JP4464872B2 - Computer system and computer control method - Google Patents

Description

  The present invention relates to a computer system and a computer control method for performing remote copy by asynchronous data transfer between disk storage devices.

  2. Description of the Related Art In recent years, computer systems for handling large amounts of data that have been difficult in the past have been realized due to dramatic improvements in hardware processing performance and higher network speeds. As the amount of data handled increases, data maintenance becomes an important issue.

  In order to preserve data, it is necessary to safely protect important data without stopping operations, and to recover the system in a short time even if a disaster occurs. One technique for satisfying such requirements is an HA (High Availability) cluster configured using a plurality of servers and a large disk array device (DKC: Disk Controller).

  In the HA cluster configuration, data is duplicated at a remote data center in preparation for a disaster such as an earthquake. That is, data stored in a volume (primary volume) of a large disk array apparatus installed at a certain site is also stored in a volume (secondary volume) of a large disk array apparatus installed at a remote location.

For this reason, a method is generally used in which large disk array devices are connected to transfer write data from a host device between large disk array devices. Making a copy of a volume by transferring data to a remote location is called remote copy.
There are two types of methods for transferring data between large disk array devices: a synchronous transfer method and an asynchronous transfer method. The synchronous transfer method is a method in which a transfer source receives information indicating receipt of data every time data is transferred, and is suitable for data transfer at a relatively short distance. On the other hand, the asynchronous transfer method transfers data unilaterally without receiving a receipt response, and is suitable for data transfer over a long distance.

  For example, Patent Document 1 discloses a technique related to a general asynchronous remote copy, and Patent Document 2 discloses a technique related to a failure recovery method for maintaining temporal consistency of copy pairs in units of groups. ing.

  In general, when remote copy is used for backup purposes, it is possible to perform operations for each application by operating multiple groups of copy pairs created by the primary volume and secondary volume that maintain consistent state. Control by group. When performing control in units of groups, a batch operation function for a plurality of copy pairs is used as a hardware function. Remote copy in groups consists of group operation functions such as creating a copy pair, suspending the copy pair status (pause), restoring the suspend status of the copy pair status, and deleting the copy pair status.

Conventionally, when a recovery operation is performed when an error occurs in a part of a copy pair in a group, the recovery operation is performed while the volume in the group remains in the group. The copy pair in the state is repaired in copy pair units.
Japanese Patent Laid-Open No. 11-85408 (paragraphs 0013 to 0015, FIG. 1) JP 2002-189570 A (paragraphs 0012 to 0013, FIG. 1)

  However, in general, in group-based asynchronous remote copy, a normal copy pair and an abnormal copy pair belong to the same group on the DKC. It is necessary to perform repair in units of copy pairs so as not to affect the process.

  As a result, the number of copy status control commands issued to the DKC also increases, which adversely affects the performance of the entire system. In addition, since a normal state copy pair and an abnormal state copy pair coexist in one group, management becomes complicated. In addition, when an abnormality due to the same cause occurs in a plurality of groups, it is necessary to perform repairs one by one, which is inefficient.

  Therefore, the present invention has been made in view of the above problems, and when an abnormal copy pair occurs in a group, the affiliation of the group of the copy pair managed by DKC is changed from the original group. The objective is to perform efficient recovery of copy pairs by changing to an abnormal group and performing error recovery for each error group.

  In order to solve the above problems, a computer system according to the present invention includes a first storage device having a plurality of logical volumes for storing data, and a second storage device having a plurality of logical volumes connected to the first storage device. And managing copy pairs for each group in which a plurality of copy pairs each composed of the logical volume of the first storage device and the logical volume of the second storage device are grouped, and the copy state within the group is temporally consistent A processing unit that remotely copies the data of the logical volume of the first storage device to the logical volume of the second storage device between the copy pairs while maintaining the copyability, the processing unit comprising: In order to collect copy pairs in an abnormal state by extracting the copy pair from the group to which the copy pair belongs when the copy pair is in an abnormal state Go to abnormal group, an abnormality recovery copy pair in the abnormal group.

  According to the present invention, an abnormal copy pair can be efficiently recovered.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a diagram showing an example of the overall configuration of a computer system according to the present invention. The computer system includes a group automatic control function program 101, a primary DKC (first storage device) 112, a secondary DKC (second storage device) 113, and a terminal 121.

  The group automatic control function program 101 is a program that is operated by a processing unit such as a CPU (Central Processing Unit: not shown) on a host device connected to the primary DKC (disk control device) 112. In this embodiment, there is one group automatic control function program 101 for performing a copy operation, as shown in the figure, and this one program can also operate on a volume existing on the secondary DKC 113. However, it may be divided into two or more.

