Restarting method using a snapshot

• US 8,024,601 B2
• Filed: 12/09/2009
• Issued: 09/20/2011
• Est. Priority Date: 01/04/2006
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Kambara, Yasuyuki, Masuda, Kouji, Hatasaki, Keisuke, Takamoto, Yoshifumi
• Primary Examiner(s): Beausoliel, Jr.; Robert W
• Assistant Examiner(s): Lottich; Joshua P
• Application Number:
  US12/633,962
• Publication Number:
  US 20100088543A1
• Time in Patent Office: 650 Days
• Field of Search: 714/13, 714/20, 714/4, 714/6, 714/4.1, 714/4.11, 714/4.12, 714/6.3
• US Class Current: 714/4.12
• CPC Class Codes: G06F 11/2025 using centralised failover control functionality
• CPC Class Codes: G06F 11/2028 eliminating a faulty processor or activating a spare
• CPC Class Codes: G06F 11/203 using migration
• CPC Class Codes: G06F 11/2046 where the redundant components share persistent storage G06F11/2043 takes precedence
• CPC Class Codes: G06F 11/2097 maintaining the standby controller/processing unit updated initialisation or re-synchronisation thereof G06F11/1658 and subgroups

Abstract

The active server A101 notifies its own operating status to the administration server 106, and the administration server 106 acquires a snapshot and copies a disk according to the policy established by a user. When the active server A101 or the active disk 113 fails, the administration server 106 chooses a backup server A104 and a status storage disk A114 to be booted and restarts them by using the snapshot according to the policy.

Claims

1. A server restarting method for a computing system comprising:

a plurality of active servers connected to a disk device comprising a plurality of storage areas;

at least one backup server connected to said disk device; and

an administration server to which each of said active servers and said backup server are connected, said administration server storing a snapshot containing memory content of said each active server in said storage areas,

the method comprising said administration server performing steps of;

receiving an alert from a first active server;

determining whether said alert corresponds to a predetermined condition;

issuing an instruction to said first active server to obtain an additional snapshot in addition to a previous snapshot obtained by said first active server when it is determined that said alert corresponds to said predetermined condition;

storing said additional snapshot in a storage area different from a storage area that stores said previous snapshot; and

storing information in a disk management storage area that relates together said additional snapshot, said storage area that stores said additional snapshot, and said predetermined condition,

wherein when a failure occurs in one of said active servers, designated a failed active server, then;

extracting a plurality of storage areas in which a snapshot corresponding to said failed active server is stored based on said information in said disk management storage area;

selecting a storage area corresponding to said predetermined condition from among said extracted storage areas using said information in said disk management storage area; and

starting up said backup server.

2. The method according to claim 1, wherein the timing for acquiring said snapshot includes timing of shutting down the OS of said failed active server after the OS of said backup server is booted.

3. The method according to claim 1, wherein, when a failure occurs in said failed active server, then the storage area connected therewith is connected with another backup server that is different from the backup server to be started and is then started.

4. The method according to claim 3, wherein, when a failure occurs in said failed active server, the failed active server acquires a memory dump, and then switches the storage area which acquired said memory dump over to a backup server.

5. The method according to claim 1, wherein each of said active servers includes policy storage means in which timing of a snapshot to be restarted when a failure occurs is stored;

and

wherein said policy storage means is referred to when said failed active server fails, and

a storage area in which timing of a snapshot associated with said failed active server and timing stored in said disk management storage area coincide with each other is chosen and connected to the backup server.

6. The method according to claim 5, wherein said policy storage means retains timing of acquiring said snapshot, determines whether or not an alert issued from the failed active server and said timing of acquiring the snapshot coincide with each other, and if the alert issued from the failed active server and said timing of acquiring the snapshot coincide with each other, acquires a snapshot for the failed active server.

7. The method according to claim 1, wherein said disk device includes a function to copy a storage area, and after acquiring said snapshot, uses said copying function to create a copy of the storage area for which said snapshot was acquired, and adds information of the copy of the storage area to said disk management storage area.

8. The method according to claim 1, wherein, said snapshot is acquired as a snapshot of a virtual server that operates on said failed active server.

9. The method according to claim 1, wherein said failed active server and said backup server are connected to a plurality of disk devices each of which is provided with a plurality of storage areas, and said disk device copies the storage area to a storage area of a different disk device.

10. The method according to claim 9, wherein said disk device copies a storage area in which a snapshot of said failed active server is stored to a storage area of a disk device to which a backup server is connected, and when said failed active server fails, said disk device connects the storage area that was connected to the failed active server, and starts up the storage area.

11. The method according to claim 9, wherein each of said active servers includes policy storage means in which timing of a snapshot to be restarted when a failure occurs is stored;

and

wherein said policy storage means is referred to when said failed active server fails, and

a storage area in which timing of a snapshot associated with said failed active server and timing stored in said disk management storage means coincide with each other is chosen, and

the chosen storage area is connected to a backup server.

12. The method according to claim 11, wherein said policy storage means retains timing of acquiring said snapshot, determines whether or not an alert issued from the failed active server and said timing of acquiring the snapshot coincide with each other, and if the alert issued from the failed active server and said timing of acquiring the snapshot coincide with each other, acquires a snapshot for the failed active server.

13. The method according to claim 9, wherein said snapshot is acquired as a snapshot of a virtual server that operates on said failed active server.

14. A computing system comprising:

a plurality of active servers connected to a disk device comprising a plurality of storage areas;

at least one backup server connected to the disk device; and

an administration server to which each of said active servers and said backup server are connected, said administration server storing a snapshot containing memory content of said each active server in said storage areas,

each active server configured to store a snapshot containing memory content of said each active server in said storage areas,

said administration server configured to;

receive an alert from a first active server;

determine whether said alert corresponds to a predetermined condition;

issue an instruction to said first active server to obtain an additional snapshot in addition to a previous snapshot obtained by said first active server when it is determined that said alert corresponds to said predetermined condition;

store said additional snapshot in a storage area different from a storage area that stores said previous snapshot; and

store information in a disk management storage area that relates together said additional snapshot, said storage area that stores said additional snapshot, and said predetermined condition,

wherein when a failure occurs in one of said active servers, designated a failed active server, then;

extract a plurality of storage areas in which a snapshot corresponding to said failed active server is stored based on said information in said disk management storage area;

select a storage area corresponding to said predetermined condition from among said extracted storage areas using said information in said disk management storage area; and

start up said backup server.

15. A computing system according to claim 14, wherein, when a failure occurs in said active server, the storage area which was connected to said failed active server is connected to another backup server that is different from the backup server to be started and is then started.

16. A computing system according to claim 14, wherein said disk device includes:

means for copying a storage area; and

means, after acquiring said snapshot, for using said copying means to create a copy of the copied storage area for which said snapshot was acquired, and adding information of the copied storage area to said disk management storage means.

17. A computing system according to claim 14, wherein said policy storage means retains timing of acquiring said snapshot, determines whether or not an alert issued from the active server and said timing of acquiring the snapshot coincide with each other, and if the alert issued from the active server and said timing of acquiring the snapshot coincide with each other, acquires a snapshot for the active server.

18. A computing system according to claim 14, wherein, an activate server includes means, when being in failure, for switching, after having acquired a dump, a storage area which acquired said dump over to a backup server.