipment 131 and VTL When some in equipment 132 breaks down, one can be switched the business to and carry out in VTL equipment.Need It is noted that various embodiments of the present invention are only illustrated with the situation of 2 VTL equipment, actually VTL Cluster can be made up of more VTL equipment.

VTL equipment can be referred to as node, and therefore VTL equipment 131 is properly termed as node 131, and VTL sets Standby 132 are properly termed as node 132, and 2 nodes are by the arbitration in heart beating network and rear end storage array 14 Information mutually monitors the duty of the other side, when a nodes break down cannot normally work, and Ke Yiyou Another one node takes over its business.It is equipped with virtual magnetic tape library equipment on node 131 and node 132 Or virtual magnetic magnetic tape controller equipment, virtual magnetic tape library equipment or virtual magnetic magnetic tape controller equipment can be referred to as void Proposing standby, for a user, these virtual units similarly are that real physical tape equipment is the most available.

Rear end storage can be the locally stored of VTL node, say, that rely on VTL node self to deposit Storage;Can also be the external storage interconnected by network and VTL node, the storage array 14 in Fig. 1 be The example of external storage.Time actually used, can create multiple LUN in rear end stores, these LUN can Directly to be used by the VTL software in VTL node, it is also possible to first created file wherein by VTL node System, is then available to VTL software and uses.The SCSI command of Server transmission and data are by virtual unit After receiving, virtual unit calls rear end read-write interface, such as block read-write interface or file read-write interface etc., Write data into rear end storage.

Those skilled in the art it should be understood that the parameters of virtual unit can be described by configuration information, As a example by virtual tape library equipment, its configuration information includes tape library ID, tape library model, tape library name The information such as title and tape library serial number, it is determined that the configuration information of a virtual unit just can re-create This virtual unit.In prior art, the node in VTL cluster is in order to carry out internodal protection, each Node can prestore the configuration information of all virtual units on other nodes, when there being nodes break down, The virtual unit on malfunctioning node can be rebuild by these configuration informations.After virtual unit is rebuild, for rebuilding After dummy node distributing equipment address, and re-establish this virtual unit and VTL equipment entry address, After mapping relations between backup server exit address, the virtual unit after this is rebuild just can accept quilt Backup server uses, and also just says and can accept the backup request of user and perform backup traffic.

But, if when virtual unit is carrying out backup traffic, node there occurs fault, due to each VTL Equipment be virtual unit distribution address process be separate, therefore rebuild after virtual unit can be by again Distribution virtual unit address.And backup server still virtual unit address generation Backup Data before continuing to use, Therefore the business that these are backing up can be failed, it is impossible to is normally sent to rebuild virtual unit.Virtual unit Address can be small computer system interface (Small Computer System Interface, SCSI) Location, such as SCSI tlv triple address, tlv triple address refer to passage channel, equipment device and Tri-information of target target carry out the address expressed.In the present embodiment, tlv triple address is at backup server In be each virtual unit distribution address.These addresses are exactly that each that see from backup server is virtual The address of equipment, so that backup server can be found by tlv triple address and use in VTL node Virtual unit.

For example assume that Fig. 1 interior joint 131 establishment has: " 131 virtual tape 0 [0:0:0] " and " 131 Virtual magnetic magnetic tape controller 0 [0:0:1] " the two virtual unit;Assume that node 132 establishment has: " 132 virtual magnetic Tape library 0 [0:0:0] ", " 132 virtual magnetic magnetic tape controller 0 [0:0:1] " and " 132 virtual magnetic magnetic tape controller equipment 1 [0:0:2] " totally three virtual units, as shown in table 1.

Wherein, in virtual unit " 131 virtual tapes [0:0:0] ", " 131 " refer to belonging to this virtual unit Nodename, " virtual tape " is the type of virtual unit, and " 0 " is the numbering of virtual unit, these letter Breath is all the configuration information of this virtual tape, it should be noted that more than these several of configuration information, due to Being only for example, remaining configuration information is not the most listed.[0:0:0] is the SCSI ground of this virtual tape Location.

The virtual unit of node 131 The virtual unit that node 132 creates 131 virtual tapes 0 [0:0:0] 132 virtual magnetic tape libraries 0 [0:0:0] 131 virtual magnetic magnetic tape controllers 0 [0:0:1] 132 virtual magnetic magnetic tape controllers 0 [0:0:1] 132 virtual magnetic magnetic tape controller equipment 1 [0:0:2]

Table 1

A kind of way is, when node 132 detect node 131 break down cannot work on time, at joint The virtual magnetic tape library in malfunctioning node, virtual magnetic magnetic tape controller equipment is re-created on point 132, and for rebuilding Virtual unit distribution scsi address.As shown in table 2 after reconstruction.

The virtual unit of node 131 The virtual unit that node 132 creates 132 virtual magnetic tape libraries 0 [0:0:0] 132 virtual magnetic magnetic tape controllers 0 [0:0:1] 132 virtual magnetic magnetic tape controller equipment 1 [0:0:2] 131 virtual tapes 0 [0:0:3] 131 virtual magnetic magnetic tape controllers 0 [0:0:4]

Table 2

As shown in table 2, two virtual units on node 131 are rebuild on node 132, VTL Node name, these two information of virtual unit type and numbering keep constant.Node 132 sets for the virtual of reconstruction When back-up joins virtual address, detect that device address [0:0:0] and [0:0:1] the two scsi address are by this The existing virtual unit of equipment takies, it is impossible to the most rebuilt virtual unit uses, in the present embodiment, and node 132 are assigned with [0:0:3] and [0:0:4] the two new equipment address for the virtual unit rebuild.

After distributing device address, rebuild these virtual magnetic tape libraries and former VTL equipment entry address, former standby Mapping relations between part server exit address, e.g. IP address, these addresses or virtual World Wide Port Title (World Wide Port Name, WWPN) address, these scripts are controlled by node 131, fault Taken over by node 132 after generation.After mapping relations have been set up, it is achieved that the adapter of malfunctioning node, The virtual unit rebuild can accept new backup request.But the backup being currently running on malfunctioning node is still It is to carry out with original equipment address, say, that server is still according to " " 131 virtual tape 0 [0:0:0] ", " " 131 virtual magnetic magnetic tape controller 0 [0:0:1] " write data, owing to the virtual unit device address rebuild is sent out Having given birth to change, therefore write operation can be failed, and server can repeatedly retry until exceeding default number of retries Or retry the time, but all can end in failure every time.In order to avoid writing the loss of data unsuccessfully caused, Must rescan in backup server (Backup Server) after failover/fault recovery completes The virtual unit mapped, then restarts the data backup/recovery tasks of failure, increases the resource consumption of system, Add the risk of loss of data.

When, after the fault recovery of node 131, the business taken over is transferred again node 131, node 131 Redistribute for the two equipment distributing equipment address, it is difficult to the device address before guarantee and fault keeps Unanimously, therefore in failover procedure, will also tend to cause the failure of backup.

Other situations beyond fault or fault recovery, if need in a VTL node is virtual Equipment is reconstituted on another one virtual unit, also there will be same problem.Prior art can only realize joining Confidence breath keeps constant, but device address is the independent information outside configuration information, distributes, manages, ties up The mode protected is all different from configuration information, and therefore prior art cannot realize rebuilding the equipment ground of front virtual unit Location with rebuild after virtual unit device address keep constant.

Embodiment one

With reference to Fig. 1, a kind of virtual unit method for reconstructing, apply in VTL cluster 13, VTL cluster 13 Including the first VTL node 131 and the second VTL node 132, backup server 11 is set by exchange Standby 12 set up with the first VTL node 131 and the second VTL node 132 and to communicate to connect.VTL system 1 Collectively constituted by cluster 13 and storage device 14.Storage device 14 and the first VTL node and second VTL node sets up communication connection, and the first VTL node creates at least one virtual unit, a VTL The device address of the virtual unit on node is unique in VTL cluster.According to backup server 11 and VTL The difference of communication mode between cluster 13, switching equipment 12 can be IP-based switching equipment or based on The switching equipment of WWPN, storage device 14 can be Redundant Array of Inexpensive Disc RAID group or and VTL Integration of equipments hard disk together.

The device address of the virtual unit in the first VTL node is unique in VTL cluster, the most just says VTL In cluster, all VTL node the most do not use the establishment virtual unit of this device address.Can have a variety of Way realizes the unique of virtual unit address, such as VTL node in the virtual unit distribution address new for Time, first inquire about the address of the existing virtual unit of other VTL node in this VTL node and cluster, select not The address used is allocated;Can also set up cluster address allocation table, each equipment is new for one During virtual unit distribution address, the address being not used by selection table is allocated, and after being assigned, The address of this sub-distribution is recorded as in table the address having been used;Further, it is also possible to be each VTL joint The point different address field of distribution, such as first equipment can distribute the set of 1000 addresses compositions, and second Individual equipment can distribute the set of other 1000 addresses composition, by that analogy, owing to not existing between set Identical element, will not conflict in the device address between VTL equipment in cluster.The virtual unit of the present embodiment Method for reconstructing comprises the steps, can be found in Fig. 2.

S21, described first VTL node are searched not by setting that arbitrary VTL equipment in described VTL cluster uses Standby address;Described first VTL node uses virtual unit described in the device address information creating found.

This is an optional step, performs this step and can ensure that the address of any two virtual unit in cluster Do not repeat.

S22, described second VTL node 131 detect whether the first VTL node 132 node breaks down.

Detection method has a variety of, such as by heart beating network, when the second VTL node 132 is in certain time In when receiving the heartbeat message of the first VTL node 131 not over heart beating network, it is determined that the first VTL node 131 there occurs fault;Can also be monitored, when storage device 14 monitors first by storage device 14 When VTL node 131 breaks down, notify the second VTL node 132.

S23, after detecting that described first VTL node 131 breaks down, described second VTL node 132 Taking over the entry address of described first VTL node 131, described second VTL node 132 obtains described virtual The configuration information of equipment, device address, entry address and backup server exit address.

Data backup to be realized, need to set up backup server exit address, VTL equipment entry address and Mapping between virtual unit, Backup Data mails to VTL node by the exit address of backup server 11 131, then mail to, by the entry address of VTL node 131, the virtual unit that VTL node 131 is set up, and By the virtual unit of VTL node 131, Backup Data is mail to storage device 14, finally realize the backup of data. The method obtaining these data can have multiple, such as can be by these after each virtual unit has created Data record is in a unified platform, and VTL node 132 obtains these information by this platform;Or Each VTL node preserves a each virtual unit configuration information of remaining VTL node, device address, Entry address and backup server exit address, and keep the renewal of these data.First VTL node 131 After fault, backup server 11 and storage device 14 also can receive fault cues, and receiving event respectively Carry out after barrier prompting and substituted the standard of virtual unit in the first VTL node by the virtual unit in the second VTL node Standby.

S24, the second VTL node use described configuration information to generate and rebuild virtual unit.

As it was previously stated, configuration information is used to describe the information of virtual unit, the such as type of a virtual unit Number, virtual unit title and the serial number of virtual unit, wherein the model of virtual unit refers to be set by virtual The model of standby virtual physical equipment, the B money equipment that such as company A produces.Obtain configuration information with The most just can re-create the virtual unit being out of order on node.It should be strongly noted that this configuration letter Breath does not include device address, entry address and backup server exit address.

S25, described device address is distributed to rebuild virtual unit, set up described reconstruction virtual unit and VTL Mapping relations between equipment entrance address, backup server exit address.

This device address is the address that in the first VTL node 131, virtual unit uses, due to each VTL Address on node is not repeated, therefore second existing this address of virtual unit use useless of VTL node 132, This address can be distributed to rebuild virtual unit and use by the second VTL node 132.Equipment entrance address, Backup server exit address also virtual unit front with reconstruction maps.Hence set up reconstruction virtual unit And after the mapping relations between VTL equipment entry address, backup server exit address, rebuild virtual unit Configuration information, device address, VTL equipment entry address, backup server exit address all with rebuild before Equally, the position simply existed has the first VTL node 131 to become the second VTL node 132.

S26, described reconstruction virtual unit, by the entry address of described reconstruction virtual unit, receive described backup Server uses the backup information that described backup server exit address sends;Described reconstruction virtual unit is by standby Number evidence mails to described storage device and backs up.After completing step S25, rebuilding virtual unit can quilt Backup server 11 uses and is managed by the second VTL node 132, and step S26 is to backup server How 11 use a kind of citing rebuilding virtual unit.

In this step S26, if fault occurs, the virtual unit in the first VTL node 131 is Perform backup tasks, this backup tasks need upon execution designated equipment address, VTL equipment entry address, Backup server exit address.Due to rebuild after the device address of virtual unit, VTL equipment entry address, Backup server exit address is all as before reconstruction, and the virtual unit after therefore rebuilding can continue executing with this Individual backup traffic.It should be noted that before step S24 completes, this virtual unit is to use , this time, backup server 11 can receive the unsuccessful prompting of backup tasks, at this moment backup server 11 Can again attempt to perform backup tasks, until step S24 has performed, i.e. can appoint with successful execution backup Business, no longer receives the unsuccessful prompting of backup tasks.

In embodiment one, if certain one malfunctions i.e. in cluster, the related service that it runs can To be taken over by other nodes in cluster.For the backup traffic being currently running, backup server 11 still edge Sending Backup Data by the parameters of virtual unit before fault, these parameters include configuration information, equipment ground Location, entry address and backup server exit address;And rebuild the configuration information of virtual unit, equipment ground Location, entry address and backup server exit address are all identical with rebuilding front virtual unit, before therefore rebuilding The backup traffic that virtual unit is processing can be switched to virtual unit after reconstruction smoothly, and data will not occur After loss adapter completes, whole cluster is still within the state that can normally work.Thus may insure that whole Cluster cannot normally will not work because of individual node fault, and the availability of cluster product has been effectively ensured.And And, during failover, the backup/restoration task that malfunctioning node is currently running will not be failed, and fault connects The process of pipe is more smooth.After failover completes, it is not necessary to rescan mapping in Backup Server Virtual unit, then restart the data backup/recovery tasks of failure again to perform.No data can be realized The service protection lost.

Embodiment two

The present embodiment is about failure recovery operation.When after the first VTL node 131 fault recovery in Fig. 1, The virtual unit migrating to the second VTL node 132 can be moved back to the first VTL node 131 again.

The principle of the present embodiment and step S22 of embodiment one, S23, S24, S25 are essentially identical.Institute is different Part is, in step S22, is after detecting that described first VTL node 131 breaks down, described Second VTL node 132 takes over the entry address of described first VTL node 131, described 2nd VTL joint The point 132 acquisition configuration information of described virtual unit, device address, entry address and backup server go out Port address.And in the corresponding steps of embodiment two, be after the first VTL node 131 fault recovery being detected, Described first VTL node 132 takes over the entry address taken over by the second VTL node 131 again.

Those skilled in the art, on the basis of having studied embodiment one carefully, can be readily accomplished embodiment two, the most not Repeat again.This is external understood embodiment one, embodiment two on the basis of, those skilled in the art should be also Other that how to be moved to by the virtual unit in any one VTL node in VTL cluster can be derived VTL node, this migration can be because fault or fault recovery, it is also possible to is other situations, such as The migration that user specifies.When the migration being not as fault or fault recovery and occur, detect whether to occur The step whether fault or detection fault are recovered is not necessarily.

Embodiment three

Based on embodiment one and embodiment two, one's duty invention also provides for a kind of virtual unit reconstructing device Embodiment.Virtual unit reconstructing device is positioned in node, such as in the second VTL node below, namely Say that virtual unit reconstructing device can be integrated in node, to have corresponding function software or hardware, It can also be node itself.

