CN100442253C - Hot mirror image base driven hot back-up method of server data - Google Patents
Hot mirror image base driven hot back-up method of server data Download PDFInfo
- Publication number
- CN100442253C CN100442253C CNB011041617A CN01104161A CN100442253C CN 100442253 C CN100442253 C CN 100442253C CN B011041617 A CNB011041617 A CN B011041617A CN 01104161 A CN01104161 A CN 01104161A CN 100442253 C CN100442253 C CN 100442253C
- Authority
- CN
- China
- Prior art keywords
- server
- data command
- command request
- master server
- backup
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Hardware Redundancy (AREA)
- Computer And Data Communications (AREA)
Abstract
The present invention relates to a hot mirror image base driven hot back-up method of server data. The present invention utilizes the characteristics that client ends in networks of Ethernet, etc. are used as occurrence sources of all data; a micro-detection module is operated on a master server and a backup server through loading a bottom layer backup drive on the client ends, which carries out intelligent mirror image operation to the master server and the backup server and detects whether the master server is in dead halt or not in order to determine whether the client ends need to carry out read-write operation again to the servers. The mode of sending the write data to the master server and the backup server can avoid the data loss caused at the moment of the dead halt of the master server in the existing hot backup mode.
Description
Technical field
What the present invention relates to is a kind of hot mirror image base driven hot back-up method of server data.
Background technology
Industry such as bank, securities broker company will guarantee that system can normally move all the time, make client-server can carry out read-write operation to server all the time, can not cause losing and mistake of data, generally all will be equipped with active and standby two servers, so the redundancy technique of server data has a wide range of applications.The server backup technology can be divided into Hot Spare and cold standby.Traditional cold standby is exactly the copy that carries out variety of way between master server and backup server, comprises some copies through data compression process.Can not play a good protection to data, particularly when master server crashes, insert backup server again after all working station need be restarted, this cold standby mode has no technology content and can say.Current main employing be the Hot Spare technology, mainly comprise dual mode: the one, all load backup module on the server stocking.When particular data file and thread internal memory change on finding master server, be about on the master server data image to backup server.When finding that master server crashes, all users switch on the backup server.The 2nd, stock server and share a disk array cabinet, active/standby server all can write this disk array cabinet, its principle is as long as the disk array cabinet does not meet accident, when Error from Master Server, network automatically switches to backup server work, its essence is multiserver and " shares " hard disk, but master server thread memory information requires constantly to pass to backup server.More than said several redundancy techniques all once be used in stock exchange, but mistake switching, loss of data can take place in existing Hot Spare technology, data are counter irritates, and too much take problem such as Internet resources.And these two kinds of backup modes are equally all kept away unavoidably at make mistakes moment workstation of server and are being write data and the loss of data and the data that cause are made mistakes with regard to principle, and its basic reason is as data the source to take place owing to ignored the client.
What at present, industry such as security, bank was taked mainly is top said two kinds of Hot Spare technology.Prior art overwhelming majority operation when data backup is to carry out on server, therefore will inevitably cause the system resource occupation rate high, causes network paralysis etc. easily.Mode below existing Hot Spare technology adopts realizes backup: at first, send instruction request and transfer data packets to master server by the client, send it back feedforward information after handling by master server again and give the client, and write hard disc data, write data simultaneously to backup server, and in the middle of practical application, when server crashes, major part occurs in the moment that just writes database, and client computer has received that feedback information shows request or transfer data packets success or failure.Most of application programs can not be differentiated master server and crash, cause in fact that data do not have enough time to write master server and backup server, so after switching to backup server, data are still lost.Existing Hot Spare technology also causes network congestion easily, and mistake is switched, the anti-filling of data etc.
Summary of the invention
The present invention will handle the server of focus from original redundancy technique and be transferred to all data generation sources---on the client, drive by load the bottom backup in client, and on master server and backup server, move micro-detecting module, and function not in the cards before reaching---data lossless loses.
The present invention drives by loading the bottom Hot Spare in client, and on master server and backup server the operation micro-detecting module, network structure is no matter be client-server, or file server mode, all data messages all are to be produced by client, and all command informations are in the inevitable preservation to some extent of client, so utilize the data source feature of client can realize really that data lossless loses.The job step of present technique is as follows: at first, when articulating master server and backup server on all working station, workstation will be preserved MAC or IP address and other parameters.Then mechanism is surveyed in such one of operation in system, in client when master server sends the data command request, if to the data read operation, then system can note described data command required parameter, and begin to open detector, and if in described data command request process, Error from Master Server or deadlock, then system automaticallyes switch, and all are connected on the backup server, lay equal stress on and newly send described data command request.Because described data command request is that data with existing on the server is operated, just do not write hard disk, and data had write backup server in the past already synchronously on the master server, so can not cause loss of data.If data are carried out write operation, then automatically technological means such as utilization address decoding is sending backup server simultaneously the described data command request of master server in system, and every corresponding parameters is also wanted corresponding correct.When backup server being carried out described data command request, client can abandon the information of server feedback, if Error from Master Server or deadlock, the microdetector module records Error from Master Server, then all connections is switched to backup server.Can not cause loss of data equally in this process, reason is no matter whether master server is made mistakes or crashed, and the described data command request msg of client is all with the concurrent backup server that writes in ablation process.Even so master server crashes, data can not lost fully yet.After master server crashes, be replaced into the chained list of backup server automatically by the chained list of all master servers during client is driven, reach all visits in the network and transfer visit to backup server to master server.Above-mentioned work all is to realize in the bottom mode, for client, has bottom layer independence, therefore can not cause any harmful effect or instability.