  The primary side DKC 112 and the secondary side DKC 113 (hereinafter referred to as “each DKC” when referring to both) are devices that control a disk array (not shown), and control I / O and various commands from the host device. Are applied to logical volumes such as volumes 1121 and 1131 (logically divided disk arrays).

Consistency groups (hereinafter referred to as “C / T group” or “group”) 114, 115, 116 are groups for maintaining data consistency among a plurality of copy pairs on hardware. A copy pair can belong to only one C / T group, and the C / T group to which the copy pair belongs is distinguished by a C / T ID.
The C / T group 114 is a normal C / T group with C / T ID “01”, and includes a copy pair 117 and a copy pair 118. The C / T group 116 is a normal C / T group with C / T ID “02”, and includes a copy pair 119 and a copy pair 120.

  The C / T group 115 is an abnormal C / T group with C / T ID “F1”, and when the copy pair 118 and the copy pair 119 are in the same abnormal state, they are composed of these two copy pairs. . In this case, the normal C / T group 114 is composed only of the copy pair 117, and the normal C / T group 116 is composed only of the copy pair 119. The copy pair 118 and the copy pair 119 are restored after the abnormal recovery. It belongs to the original C / T group.

The group automatic control function program 101 is configured to function as a pair status control command issuing unit 102, a status abnormality monitoring unit 103, a recovery function 105, and a control unit 110 by the CPU on the host device.
The pair status control command issuing unit 102 has an operation command function for issuing a designated command to each DKC.

The state abnormality monitoring unit 103 determines whether the copy pair is normal or abnormal from the copy pair information acquired from each DKC. Further, the state abnormality monitoring unit 103 includes a group control function 104 that changes the C / T group according to the copy pair state. The processing of the state abnormality monitoring unit 103 will be described later with reference to FIG.
The recovery function 105 is a function for executing recovery processing for an abnormal copy pair. The processing of the recovery function 105 will be described later with reference to FIG.

  The control unit 110 controls each function in the group automatic control function program 101. Further, a screen display function for creating a screen shown to the user by the terminal 121 such as a group status display screen 201 (see FIG. 2), a group details display screen 301 (see FIG. 3), an abnormal group details display screen 401 (see FIG. 4), etc. 111 is included.

  The pair status control command issuing unit 102, the status error monitoring unit 103, the recovery function 105, and the control unit 110 are respectively a group management table 106 that manages groups, a pair management table 107 that manages copy pairs, and an error recovery procedure (error recovery). Data is exchanged between the recovery procedure table 108 that manages the method) and the C / T ID management table that manages the C / T ID.

  The terminal 121 displays a screen created by the screen display function 111 and allows a user to input data. For example, the terminal 121 can be realized by an information device (including a display unit) such as a personal computer. .

  2 shows all groups managed by the group automatic control function program 101 created by the screen display function 111 based on the information set in the group management table 106, the pair management table 107, and the recovery procedure table 108 of FIG. A status display screen image 201 is displayed on the terminal 121 (see FIG. 1).

  In the normal group display unit 202, the group name 203 of all normal groups, the C / T ID 204, the number of copy pairs 205 in the normal state currently belonging to the normal group, originally belonging to the normal group, but currently in the abnormal state Therefore, the number 206 of copy pairs that belong to an abnormal group and are waiting for recovery, and the number 207 of copy pairs that originally belonged to the normal group but are currently recovering from an abnormality, belong to the abnormal group. it's shown. When the user presses the button 208 by operating the terminal 121 (see FIG. 1), a transition is made to a group detail display screen 301 (see FIG. 3) showing details of the target normal group.

  In the abnormal group display unit 216, the group name 209, C / T ID 210, abnormal status code 211 indicating the type of abnormal status, the number of copy pairs 212 currently belonging to the abnormal group, and the status within the group are displayed. The status group 213 to be shown and the recovery method 214 for the target abnormal group are displayed. When the user presses the button 215 by operating the terminal 121 (see FIG. 1), the screen transitions to an abnormal group details display screen 401 (see FIG. 4) showing details of the target abnormal group.

  3 is a detailed display screen image 301 of a specific group created by the screen display function 111 based on information set in the group management table 106, the pair management table 107, and the recovery procedure table 108 of FIG. 121 (see FIG. 1).

The group status display unit 302 displays the same group name, C / T ID, number of normal copy pairs, number of copy pairs waiting for recovery, and number of copy pairs being recovered, as the group name 203 of the normal group display unit 202. .
In the pair list display unit 303, the normal pair display unit 304 displays the details of the copy pair in the normal state, the DKC device number (primary SN) of the copy source (primary side), and the volume number of the primary target volume. (Primary VN), copy destination DKC device number (secondary SN), volume number of the secondary target volume (secondary VN), and copy pair status (such as “pause”).

  The abnormal pair display unit 305 displays details of the copy pair that is in an abnormal state. The DKC device number (primary SN) of the copy source (primary side), the volume number (primary VN) of the primary target volume, and the copy destination DKC device number (secondary SN), volume number of the secondary target volume (secondary VN), copy pair status (such as “Recovering (6A)”), C / T ID of the registration destination currently registered as an abnormal group (Current ID).

4 is a detailed display screen image 401 of a specific abnormality group created by the screen display function 111 based on the information set in the group management table 106, the pair management table 107, and the recovery procedure table 108 of FIG. It is displayed on the terminal 121 (see FIG. 1).
In the group status display unit 402, the same group name, C / T ID, error code (error status code: type of abnormal status) as the group name 209 of the abnormal group display unit 216 (see FIG. 2), and a copy belonging to the group The number of pairs is displayed.

The cause display unit 403 displays the cause of the abnormality corresponding to the error code of the target abnormality group.
The recovery method display unit 404 displays a recovery method for the target abnormality. Here, a recovery button 4041 and a reduced operation button 4042 are displayed. The recovery button 4041 operates the recovery function 105 to perform abnormal state recovery processing. The reduced operation button 4042 operates the recovery function 105, performs an abnormal state recovery process, and assigns a path used in a certain group to another group, thereby causing the entire computer system to perform a reduced operation. It is.

  In the pair detail display unit 405, the primary DKC device number (primary device), the volume number of the primary target volume, the secondary DKC device number (secondary device), the volume number of the secondary target volume, and the copy pair status (status) ), The C / T ID (original C / T ID) of the normal group that belonged before belonging to the abnormal group is displayed.

Next, the group management table 106 (see FIG. 1) that manages the status of each group will be described with reference to FIG. FIG. 5 is a diagram illustrating a data configuration of the group management table 106.
The group management table 106 includes an item number 501, a C / T ID 502 for each group, a group name 503 for each group, a group status 504 indicating the status of each group (normal / recovering / standby), and an error status code for an abnormal group 505, the number of copy pairs belonging to the C / T ID of each group 506, and a recovery procedure code 507 indicating a recovery procedure code that has been applied to the abnormal group before.

Next, the pair management table 107 (see FIG. 1) that manages the status of each copy pair will be described with reference to FIG. FIG. 6 is a diagram illustrating a data configuration of the pair management table 107.
The pair management table 107 includes an item number 601 indicating a copy pair number, a primary DKC device number (primary device number) 602, a primary volume number 603, a secondary DKC device number (secondary device number) 604, and a secondary volume number. 605, the copy pair status code 606, the current C / T ID 607 of the C / T group to which the copy pair belongs, and the original C / T ID 608 of the normal group to which the copy group belonged before changing to the abnormal group Has been. Among these, when the current C / T ID 607 and the original C / T ID 608 have the same value, it indicates that the target copy pair belongs to the normal group, and if different, it belongs to the abnormal group. That is, the current C / T ID 607 can be changed as appropriate, but the original C / T ID 608 indicates the C / T ID of the original normal group and is unchanged.

Next, the recovery procedure table 108 (see FIG. 1) managing the recovery procedure will be described with reference to FIG. FIG. 7 is a diagram illustrating a data configuration of the recovery procedure table 108.
The recovery procedure table 108 includes a recovery procedure code 701, a recovery method 702 for each recovery procedure code, an issuance command 703 issued according to the recovery procedure, a target status code 704 that is a target of the recovery procedure, and a priority indicating an execution priority for each recovery procedure. It is comprised from degree 705.

Next, the C / T ID management table 109 (see FIG. 1) that manages C / T ID will be described with reference to FIG. FIG. 8 is a diagram illustrating a data configuration of the C / T ID management table 109.
The C / T ID management table 109 includes an item number 801, a target C / T ID 802, a status 803 indicating the usage status of the C / T ID, and a use group indicating a group name of the group using the C / T ID. 804.

  Next, state abnormality monitoring processing by the state abnormality monitoring unit 103 will be described with reference to FIG. 9 (see FIGS. 1 to 8 as appropriate). FIG. 9 is a flowchart illustrating a state abnormality monitoring process performed by the state abnormality monitoring unit 103.

  First, when the state abnormality monitoring unit 103 starts the state abnormality monitoring process (step 901), the pair state control command issuing unit 102 is used for all copy pairs (all pairs) in the pair management table 107. A status acquisition command is issued, and the status codes 606 of all copy pairs are updated to the latest one (step 902). If an error occurs as a result of issuing the status acquisition command, the error code is set in the status code 606 of the pair management table 107.

  Subsequently, the status error monitoring unit 103 sets all copy pairs in the pair management table 107 as monitoring targets (step 903), and determines whether the copy pairs are normal or abnormal by performing the processing from step 904 on for all the copy pairs. Do. Specifically, first, it is determined whether the status code 606 of the copy pair is a normal code or an abnormal code (step 904). If it is abnormal ("No" in step 904). )), The condition whether the current C / T ID 607 and the original C / T ID 608 of the target copy pair are the same, and the error status code 505 of the group having the same C / T ID 502 as the current C / T ID 607 is the copy pair. A determination is made based on two conditions, which are different from the status code 606 (step 908).

  If both conditions are not satisfied (“No” in step 908), there is no change in the state, so there is no need to change the current C / T ID 607 of the copy pair, and the process proceeds to determination for the next copy pair (step 906). When either condition is satisfied (“Yes” in step 908), copy pair change processing is performed by the processing from step 909 onward.

That is, if the current C / T ID 607 and the original C / T ID 608 are the same (“Yes” in step 908), the current copy pair is abnormal (because “No” in step 904), but the current copy pair is abnormal. Since the C / T ID 607 is registered as a normal group, it is necessary to change the current C / T ID 607 to the C / T ID of the abnormal group by the processing from step 909 onward.
If the error status code 505 of the group having the same C / T ID 502 as the current C / T ID 607 is different from the status code 606 of the target copy pair (“Yes” in step 908), the status code 606 of the target copy pair is different. It has been changed to the abnormal state code, and it is necessary to change the current C / T ID 607 to the C / T ID of the abnormal group corresponding to the current abnormal state code by the processing after step 909.

  In step 909, as the copy pair change processing, the status error monitoring unit 103 first determines whether it is necessary to create a new error group, and the copy pair status code 606 in the group management table 106 is determined. A search is made for a group having the error status code 505 having the same value as, specifically, it is determined whether there is a group having the same code as the pair.

  If no matching group is found (“No” at step 909), the C / T ID management table 109 is searched for a C / T ID 802 whose status 803 is an abnormal group spare (step 910). After that, in step 911, an abnormal group having the found C / T ID 802 as C / T ID 502 is added to the group management table 106, and the abnormal group added to the use group 804 of the C / T ID management table 109 is added. Set the name.

  On the other hand, the state abnormality monitoring unit 103 finds a group having the C / T ID 502 as a change destination (“Yes” in Step 909) or creates a group having the C / T ID 502 (Step 911). After that, a pair control command for changing the C / T ID of the target copy pair (target pair) is issued to change the C / T ID of the target copy pair (step 912).

  Next, the number of copy pairs 506 of the group having the original C / T ID of the copy pair in the group management table 106 is reduced by one, and the number of copy pairs 506 of the group having the changed C / T ID is decreased by one. Increase (step 913). Then, the current C / T ID 607 of the pair management table 107 is changed to the changed C / T ID (step 914). After the change, the process proceeds to step 906 and proceeds to the process for the next copy pair.

  If the status code of the copy pair is normal (“Yes” in step 904), the status error monitoring unit 103 determines whether the current C / T ID 607 and the original C / T ID 608 of the copy pair are the same. Then, it is determined whether the target copy pair belongs to a normal group or an abnormal group (step 905). If it belongs to the normal group (“Yes” in step 905), there is no need to change the C / T ID, the process proceeds to step 906, and the process for the next copy pair is started.

If the target copy pair belongs to the abnormal group (“No” in step 905), the status error monitoring unit 103 has recovered the C / T of the target copy pair because the target copy pair has recovered to the normal status. A pair control command for changing the ID is issued, and the C / T ID of the target copy pair is changed, that is, returned to the original C / T ID (step 915).
Next, the number of copy pairs 506 of the group having the current C / T ID of the copy pair in the group management table 106 is decreased by 1, and the number of copy pairs 506 of the group having the changed C / T ID is increased by 1. (Step 916). Then, the current C / T ID 607 in the pair management table 107 is changed to the changed C / T ID (step 917). After the change, the process proceeds to step 906 and proceeds to the process for the next copy pair.

  When the status code determination is completed for all copy pairs (“Yes” in step 906), the status abnormality monitoring unit 103 ends the process. In FIG. 9, copy pairs having the same status code 606 in the abnormal state are registered in the same C / T group. However, copy pairs in the abnormal state having the same recovery procedure instead of the status code are registered in the same C / T group. There is also a way to register. Detailed processing will be described later with reference to FIG.

Next, the processing of the recovery function 105 will be described with reference to FIG. 10 (see FIGS. 1 to 8 as appropriate). FIG. 10 is a flowchart showing processing of the recovery function 105.
The recovery function 105 is a function for recovering an abnormal state group in units of groups, and there are two types of calling: a calling method specifying a recovery method and a calling method without specifying a recovery method. When the recovery method is designated, the user operates the terminal 121.

  When the recovery function 105 starts processing (step 1001), it first determines whether or not a recovery method is specified (step 1002). If the recovery method is specified ("Yes" in step 1002). ) Proceed to the recovery operation (step 1006), and if not designated (“No” in step 1002), proceed to the recovery method search process (step 1003).

  In step 1003, the recovery function 105 first determines whether or not a value is set in the applied recovery procedure code 507 of the group management table 106 of the target group as the recovery method search process. If the applied recovery procedure code 507 is not set (“No” in step 1003), the error status code 505 of the group is included in the target status code 704 from the recovery procedure table 108, and the highest priority 705 is set. A recovery procedure code 701 having a higher value is searched (step 1004).

  When the applicable recovery procedure code 507 is set (“Yes” in step 1003), the recovery function 105 includes the group error status code 505 in the target status code 704 from the recovery procedure table 108. Further, a recovery procedure code 701 having a priority 705 higher than the priority 705 of the applied recovery procedure code 507 is searched. After the search (step 1005), the process proceeds to the recovery operation (step 1006) as in the case where the recovery method is designated.

  That is, if a certain recovery target group cannot be recovered by the recovery method 702 corresponding to the recovery procedure code 701 searched in step 1004, once the processing of the flowchart of FIG. When the processing of the flowchart of FIG. 10 starts and proceeds from step 1003 to step 1005, recovery is attempted by the recovery method 702 corresponding to the recovery procedure code 701 having the next highest priority searched in step 1005. If recovery is still not possible, similarly, when proceeding to step 1005 again, recovery is attempted by the recovery method 702 corresponding to the recovery procedure code 701 having the next highest priority searched in step 1005, and That will be done until it recovers.

  Returning to the flowchart of FIG. 10, the recovery function 105 issues a pair control command using the pair status control command issuing unit 102 according to the issue command 703 of the recovery procedure code 701 from the recovery procedure table 108 as a recovery operation. (Step 1006) The executed recovery procedure 701 is set for the applied recovery procedure code 507 in the group management table 106 (Step 1007).

Subsequently, the recovery function 105 determines whether or not the pair control command executed previously has been completed normally (step 1008). If the recovery function 105 has completed normally (“Yes” in step 1008), the group management table 106 Recovery status is set for the group state 504 (step 1009).
If the pair control command does not end normally in step 1008 (“NO” in step 1008), the process of step 1009 is skipped and the process proceeds to step 1010.

Thereafter, the recovery function 105 calls the state abnormality monitoring unit 103 (step 1010), ends the processing by updating the state and changing the C / T ID.
In the process according to the flowchart of FIG. 10, every time one recovery procedure for one abnormality group is attempted, the state abnormality monitoring unit 103 updates each state (process according to the flowchart of FIG. 9) in step 1010. However, the frequency of the processing in step 1010 may be changed (reduced) as appropriate.

Next, a description will be given of the processing of the group management table 106 and the state abnormality monitoring unit 103 when registering a copy pair in the same C / T group that is not in the status code but has the same recovery procedure.
First, the group management table 106 when collecting copy pairs having the same recovery procedure will be described with reference to FIG. FIG. 11 is a diagram showing a data configuration of the group management table 106A in that case.

  The group management table 106A includes an item number 1101, a C / T ID 1102 of each group, a group name 1103 of each group, a group status 1104 indicating the status (normal / recovering / standby) of each group, a recovery procedure code 1105, each It consists of the number 1106 of copy pairs belonging to the C / T ID of the group and the recovery procedure code 1107 of the recovery procedure applied to the abnormal group. The change from the group management table 106 in FIG. 5 is that the error status code 505 is deleted and a recovery procedure code 1105 is added.

Next, with reference to FIG. 12, a description will be given of a state abnormality monitoring process when the state abnormality monitoring unit 103 collects copy pairs having the same recovery procedure. FIG. 12 is a flowchart showing a monitoring process for a state abnormality in that case.
First, the status abnormality monitoring unit 103 starts a recovery procedure search (step 1201), and issues a status acquisition command to all copy pairs in the pair management table 107 using the pair status control command issuing unit 102. The status codes 606 of all copy pairs are updated to the latest one (step 1202). If an error occurs as a result of issuing the status acquisition command, the error code is set in the status code 606 of the pair management table 107.

  Subsequently, the status error monitoring unit 103 sets all copy pairs in the pair management table 107 as monitoring targets (step 1203), and performs normal / normal determination of copy pairs by the processing from step 1204 on all the copy pairs. Do. Specifically, first, it is determined whether the status code 606 of the copy pair is a normal code or an abnormal code (step 1204). If it is abnormal (“NO” in step 1204). ), It is determined whether there is an application recovery procedure code 1107 for the group to which the target copy pair currently belongs (step 1205).

  If the applicable recovery procedure code 1107 is not set (“No” in step 1205), the state abnormality monitoring unit 103 includes the copy pair status code 606 in the target status code 704 from the recovery procedure table 108. The recovery procedure code 701 having the highest priority 705 is searched (step 1207). If the applied recovery procedure code 1107 is set (“Yes” in step 1205), the status code 606 of the copy pair is included in the target status code 704 from the recovery procedure table 108, and the applied recovery procedure code. A recovery procedure code 701 having a priority 705 higher than the priority 705 of 1107 is searched (step 1206).

  Next, the state abnormality monitoring unit 103 determines whether the current C / T ID 607 and the original C / T ID 608 of the target copy pair are the same, and a recovery procedure for the group having the same C / T ID 1102 as the current C / T ID 607. Judgment is made based on two conditions, whether 1106 is different from the retrieval result of the recovery procedure code (step 1208). If both conditions are not satisfied (“No” in step 1208), there is no change in the state, so there is no need to change the current C / T ID 607 of the copy pair, and the process proceeds to step 1219 to process the next copy pair. Move on. If either condition is satisfied (“Yes” in step 1208), copy pair change processing is performed (step 1209).

  In step 1209, the status abnormality monitoring unit 103 first determines whether it is necessary to create a new abnormality group as a copy pair change process. The group having the recovery procedure code 1105 having the same value is searched. If no matching group is found (“No” in step 1209), the C / T ID management table 109 is searched for a C / T ID 802 whose status 803 is an abnormal group spare (step 1210). After that, an abnormal group having the found C / T ID 802 as C / T ID 1102 is added to the group management table 106A, and the added abnormal group name is set in the use group 804 of the C / T ID management table 109. (Step 1211).

  On the other hand, the state abnormality monitoring unit 103 finds a group having the C / T ID 502 as a change destination (“Yes” in Step 1209) or creates a group having the C / T ID 502 (Step 1211). Thereafter, a pair control command for changing the C / T ID of the target copy pair is issued to change the C / T ID of the target copy pair (step 1212).

  Next, the number of copy pairs 1106 of the group having the original C / T ID of the copy pair in the group management table 106A is reduced by 1, and the number of copy pairs 1106 of the group having the changed C / T ID is increased by 1. (Step 1213). Then, the current C / T ID 607 in the pair management table 107 is changed to the changed C / T ID (step 1214). After the change, the process proceeds to step 1219, and the process for the next copy pair is started.

  If it is normal in step 1204 (“Yes”), the state abnormality monitoring unit 103 determines whether the target copy pair is the same depending on whether the current C / T ID 607 and the original C / T ID 608 of the copy pair are the same. It is determined whether it belongs to a normal group or an abnormal group (step 1215). If it belongs to the normal group (“Yes” in step 1215), it is not necessary to change the C / T ID, the process proceeds to step 1219, and the process for the next copy pair is started.

  If the target copy pair belongs to the abnormal group (“No” in step 1215), the status error monitoring unit 103 changes the C / T ID of the target copy pair because the target copy pair has recovered to the normal status. Is issued and changed (step 1216). Next, the copy pair number 1106 of the group having the current C / T ID of the copy pair in the group management table 106A is decreased by 1, and the copy pair number 1106 of the group having the changed C / T ID is increased by 1. (Step 1217). Then, the current C / T ID 607 of the pair management table 107 is changed to the changed C / T ID (step 1218). After the change, the process proceeds to step 1219, and the process for the next copy pair is started.

When the status code determination is completed for all copy pairs (“Yes” in step 1219), the status abnormality monitoring unit 103 ends the process.
By performing the process of FIG. 12, copy pairs having different causes of anomalies can belong to the same C / T group and can be simultaneously recovered.

Next, the internal configuration of the DKC having the function of changing the C / T group in asynchronous remote copy will be described with reference to FIG. FIG. 13 is a logical block diagram showing a configuration of a DKC having a group change function.
When the control device 1302 in the primary DKC 112 (see FIG. 1) receives a write request 1303 from the host 1301 to the primary volume 1305 of the copy pair (corresponding to 1121 in FIG. 1), it follows the DKC group management table 1304. Determine the write destination. Details of the in-DKC group management table 1304 will be described with reference to FIG.