See accompanying drawing 3, in the present embodiment, virtual unit reconstructing device 3, it is positioned in the second VTL node, Virtual magnetic tape library VTL cluster includes the first VTL node and described second VTL node, backup services Device sets up communication connection with described first VTL node and described second VTL node, and storage device is with described First VTL node and described second VTL node set up communication connection, first VTL node create have to A few virtual unit, the device address of the virtual unit in the first VTL node is unique in VTL cluster, Virtual unit reconstructing device includes: acquisition module 33, reconstruction module 34 and enable module 35.Optionally, Creation module 31 and fault detection module 32 can also be included.

Creation module 31, for searching the equipment ground not used by arbitrary VTL equipment in described VTL cluster Location, creates described virtual unit for described second VTL equipment.

Fault detection module 32, and whether creation module 31 connects, occur for detection the first VTL node Fault, and after the first VTL node fault being detected, start acquisition module 33.

Acquisition module 33, and fault detection module 32 connects, for taking over entering of described first VTL node Port address, and obtain the configuration information of described virtual unit, device address, entry address and backup clothes Business device exit address.

Data backup to be realized, need to set up backup server exit address, VTL equipment entry address and Mapping between virtual unit, Backup Data mails to VTL node by the exit address of backup server 11 131, then mail to, by the entry address of VTL node 131, the virtual unit that VTL node 131 is set up, and By the virtual unit of VTL node 131, Backup Data is mail to storage device 14, finally realize the backup of data. The method obtaining these data can have multiple, such as can be by these after each virtual unit has created Data record is in a unified platform, and VTL node 132 obtains these information by this platform;Or Each VTL node preserves a each virtual unit configuration information of remaining VTL node, device address, Entry address and backup server exit address, and keep the renewal of these data.First VTL node 131 After fault, backup server 11 and storage device 14 also can receive fault cues, and receiving event respectively Carry out after barrier prompting and substituted the standard of virtual unit in the first VTL node by the virtual unit in the second VTL node Standby.

Rebuild module 34, and acquisition module 33 connects, use described configuration information for the second VTL node Generate and rebuild virtual unit.

As it was previously stated, configuration information is used to describe the information of virtual unit, the such as type of a virtual unit Number, virtual unit title and the serial number of virtual unit, wherein the model of virtual unit refers to virtual unit The model of virtual physical equipment.Just can re-create after obtaining configuration information and be out of order on node Virtual unit.It should be strongly noted that this configuration information do not include device address, entry address with And backup server exit address.

Enable module 35, and reconstruction module 34 connects, for distributing to rebuild virtual setting by described device address Standby, and set up described reconstruction virtual unit and VTL equipment entry address, backup server exit address it Between mapping relations.

This device address is the address that in the first VTL node 131, virtual unit uses, due to each VTL Address on node is not repeated, therefore second existing this address of virtual unit use useless of VTL node 132, This address can be distributed to rebuild virtual unit and use by the second VTL node 132.Equipment entrance address, Backup server exit address also virtual unit front with reconstruction maps.Hence set up reconstruction virtual unit And after the mapping relations between VTL equipment entry address, backup server exit address, rebuild virtual unit Configuration information, device address, VTL equipment entry address, backup server exit address all with rebuild before Equally, the position simply existed has the first VTL node 131 to become the second VTL node 132.

Enable module 35, it is also possible to for stating the virtual unit entry address by described reconstruction virtual unit, Receive the backup information that described backup server uses described backup server exit address to send;And make institute State reconstruction virtual unit the Backup Data received is mail to described storage device to back up.

If fault occurs, the virtual unit in the first VTL node 131 is carrying out backup tasks, this Individual backup tasks needs designated equipment address, VTL equipment entry address, backup server outlet upon execution Address.Owing to the device address of virtual unit after rebuilding, VTL equipment entry address, backup server go out Port address is all as before reconstruction, and the virtual unit after therefore rebuilding can continue executing with this backup traffic. It should be noted that before step S24 completes, this virtual unit cannot use, this time Backup server 11 can receive the unsuccessful prompting of backup tasks, and at this moment backup server 11 can be tasted again Examination performs backup tasks, until step S24 has performed, i.e. no longer can receive with successful execution backup tasks To the unsuccessful prompting of backup tasks.

In embodiment three, if certain one malfunctions i.e. in cluster, the related service that it runs can To be taken over by other nodes in cluster.For the backup traffic being currently running, backup server 11 still edge Sending Backup Data by the parameters of virtual unit before fault, these parameters include configuration information, equipment ground Location, entry address and backup server exit address;And rebuild the configuration information of virtual unit, equipment ground Location, entry address and backup server exit address are all identical with rebuilding front virtual unit, before therefore rebuilding The backup traffic that virtual unit is processing can be switched to virtual unit after reconstruction smoothly, and data will not occur After loss adapter completes, whole cluster is still within the state that can normally work.Thus may insure that whole Cluster cannot normally will not work because of individual node fault, and the availability of cluster product has been effectively ensured.And And, during failover, the backup/restoration task that malfunctioning node is currently running will not be failed, and fault connects The process of pipe is more smooth.After failover completes, it is not necessary to rescan mapping in Backup Server Virtual unit, then restart the data backup/recovery tasks of failure again to perform.No data can be realized The service protection lost.

Owing to embodiment three is performed for the device of embodiment one.Obviously, in embodiment one, two, three On the basis of, equally there is fault recovery device or Failure Recovery Module, perform in embodiment two Method, owing to having been carried out describing in embodiment two, the most no longer describes in detail.

Embodiment four

Seeing accompanying drawing 4, the present embodiment is a kind of first VTL node 4.Virtual magnetic tape library VTL cluster bag Include the first VTL node and the second VTL node, backup server and described first VTL node and institute State the second VTL node and set up communication connection, storage device and described first VTL node and described second VTL node sets up communication connection, and the first VTL node creates at least one virtual unit, a VTL The device address of the virtual unit on node is unique in VTL cluster, and described first VTL node includes CPU44, the first communication interface 41 communicated with CPU, the second communication interface 42 communicated with CPU, with And the internal memory 43 communicated with CPU, described first communication interface 41, for communicating with described backup server; Described second communication interface 42, is used for and described memory device communication;Described internal memory 43, is used for storing calculating Machine instructs;And described processor 44, and the coupling of described internal memory, for by running described computer instruction Perform the operation described in following example one and embodiment two.

Through the above description of the embodiments, those skilled in the art is it can be understood that arrive this Bright can add the mode of required common hardware by software and realize, naturally it is also possible to by hardware, but a lot In the case of the former is more preferably embodiment.Based on such understanding, technical scheme substantially or Person says that the part contributing prior art can embody with the form of software product, and this computer is soft Part product is stored in the storage medium that can read, such as the floppy disk of computer, and hard disk or CD etc., if including Dry instruction is with so that a computer equipment (can be personal computer, server, or the network equipment Deng) perform the method described in each embodiment of the present invention.

The above, the only detailed description of the invention of the present invention, but protection scope of the present invention is also Being not limited to this, any those familiar with the art is at the technology model that the invention discloses In enclosing, the change that can readily occur in or replacement, all should contain within protection scope of the present invention. Therefore, protection scope of the present invention should be as the criterion with described scope of the claims.

Claims (9)

1. a virtual unit method for reconstructing, it is characterised in that virtual magnetic tape library VTL cluster includes One VTL node and the second VTL node, backup server and described first VTL node and described the Two VTL node set up communication connection, storage device and described first VTL node and described 2nd VTL Node sets up communication connection, and the first VTL node creates has at least one virtual unit, in the first VTL node The device address of virtual unit unique in VTL cluster, the method includes:
Described second VTL node takes over the entry address of described first VTL node, described 2nd VTL joint Point obtains the configuration information of described virtual unit, device address, entry address and backup server export place Location;Wherein, before described second VTL node takes over the entry address of described first VTL node, further Including: described first VTL node searches the equipment not used by arbitrary VTL equipment in described VTL cluster Address;Described first VTL node uses virtual unit described in the device address information creating found;
Second VTL node uses described configuration information to generate and rebuilds virtual unit;
Distribute to described device address rebuild virtual unit, set up described reconstruction virtual unit and VTL equipment Mapping relations between entry address, backup server exit address.
2. the method for claim 1, it is characterised in that farther include after setting up mapping relations:
Described reconstruction virtual unit, by the entry address of described reconstruction virtual unit, receives described backup services Device uses the backup information that described backup server exit address sends;
Backup Data is mail to described storage device and backs up by described reconstruction virtual unit.
3. the method for claim 1, it is characterised in that described second VTL node adapter is described Before the entry address of the first VTL node, farther include:
Described second VTL node detects that the first VTL node breaks down.
4. a virtual unit reconstructing device, is positioned in the second VTL node, it is characterised in that virtual magnetic Tape library VTL cluster includes the first VTL node and described second VTL node, backup server and institute State the first VTL node and described second VTL node sets up communication connection, storage device and described first VTL node and described second VTL node set up communication connection, and the first VTL node creates at least one Individual virtual unit, the device address of the virtual unit in the first VTL node is unique in VTL cluster, described Virtual unit reconstructing device includes:
Creation module, for searching the device address not used by arbitrary VTL equipment in described VTL cluster, Described virtual unit is created for described second VTL equipment;
Acquisition module, for taking over the entry address of described first VTL node, and obtains described virtual set Standby configuration information, device address, entry address and backup server exit address;
Rebuild module, use described configuration information to generate for the second VTL node and rebuild virtual unit;
Enabling module, for described device address being distributed to reconstruction virtual unit, and setting up described reconstruction Mapping relations between virtual unit and VTL equipment entry address, backup server exit address.
5. device as claimed in claim 4, it is characterised in that described enable module is additionally operable to:
Make described virtual unit by the entry address of described reconstruction virtual unit, receive described backup server Use the backup information that described backup server exit address sends;And
Make described reconstruction virtual unit that the Backup Data received is mail to described storage device to back up.
6. device as claimed in claim 4, it is characterised in that described device farther includes:
Whether fault detection module, break down for detection the first VTL node.
7. a VTL node, it is characterised in that virtual magnetic tape library VTL cluster includes a VTL Node and the second VTL node, backup server and described first VTL node and described 2nd VTL Node sets up communication connection, and storage device is set up with described first VTL node and described second VTL node Communication connection, the first VTL node creates at least one virtual unit, and virtual in the first VTL node sets Standby device address is unique in VTL cluster, and described second VTL node includes CPU, communicates with CPU The first communication interface, the second communication interface communicated with CPU, and the internal memory communicated with CPU:
Described first communication interface, for communicating with described backup server;
Described second communication interface, is used for and described memory device communication;
Described internal memory, is used for storing computer instruction;And
Described CPU, and the coupling of described internal memory, for performing following behaviour by the described computer instruction of operation Make:
Take over the entry address of described first VTL node, obtain the configuration information of described virtual unit, equipment Address, entry address and backup server exit address;
Use described configuration information to generate and rebuild virtual unit;
Distribute to described device address rebuild virtual unit, set up described reconstruction virtual unit and VTL equipment Mapping relations between entry address, backup server exit address;
Described CPU is further used for performing:
Make described reconstruction virtual unit by the entry address of described reconstruction virtual unit, receive described backup clothes Business device uses the backup information that described backup server exit address sends;
Make described reconstruction virtual unit that Backup Data to mail to described storage device to back up.
Second VTL node the most as claimed in claim 7, it is characterised in that described CPU uses further In execution:
Before described adapter step, detect whether the first VTL node breaks down.
Second VTL node the most as claimed in claim 7, it is characterised in that described CPU uses further In execution:
Before described adapter step, whether the fault detecting the first VTL node is recovered.
CN201280005835.5A 2012-12-31 2012-12-31 A kind of virtual unit method for reconstructing and device CN103814352B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2012/088096 WO2014101211A1 (en) 2012-12-31 2012-12-31 Virtual device reestablishing method and device