Advantage of the present invention is: the hot mirror image base driven hot back-up technology is based on data source---client, and write master server and backup server concomitantly, so the system resource occupation rate is extremely low.Loading backup in client drives, concurrent master server and the backup server of writing of all data that makes the client generation, even server crashes, but data still can write backup server, and carry out bottom operation such as chained list conversion and make client all switch to backup server to all operations of master server, lose thereby reach data lossless.Bottom layer driving Hot Spare technical resource occupation rate is extremely low, has overcome the traditional hot redundancy technique and has caused defectives such as network congestion, mistake are switched, the anti-filling of data easily, has solved the data degradation problem from principle, steadily switch in 15 seconds, and bottom layer realization, stable.
Description of drawings
Heat backup method process flow diagram among Fig. 1 the present invention.
Embodiment
Reach the technological means that predetermined purpose is taked for further setting forth the present invention,, be described in detail below in conjunction with the concrete treatment step of accompanying drawing to the hot mirror image base driven hot back-up method of a kind of server data of the present invention's proposition.
As shown in Figure 1, frame 1: workstation begins to network.
Frame 2: workstation articulates master server and backup server.
Frame 3: workstation is preserved network join dependency address, comprises NIC address, IP address, and correlation parameter.
Frame 4: workstation begins to send the data command request to master server and backup server.
Frame 5: judge that described data command request is read operation or write operation.
Frame 5.1: be judged as read operation, then only send described data command request to master server.And whether the unlatching detector crashes in the read operation process to judge master server.
Frame 5.1.1: judge whether master server crashes.
Frame 5.1.1.1: if crash, workstation carries out the chained list conversion, with all to the operation map of master server to backup server.
Frame 5.1.1.1.1: all workstations that loaded native system all switch and are connected to backup server.
Frame 5.1.1.1.2: described data command request is sent to backup server again.
Frame 5.1.1.1.3:, begin next described data command request to the described data command request success of backup server.
Frame 5.1.1.2: if do not crash, described data command request success, the described data command request of following secondary data.
Frame 5.2: be judged as write operation, then workstation carries out chained list conversion, bottom operations such as address decoding.
Frame 5.2.1: described data command Intra-request Concurrency ground is sent to backup server.
Frame 5.2.2: judge whether master server crashes.
Frame 5.2.2.1: if master server crashes, all workstations that loaded native system all switch and are connected to backup server.
Frame 5.2.2.1.1: send described data command request success to backup server, begin next described data command request.
Frame 5.2.2.2: the master server operate as normal, next described data command request is carried out in described data command request success.
A is a master server among the figure, and B is a backup server.
The computer software of writing according to this method, can adopt different language, write out different original programs, in the time will arriving the concrete webserver according to the software application that this method is write out, can do suitable adjustment in conjunction with concrete condition, all should be as use to this method.