  Referring to FIG. 14, the DKC group management table 1304 for managing the copy pair and group for each volume in the DKC will be described (see FIG. 13 as appropriate). FIG. 14 is a diagram showing the configuration of the DKC group management table 1304.

  The intra-DKC group management table 1304 includes an item number 1401, a volume number 1402 that is the number of each volume, a DKC device number 1403 that is paired with each volume, and a volume that is paired with each volume. It includes a number 1404, a C / T ID 1405 of each group, and a source C / T ID 1406 when the group is changed in the target volume.

  When the control device 1302 receives a data write request 1303 from the host 1301 to the primary volume 1305 of the copy pair at normal time, the control device 1302 writes to the cache memory 1307 simultaneously with the primary volume 1305 of the copy pair. The cache memory 1307 has an area set for each C / T number, and holds a write time, a write destination, and a value to be written for each write data 1308.

  In addition, in order to make the writing order consistent in the C / T group, the write data 1308 is arranged in time series in the cache memory 1307. When writing the write data 1308 from the cache memory 1307 to the secondary volume of the copy pair, the control device 1302 writes the write data 1308 to the secondary volume via the inter-DKC transfer cable 1306 in chronological order. , Keep the writing order consistent within the group.

  When the group is changed, the control device 1302 sets the change destination C / T ID in the C / T ID 1405 of the DKC group management table 1304, and also sets the original C / T ID 1406 in the C / T ID before change. Set. In addition, an area corresponding to the changed C / T ID is also secured in the cache memory 1307, and when the data write request 1303 is received after the change, writing is performed to the changed area in the cache memory.

  When the controller 1302 writes data from the cache memory 1307 to the secondary volume of the copy pair, the write data 1308 having the same target volume disappears in the cache memory 1307 corresponding to the original C / T ID. By confirming this, the consistency of the writing order of the original C / T group is prevented from being lost.

  As described above, according to the computer system of this embodiment, when an abnormal copy pair occurs in a group, the group belonging to the copy pair managed by DKC is changed from the original group to the abnormal group. At that time, copy pairs with the same type of error status or copy pairs with the same error recovery method are collected in the same error group, and error recovery is performed in units of the error group, thus enabling efficient error recovery of copy pairs. It can be carried out.

  This is the end of the description of the embodiments, but the aspects of the present invention are not limited to these. For example, specific configurations of hardware, programs, tables, and the like can be changed as appropriate without departing from the spirit of the present invention, such as division and integration.

It is a figure which shows the example of whole structure of the computer system of this embodiment. It is a screen image figure which displays the state of the whole group. It is a screen image figure which displays the detailed state of a specific group. It is a screen image figure showing the detailed state of a specific abnormal group. It is a figure which shows the structure of a group management table. It is a figure which shows the structure of a pair management table. It is a figure which shows the structure of a recovery procedure table. It is a figure which shows the structure of a C / T ID management table. It is a flowchart which shows the group change process performed using an error code as a key. It is a flowchart which shows an abnormal condition recovery process. It is a figure which shows the structure of a group management table. It is a flowchart which shows the group change process performed using a recovery procedure as a key. It is a logic block diagram which shows the structure of DKC with a group change function. It is a figure which shows the structure of the group management table in DKC.

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 Group automatic control function program 102 Pair status control command issuing part 103 State abnormality monitoring part 104 Group control function 105 Recovery function 106 Group management table 107 Pair management table 108 Recovery procedure table 109 C / T ID management table 110 Control part 111 Screen display Functional part 112 Primary side DKC (disk controller)
113 Secondary DKC (disk controller)
117 Normal copy pair 118 Abnormal copy pair 119 Abnormal copy pair 120 Normal copy pair 121 Terminal

Claims (19)

A first storage device having a plurality of logical volumes for storing data, a second storage device having a plurality of logical volumes connected to the first storage device, a logical volume of the first storage device, and the second storage device The copy pair is managed for each group in which a plurality of copy pairs configured with the logical volume are grouped, and the first storage device is copied between the copy pairs while maintaining the temporal consistency of the copy status within the group. A processing unit for remotely copying data of a logical volume to the logical volume of the second storage device,
The processing unit updates the normal or abnormal state of all the copy pairs, and the condition whether or not the move to the abnormal group has not yet been made for one of the abnormal copy pairs, and has already been in the abnormal group. If it has moved, determine the two conditions of whether the type of abnormal state has changed since moving to the abnormal group,
If any of the above conditions is met, it is determined whether there is an abnormal group corresponding to the type of abnormal state at that time. If the abnormal group does not exist, a new abnormal group is set, and the abnormal state is set. Move one of the copy pairs from the group to which it belongs to the abnormal group
In addition, update the number of copy pairs of the group management table that manages the number of copy pairs of the group and the abnormal group, update the information of the pair management table that manages the group and the abnormal group to which each copy pair belongs, Further, the computer system is characterized in that an abnormal recovery of a copy pair is performed in units of the abnormal group. A first storage device having a plurality of logical volumes for storing data, a second storage device having a plurality of logical volumes connected to the first storage device, a logical volume of the first storage device, and the second storage device The copy pair is managed for each group in which a plurality of copy pairs configured with the logical volume are grouped, and the first storage device is copied between the copy pairs while maintaining the temporal consistency of the copy status within the group. A processing unit for remotely copying data of a logical volume to the logical volume of the second storage device,
When the copy pair is in an abnormal state, the processing unit extracts the copy pair from the group to which the copy pair belongs, and moves to an abnormal group for collecting the abnormal copy pairs. A computer system characterized by performing abnormal recovery.   3. The computer system according to claim 2, wherein the processing unit performs an error recovery of the copy pair in units of the error group, and then returns the error-recovered copy pair to the group to which it belongs.   The processing unit, when extracting a copy pair in an abnormal state from the group to which it belongs and moving to an abnormal group, moves the copy pair having the same abnormal state type to the same abnormal group. Item 3. The computer system according to Item 2.   When the abnormal state type of the copy pair moved to the abnormal group changes, the processing unit further moves the copy pair to an abnormal group for collecting the copy pairs of the abnormal state type after the change. The computer system according to claim 4.   When the processing unit extracts a copy pair in an abnormal state from the group to which it belongs and moves to an abnormal group, the error recovery method for recovering the error moves the same copy pair to the same abnormal group. The computer system according to claim 2.   3. The computer according to claim 2, wherein when there are a plurality of groups, the processing unit extracts a copy pair in an abnormal state from the group belonging to the plurality of groups and moves to the abnormal group. system.   When performing error recovery of a copy pair in units of the error group, priority is given to the error recovery method for error recovery, and the processing unit has priority when there are a plurality of error recovery methods that can be used. The computer system according to claim 2, wherein the abnormality recovery is performed by using the abnormality recovery method having a high level.   The computer system according to claim 2, wherein the processing unit further displays the group information and the abnormal group information on a display unit.   The computer system according to claim 2, wherein the processing unit further displays one or more abnormality recovery methods for the abnormality recovery on a display unit. A first storage device having a plurality of logical volumes for storing data, a second storage device having a plurality of logical volumes connected to the first storage device, a logical volume of the first storage device, and the second storage device The copy pair is managed for each group in which a plurality of copy pairs configured with the logical volume are grouped, and the first storage device is copied between the copy pairs while maintaining the temporal consistency of the copy status within the group. A computer control method by a computer system comprising: a processing unit that remotely copies data of a logical volume to a logical volume of the second storage device;
When the copy pair is in an abnormal state, the processing unit extracts the copy pair from the group to which the copy pair belongs, and moves to an abnormal group for collecting the abnormal copy pairs. A computer control method characterized by performing abnormality recovery.   The computer control method according to claim 11, wherein the processing unit performs an error recovery of the copy pair in units of the error group, and then returns the error-recovered copy pair to the group to which it belongs.   The processing unit, when extracting a copy pair in an abnormal state from the group to which it belongs and moving to an abnormal group, moves the copy pair having the same abnormal state type to the same abnormal group. Item 12. The computer control method according to Item 11.   When the abnormal state type of the copy pair moved to the abnormal group changes, the processing unit further moves the copy pair to an abnormal group for collecting the copy pairs of the abnormal state type after the change. The computer control method according to claim 13.   When the processing unit extracts a copy pair in an abnormal state from the group to which it belongs and moves to an abnormal group, the error recovery method for recovering the error moves the same copy pair to the same abnormal group. The computer control method according to claim 11.   12. The computer according to claim 11, wherein when there are a plurality of groups, the processing unit extracts a copy pair in an abnormal state from the group to which the plurality of groups belong and moves to the abnormal group. Control method.   When performing error recovery of a copy pair in units of the error group, priority is given to the error recovery method for error recovery, and the processing unit has priority when there are a plurality of error recovery methods that can be used. The computer control method according to claim 11, wherein the abnormality recovery is performed by using the abnormality recovery method having a high value.   The computer processing method according to claim 11, wherein the processing unit further displays the group information and the abnormal group information on a display unit.   12. The computer control method according to claim 11, wherein the processing unit further displays one or more abnormality recovery methods for the abnormality recovery on a display unit.

Images