Publications (2)

Publication Number Publication Date
CN103814352A CN103814352A (en) 2014-05-21
CN103814352B true CN103814352B (en) 2016-11-02

Family

ID=50709723

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201280005835.5A CN103814352B (en) 2012-12-31 2012-12-31 A kind of virtual unit method for reconstructing and device

Country Status (2)

Country Link
CN (1) CN103814352B (en)
WO (1) WO2014101211A1 (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106921508A (en) * 2015-12-25 2017-07-04 中兴通讯股份有限公司 Virtualization network element failure self-healing method and device
CN106557280B (en) * 2016-10-31 2019-10-15 北京华为数字技术有限公司 Date storage method, storage array and host server
CN108234410B (en) * 2016-12-15 2019-02-22 视联动力信息技术股份有限公司 A kind of virtual terminal distribution method and device

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101799780A (en) * 2010-03-10 2010-08-11 成都市华为赛门铁克科技有限公司 Configuration information file processing method, system, server, and virtual tape library equipment
CN201789513U (en) * 2010-09-14 2011-04-06 北京同有飞骥科技股份有限公司 High usable VTL (Virtual Tape Library) system capable of fast recovery
CN102521072A (en) * 2011-11-25 2012-06-27 成都市华为赛门铁克科技有限公司 Virtual tape library equipment and data recovery method

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4448719B2 (en) * 2004-03-19 2010-04-14 株式会社日立製作所 Storage system
US8843725B2 (en) * 2005-09-19 2014-09-23 Spectra Logic Corporation Virtual interchangeable storage device

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101799780A (en) * 2010-03-10 2010-08-11 成都市华为赛门铁克科技有限公司 Configuration information file processing method, system, server, and virtual tape library equipment
CN201789513U (en) * 2010-09-14 2011-04-06 北京同有飞骥科技股份有限公司 High usable VTL (Virtual Tape Library) system capable of fast recovery
CN102521072A (en) * 2011-11-25 2012-06-27 成都市华为赛门铁克科技有限公司 Virtual tape library equipment and data recovery method

Also Published As

Publication number Publication date
WO2014101211A1 (en) 2014-07-03
CN103814352A (en) 2014-05-21

Similar Documents

Publication Publication Date Title
US8533157B2 (en) Snapshot management apparatus and method, and storage system
US6912668B1 (en) Methods and systems for implementing shared disk array management functions
US7447933B2 (en) Fail-over storage system
US9135120B1 (en) Consistency group moving
US9575857B1 (en) Active/active replication
US8856319B1 (en) Event and state management in a scalable cloud computing environment
JP4634812B2 (en) A storage system having the ability to allocate virtual storage segments between multiple controllers
CA2498174C (en) Dynamic storage device pooling in a computer system
JP5026305B2 (en) Storage and server provisioning for visualization and geographically distributed data centers
US6538669B1 (en) Graphical user interface for configuration of a storage system
CN104081353B (en) Balancing dynamic load in scalable environment
US8572330B2 (en) Systems and methods for granular resource management in a storage network
JP2005275525A (en) Storage system
JP2009232207A (en) Network-switching apparatus, server system, and server migration method in the server system
CN100414542C (en) Method for managing memory source in cluster computer system
US8683172B2 (en) Method and apparatus for management between virtualized machines and virtualized storage systems
DE10014448B4 (en) Storage management system
US8392378B2 (en) Efficient backup and restore of virtual input/output server (VIOS) cluster
JP2004252686A (en) Information processing system
US10073745B2 (en) Automated disaster recovery system and method
US7870093B2 (en) Storage subsystem
EP1357465A2 (en) Storage system having virtualized resource
US8166241B2 (en) Method of improving efficiency of capacity of volume used for copy function and apparatus thereof
US8504741B2 (en) Systems and methods for performing multi-path storage operations
US8498967B1 (en) Two-node high availability cluster storage solution using an intelligent initiator to avoid split brain syndrome

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
C10 Entry into substantive examination
GR01 Patent grant
C14 Grant of patent or utility model