Claims (1)
1, a kind of hot mirror image base driven hot back-up method of server data is characterized in that having taked the following step:
(1) workstation access network;
(2) workstation articulates master server and backup server, preserves MAC or IP address;
(3) workstation is preserved NIC address, IP address;
(4) workstation begins to send out the data command request to master server and backup server;
(5) judge that described data command request is read operation or write operation;
(6) when being judged as read operation, then only send out the data command request described, and open detector to master server, whether crash in the read operation process to judge master server;
(7) if master server crashes, workstation carries out the chained list conversion, with all to the operation map of master server to backup server, all workstations that loaded native system all switch and are connected to backup server, described data command request is sent to backup server again, after the backup server success of sending out the data command request described, begin next described data command request;
(8) if do not crash, next described data command request is carried out in described data command request success;
(9) when being judged as write operation, then workstation carries out chained list conversion and two bottom operations of address decoding, and described data command Intra-request Concurrency ground is sent to backup server;
(10) open detector and judge whether master server crashes;
(11) if master server crashes, all workstations that loaded native system all switch and are connected to backup server, after backup server sends out data command request success described, begin next described data command request; If master server is working properly, next described data command request is carried out in described data command request success.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB011041617A CN100442253C (en) | 2001-02-22 | 2001-02-22 | Hot mirror image base driven hot back-up method of server data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB011041617A CN100442253C (en) | 2001-02-22 | 2001-02-22 | Hot mirror image base driven hot back-up method of server data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1371058A CN1371058A (en) | 2002-09-25 |
| CN100442253C true CN100442253C (en) | 2008-12-10 |
Family
ID=4653719
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB011041617A Expired - Fee Related CN100442253C (en) | 2001-02-22 | 2001-02-22 | Hot mirror image base driven hot back-up method of server data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100442253C (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100424687C (en) * | 2004-06-18 | 2008-10-08 | 中国建设银行股份有限公司 | On-line processing system and method based on network |
| CN1329829C (en) * | 2004-11-29 | 2007-08-01 | 中兴通讯股份有限公司 | Back-up restoring method for user database |
| CN101876924B (en) * | 2009-04-30 | 2013-10-30 | 升东网络科技发展(上海)有限公司 | Database fault automatic detection and transfer method |
| CN102065138A (en) * | 2010-12-22 | 2011-05-18 | 上海锐起信息技术有限公司 | Method for realizing thermal backup control of servers in disk-less system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH08234822A (en) * | 1995-02-27 | 1996-09-13 | Fanuc Ltd | Method and device for controlling vibration of machine tool |
| JPH09282237A (en) * | 1996-04-19 | 1997-10-31 | Fujitsu General Ltd | Storage device using eeprom |
| US5699510A (en) * | 1994-12-15 | 1997-12-16 | Hewlett-Packard Company | Failure detection system for a mirrored memory dual controller disk storage system |
-
2001
- 2001-02-22 CN CNB011041617A patent/CN100442253C/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5699510A (en) * | 1994-12-15 | 1997-12-16 | Hewlett-Packard Company | Failure detection system for a mirrored memory dual controller disk storage system |
| JPH08234822A (en) * | 1995-02-27 | 1996-09-13 | Fanuc Ltd | Method and device for controlling vibration of machine tool |
| JPH09282237A (en) * | 1996-04-19 | 1997-10-31 | Fujitsu General Ltd | Storage device using eeprom |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1371058A (en) | 2002-09-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7606839B2 (en) | Systems and methods for providing client connection fail-over | |
| US7340637B2 (en) | Server duplexing method and duplexed server system | |
| US7512682B2 (en) | Database cluster systems and methods for maintaining client connections | |
| KR100471567B1 (en) | Transaction Management Method For Data Synchronous In Dual System Environment | |
| US7631021B2 (en) | Apparatus and method for data replication at an intermediate node | |
| US20040162836A1 (en) | System and method for altering database requests and database responses | |
| US7100070B2 (en) | Computer system capable of fast failover upon failure | |
| AU2011338485B2 (en) | Providing transparent failover in a file system | |
| CA2830530C (en) | System and method for failover | |
| US7188125B1 (en) | Replication using a special off-host network device | |
| US5828569A (en) | Method and apparatus for maintaining network connections across a voluntary process switchover | |
| CN101079896B (en) | A method for constructing multi-availability mechanism coexistence framework of concurrent storage system | |
| RU2746042C1 (en) | Method and the system for message transmission | |
| KR100450400B1 (en) | A High Avaliability Structure of MMDBMS for Diskless Environment and data synchronization control method thereof | |
| KR20040022433A (en) | Mirroring network data to establish virtual storage area network | |
| CN106790563A (en) | Distributed memory system and method | |
| CN100442253C (en) | Hot mirror image base driven hot back-up method of server data | |
| US20040216126A1 (en) | Method, system, and article of manufacture for agent processing | |
| KR100492167B1 (en) | Shared-nothing database cluster server system and On-line scaling method | |
| EP3816781A1 (en) | Storage area network controller | |
| CN112346912A (en) | Stateful service master and standby high availability system and method based on network file system | |
| JP2002202906A (en) | Database duplex system and method therefor | |
| CN111813334A (en) | Write performance optimization and double-control node combination method for Ceph | |
| CN111414411A (en) | High availability database system | |
| EP1564975B1 (en) | Service request handling method and storage system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C56 | Change in the name or address of the patentee |
Owner name: BEIJING VRV SOFTWARE CO., LTD. Free format text: FORMER NAME: BEIJING BEIXINYUAN AUTOMATION TECHNOLOGY CO., LTD. |
|
| CP01 | Change in the name or title of a patent holder |
Address after: 100083, Beijing North four ring road, 29 square German building, six floor Patentee after: Beijing VRV Software Co., Ltd. Address before: 100083, Beijing North four ring road, 29 square German building, six floor Patentee before: Beijing Beixinyuan Automation Technology Co., Ltd. |
|
| C56 | Change in the name or address of the patentee | ||
| CP02 | Change in the address of a patent holder |
Address after: 100081 Beijing, Zhongguancun, South Street, No. 34, Zhongguancun science and technology development building, room C, block, room 1602 Patentee after: Beijing VRV Software Co., Ltd. Address before: 100083, Beijing North four ring road, 29 square German building, six floor Patentee before: Beijing VRV Software Co., Ltd. |
|
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20081210 Termination date: 20190222 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |