HP StorageWorks Continuous Access XP Journal user guide Part number: T1778-96001 Fourth edition: June 2006
Legal and notice information © Copyright 2005, 2006 Hewlett-Packard Development Company, L.P. Confidential computer software. Valid license from HP required for possession, use or copying. Consistent with FAR 12.211 and 12.212, Commercial Computer Software, Computer Software Documentation, and Technical Data for Commercial Items are licensed to the U.S. Government under vendor’s standard commercial license. The information contained herein is subject to change without notice.
Contents About this guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Supported storage platforms and firmware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Intended audience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Prerequisites. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3 Continuous Access XP Journal operations requirements and restrictions . . . . . . . . . . . . . 39 System requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Requirements and Restrictions for Continuous Access XP Journal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . One-to-one volume copy operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Displaying information about logical paths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Displaying information about ports on local arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Configuring port attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Associating primary and remote arrays and configuring logical paths (Add DKC) . . . . . . . . . . . . . . . . . .
If Continuous Access XP Journal has already been started. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 Optional Operation pane components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 Changing array options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146 10Disaster recovery operations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Calling HP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 A Power management for arrays and network relay devices . . . . . . . . . . . . . . . . . . . . . . 187 When power is removed due to an unexpected reason . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . When power is removed from primary arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 8 Transferring business tasks from Continuous Access XP Journal secondary sites to primary sites . . . . . . 75 Copying access attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 DKC Operation pane, DKC option. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Tables 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 Document conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Emulation types for journal volumes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Metadata information. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
About this guide This guide provides information about: • Understanding Continuous Access XP Journal • Preparing for Continuous Access XP Journal operations • Configuring arrays and logical paths • Using journal groups • Performing Continuous Access XP Journal pair operations • Monitoring remote copy operations • Viewing history of data volume pair operations • Performing optional operations • Performing disaster recovery operations • Using Continuous Access XP Journal with NAS Blade • Troubleshooting • Man
• Hitachi TrueCopy for z/OS user guide You can find the above documents at http://www.hp.com/support/rwc/manuals • HP StorageWorks RAID Manager XP user guide You can find the above document at http://www.hp.com/support/manuals Document conventions and symbols Document conventions Table 1 Convention Element Blue text: Table 1 Cross-reference links and e-mail addresses Blue, underlined text: (http://www.hp.
For continuous quality improvement, calls may be recorded or monitored. Subscription service HP recommends that you register your product at the Subscriber’s Choice for Business web site: http://www.hp.com/go/e-updates. After registering, you will receive e-mail notification of product enhancements, new driver versions, firmware updates, and other product resources. HP web sites For additional information, see the following HP web sites: • http://www.hp.com • http://www.hp.com/go/storage • http://www.hp.
1 Overview of HP copy solutions Continuous Access XP Journal You can use HP StorageWorks Continuous Access XP Journal to create and maintain duplicate copies of all user data stored on a distant array for data duplication, backup, and disaster recovery purposes. This enables you to recover valuable data and protect it from a disaster, such as an earthquake, fire, or system failure.
This guide does not cover Business Copy XP operations. For information and instructions on performing Business Copy XP operations, see the HP StorageWorks Business Copy XP user guide for the XP12000/XP10000/SVS200 or contact your HP account support representative. Continuous Access XP and TrueCopy® for z/OS You can use the Continuous Access XP and TrueCopy for z/OS data duplication features to create copies of logical volumes in two different arrays at remote locations.
2 Overview of Continuous Access XP Journal Continuous Access XP Journal provides a RAID storage-based hardware solution for disaster recovery that enables fast and accurate system recovery. After establishing Continuous Access XP Journal operations, duplicate copies of data are automatically maintained for backup and disaster recovery purposes. During normal Continuous Access XP Journal operations, primary data volumes remain online to all hosts and continue processing read and write I/O operations.
Figure 1 provides an overview of Continuous Access XP Journal operations. Figure 1 Continuous Access XP Journal components for Fibre Channel connections Journal obtain This function stores existing primary volume data as a base-journal in the primary site journal volume. Then this function stores write data in the journal volume every time the host updates the primary data volume. The journal obtain operation is performed when a paircreate or pairresync operation command is issued from the primary site.
secondary site through dedicated Fibre Channel remote copy connections using a journal volume. You can operate Continuous Access XP Journal with the user-friendly GUI environment using Command View XP or XP Remote Web Console. NOTE: Host failover software is required for effective disaster recovery with Continuous Access XP Journal.
Figure 3 shows Continuous Access XP Journal’s connection configuration for multiple secondary arrays. By connecting one primary array with more than one secondary array, you can create a volume pair that has a one-to-one relationship for each journal group. Figure 3 Connection configuration for multiple secondary arrays Local arrays Continuous Access XP Journal operations involve local arrays at the primary and secondary sites. The primary array consists of the main control unit (primary array) and SVP.
XP Journal secondary data volumes). The secondary array issues read journal commands to the primary array in order to copy journals. For Command View XP, the secondary management station should be connected to the secondary arrays at the secondary site on a separate LAN. The secondary arrays should also be attached to a host system to: • Report sense information if a problem with a secondary data volume or secondary array occurs. • Provide disaster recovery capabilities.
Because journal data is stored in journal volumes, you can perform and manage highly reliable remote copy operations without suspension of remote copy operations. For example, even if a communications path between the primary and secondary arrays temporarily fails, remote copy operations continue after the communications path is recovered.
• Processing for changing a data volume pair’s status (for example, deleting or suspending a data volume pair) is not in progress NOTE: If a path from a host to a volume is defined, you cannot register the volume as a journal volume. You can use Command View XP management stations or XP Remote Web Console to register journal volumes.
Consistency group number specifications are different from journal group number specifications. Each array has the ability to manage a total of 128 consistency groups for the primary and secondary data volumes (No. 0 to 127 are assigned). NOTE: same.
See ”Remote copy connections” on page 51 for more information about installing and configuring FC remote copy connections. The Continuous Access XP Journal remote copy configuration between the primary and secondary arrays has the following requirements (see ”Remote copy connections” on page 51): • Continuous Access XP Journal supports a 1-to-1 remote copy connection in one journal group pair. In one journal group pair, one primary array can be connected to only one secondary array.
Host failover software Host failover software, which transfers information between host servers at the primary and remote sites, is a critical component of any disaster recovery effort. Host failover is configured using the desired host failover software product for the platform (for example, Microsoft Cluster Server or HP Serviceguard), depending on your installation requirements and standards. The Command View XP or XP Remote Web Console software products do not provide host failover functions.
suspended. When a data volume pair is suspended, a representation of host data writes to the primary and secondary data volumes is recorded in difference bitmaps. The primary array’s base-journal data is sent to the secondary array journal volume after a read command is issued from the secondary array. After that, base-journal data is restored from the journal volume to the secondary data volume. The initial copy operation finishes when all base-journals have been destaged to the secondary volume.
volume write option” on page 28). You can only enable the secondary data volume’s write option when splitting the pair from the primary array. NOTE: To reduce the overhead associated with these remote copy activities and maximize data transfers, the local array uses a special write command that is allowed only for Continuous Access XP Journal initial and update copy operations.
Table 3 Metadata information (continued) Type Description Journal sequence number Sequence number assigned when the journal is obtained Timestamp Time journal data is obtained The journal sequence number indicates the primary data volume write sequence that the primary array created for each journal group. Journal data is transferred to the secondary array asynchronously with regards to the host I/O.
secondary array selects journal data S1 to be settled, because S1 is the lowest sequence number. When S1 is removed from the sequence number queue, journal data S2 becomes the top entry, but it has not arrived yet. Journal data S2’s management information is 0. The secondary array waits for journal data S2. When journal data S2 arrives, the secondary array selects S2 as the next journal data to be settled. The journal data that the secondary disk selects is marked as “host-dirty” and treated as formal data.
• Split all pairs in a journal group. See ”Splitting mirrors (Pairsplit-r: suspending copy operations)” on page 111 for a description of the Suspend Range-Group pairsplit option. • Resume all suspended pairs in a journal group. See ”Restoring mirrors (Pairresync: resuming copy operations)” on page 112 for a description of the Resume Range-Group pairresync option. • Delete all pairs in a journal group.
• C is the master journal volume’s total capacity (in megabytes). • u is the master journal volume’s data usage rate (as a percentage). • V is the data transfer speed between the primary and secondary arrays. The unit is MB/s (megabytes per second). To find a journal volume’s usage rate, use the monitoring feature (see ”Monitoring remote copy operations” on page 135).
Table 4 Continuous Access XP Journal data volume pair status (continued) Pair status Description Data volume pair is not synchronized because the PSUE (pair suspended-error) primary or secondary array suspended the pair due to an error condition. For Continuous Access (see Table 7 on XP Journal pairs, the primary and secondary page 36 for suspend arrays keep track of any journal data discarded types) during the suspension operation.
Table 4 Continuous Access XP Journal data volume pair status (continued) Pair status Description Primary data volume access Secondary data volume access Suspending Pair is not synchronized and is in transition from PAIR or COPY to PSUS/PSUE. When the split/suspend pair operation is requested, the status of all affected pairs changes to Suspending. When the split/suspend operation is complete, the status changes to PSUS/PSUE.
Table 5 Continuous Access XP Journal split types (PSUS) (continued) Type Applies to Description PSUS, Pairsplit-S to RCU Primary data volume The primary array detected that the secondary data volume’s status changed to SMPL because you deleted the pair (pairsplit-S) from the secondary array. The pair cannot be resumed because the secondary data volume does not have the PSUS/PSUE status.
is suspended. Primary and secondary arrays also keep track of any journal data discarded during the pair suspension. When the suspended Continuous Access XP Journal pair is resumed (pairresync), the secondary array sends the secondary data volume’s bitmap to the primary array, and the primary array merges the primary and secondary data volumes’ bitmaps to determine which cylinders/tracks are out-of-sync.
• Primary data volume records updated by host-requested write I/Os after the pair was suspended Table 8 describes Continuous Access XP Journal suspension conditions and indicates which CU detects the condition and which pairs are suspended. See ”Displaying the History Operation pane” on page 141 for troubleshooting information for Continuous Access XP Journal suspension conditions.
Overview of Continuous Access XP Journal
3 Continuous Access XP Journal operations requirements and restrictions System requirements Continuous Access XP Journal operations involve primary and secondary arrays containing the primary and secondary volumes, remote copy connections between primary and secondary arrays, hosts at the primary and remote sites, and the licensed Continuous Access XP Journal.
Requirements and Restrictions for Continuous Access XP Journal One-to-one volume copy operations NOTE: Continuous Access XP Journal does not support operations in which one primary data volume is copied to more than one secondary data volume, or more than one primary data volume is copied to one secondary data volume. Because Continuous Access XP Journal operates on logical volumes rather than files, multivolume files require special attention.
NOTE: • Specify a port in SLPR 0 when using LUN Manager to define an LU path to a data volume. It is possible, but not recommended, that you specify a port in another SLPR when defining an LU path. For more information about SLPR 0, see the HP StorageWorks XP Disk/Cache Partition user guide. • Volumes containing a volume management area (VMA) cannot be used as journal volumes. For more information about VMA, see the HP StorageWorks LUN Security XP Extension user guide.
specifies. A limit on the number of pairs is applied to the primary and secondary arrays. Use the following formula to calculate the number of cylinders for different emulation types: • For OPEN-V Number of cylinders = (↑ ( (↑ (Max. LBA/512) ↑ ) /15) ↑ ) + 1 The ↑ …↑ symbols around a value indicate the value should be rounded up to the nearest integer. • Number of bitmap areas: Calculate the number of bitmap areas to be used by all data volumes forming pairs from the number of cylinders.
• Table 14 lists journal group specifications. Table 14 Journal group specifications Item Support specifications Number of journal groups Up to 256 journal groups (No. 0 - 255) per array if all the following conditions are satisfied: • • • • Array is an XP12000 (not an XP10000). Microprogram version is 50-05-00-00 or later. Shared memory options #3, #4, and #5 are already installed. High Performance mode is applied to shared memories.
NOTE: When Continuous Access XP Journal and Universal Replicator for z/OS coexist in the same local array, each journal group must contain either Continuous Access XP Journal pairs or Universal Replicator for z/OS pairs (not both).
Access XP Journal primary site. Also, when configuring a system in a 3DC cascading configuration, you must avoid using illegal combinations of an RCU target port in the Continuous Access XP Journal primary site and an RCU target port in the Continuous Access XP secondary site. If you use illegal port combinations, Continuous Access XP Journal remote copy operations are likely to be suspended due to failures.
Table 16 Illegal port combinations (for FRONT CL1) (continued) Location Illegal port combinations Location Illegal port combinations (channel adapter number) RCU target port in Continuous Access XP Journal primary site Target port in Continuous Access XP Journal primary site, or RCU target port in Continuous Access XP secondary site (channel adapter number) RCU target port in Continuous Access XP Journal primary site Target port in Continuous Access XP Journal primary site, or RCU target port in
Table 16 Illegal port combinations (for FRONT CL1) (continued) Location Illegal port combinations (channel adapter number) RCU target port in Continuous Access XP Journal primary site Target port in Continuous Access XP Journal primary site, or RCU target port in Continuous Access XP secondary site CLD-Q 1L (Add.
Table 17 Illegal port combinations (for REAR CL2) Location Illegal port combinations Location Illegal port combinations (channel adapter number) RCU target port in Continuous Access XP Journal primary site Target port in Continuous Access XP Journal primary site, or RCU target port in Continuous Access XP secondary site (channel adapter number) RCU Target port in Continuous Access XP Journal primary site Target port in Continuous Access XP Journal primary site, or RCU target port in Continuous Ac
Table 17 Illegal port combinations (for REAR CL2) (continued) Location Illegal port combinations (channel adapter number) RCU target port in Continuous Access XP Journal primary site Target port in Continuous Access XP Journal primary site, or RCU target port in Continuous Access XP secondary site CL2-F 2W (Add.4) 2X (Add.
Table 17 Illegal port combinations (for REAR CL2) (continued) Location Illegal port combinations Location Illegal port combinations (channel adapter number) RCU target port in Continuous Access XP Journal primary site Target port in Continuous Access XP Journal primary site, or RCU target port in Continuous Access XP secondary site (channel adapter number) RCU Target port in Continuous Access XP Journal primary site Target port in Continuous Access XP Journal primary site, or RCU target port in C
management for arrays and network relay devices” on page 187 for more information about powering off and on the primary and secondary arrays. 6. HP Representative: Install Continuous Access XP Journal remote copy connections between the primary and secondary arrays. The user supplies this hardware (optical fibre cables, switches, and so on). See ”Remote copy connections” on page 51 for remote copy configurations.
• Direct connections Figure 8 Direct remote copy connections • Switch connections Figure 9 Switch remote copy connections • Extender connections Figure 10 Extender remote copy connections CAUTION: When an MCU and RCU are connected via switches with a channel extender and multiple remote copy paths are assembled, the majority of data to be transmitted might be concentrated on particular switches, depending on the switch routing’s configuration and settings.
Planning for journal volumes The data transfer speed for journal groups is affected by the journal volume specifications the journal groups use. Therefore, to achieve the desired data transfer speed, you must consider these journal volume specifications. Considerations This section explains what you need to consider when deciding on journal volume specifications.
RAID group and journal group configurations A RAID group can consist of physical volumes with a different number of revolutions, physical volumes with different capacities, and physical volumes with different RAID configurations (for example, RAID-1 and RAID-5). A RAID group’s data transfer speed is affected by the physical volumes and RAID configurations. A journal volume’s data transfer speed depends on the data transfer speed of the RAID group to which the journal volume belongs.
Journal volume capacity > VH-M - VM-R x t where: • VH-M is the data transfer speed between the hosts and primary array • VM-R is the data transfer speed between the primary and secondary arrays • t is the length of time during which data transfer can continue To calculate the journal volume capacity required when a communications path failure occurs between the primary and secondary arrays, assign 0 (zero) to VM-R.
Journal transfer speed of DKCs A DKC’s journal transfer speed depends on the number of paths used for transferring journals. The number of paths for transferring journals must be large enough that the journal transfer speed exceeds the data transfer speed in the “normal status” mentioned previously. The maximum number of paths for each pair of the primary and secondary DKC is 8. NOTE: If the journal transfer speed falls short of the desired speed, check the channel processors’ usage rate.
Combining Continuous Access XP Journal with other data management functions Some volume types that non-Continuous Access XP Journal functions use can also be used as Continuous Access XP Journal data volumes and/or journal volumes. Table 18 explains whether non-Continuous Access XP Journal volumes can also be used as Continuous Access XP Journal volumes.
Table 18 Whether non-Continuous Access XP Journal volumes can be used as Continuous Access XP Journal volumes (continued) Functions and volumes Can volumes be used as primary data volumes? Can volumes be used as secondary data volumes? Can volumes be used as journal volumes? No. Yes. No. NOTE: If the source volume is used as a primary data volume, volume migration stops. NOTE: If the source volume is used as a secondary data volume, volume migration stops.
Table 18 Whether non-Continuous Access XP Journal volumes can be used as Continuous Access XP Journal volumes (continued) Functions and volumes Can volumes be used as primary data volumes? Can volumes be used as secondary data volumes? Can volumes be used as journal volumes? Volume that is disabled for use as an S-VOL Yes. No. No. Volume including VMA Yes. Yes. No. No. No. No. No. No. No. Yes. Yes. Yes. Yes. Yes. No. Yes. Yes. Yes. Volume to which a path is defined Yes. Yes.
Journal pairs. You can use LUN Manager to protect volumes assigned to Continuous Access XP Journal pairs. NOTE: A UNIX or PC server cannot access Continuous Access XP Journal secondary data volumes unless the pair is split. Combining Continuous Access XP Journal and Performance Control XP Initiator ports in the primary array do not support the HP StorageWorks Performance Control XP option.
Figure 14 through Figure 17 show various Continuous Access XP Journal/Business Copy XP configurations that share volumes. • Continuous Access XP Journal/Business Copy XP configurations that share the Continuous Access XP Journal primary data volume and Business Copy XP P-VOL Figure 14 shows an example of a Continuous Access XP Journal primary data volume that also functions as a Business Copy XP P-VOL.
Figure 16 combines the configurations shown in Figure 14 and Figure 15. Within a single Continuous Access XP Journal pair, the primary and secondary data volumes function as Business Copy XP P-VOLs, providing multiple copies at the primary and remote sites.
2. Execute the RAID Manager XP horctakeover command on the Continuous Access XP Journal pair to reverse the copy direction and re-establish the pair. Figure 19 Restoring Business Copy XP P-VOLs, step 2 3. Execute the RAID Manager XP horctakeover command on the Continuous Access XP Journal pair to change the copy direction to the original direction and re-establish the pair. Figure 20 Restoring Business Copy XP P-VOLs, step 3 4.
6. Execute the RAID Manager XP pairsplit -S command on the Business Copy XP pair to release the pair. Figure 23 Restoring Business Copy XP P-VOLs, step 6 7. Execute the RAID Manager XP paircreate command on the Business Copy XP pair to copy in the original direction Figure 24 Restoring Business Copy XP P-VOLs, step 7 8. Execute the RAID Manager XP pairsplit -r command on the Business Copy XP pair to put the pair in PSUS status. Figure 25 Restoring Business Copy XP P-VOLs, step 8 9.
operations are performed using cascading connections and a three data center (3DC) configuration. In a 3DC configuration, an intermediate site is located between the primary and secondary sites. Figure 27 shows an example of a 3DC cascading configuration in which Continuous Access XP Journal and Continuous Access XP Synchronous are combined. In this configuration, the volume in the primary site is a Continuous Access XP Synchronous primary volume (P-VOL).
secondary data volumes in the secondary site are made in the same order that updates to Continuous Access XP Synchronous P-VOL in the primary site are made. As explained previously, when the host in the primary site issues an I/O request to the Continuous Access XP Synchronous P-VOL, data is restored asynchronously to the Continuous Access XP Journal secondary data volume in the secondary site in a remote location.
2. Issue a request for creating a Continuous Access XP Synchronous pair to the primary array where Continuous Access XP Synchronous is installed. 3. Wait until the status of the Continuous Access XP Synchronous pair becomes PAIR. 4. Issue a request for creating a Continuous Access XP Journal pair to the Continuous Access XP Journal primary array. When creating a Continuous Access XP Journal pair, ensure that you select a value from 1 to 3 as the mirror ID. 5.
Figure 29 illustrates a 3DC multi-target configuration where Continuous Access XP Journal and Continuous Access XP Synchronous are combined.
Hardware configurations A 3DC multi-target configuration requires the following sites: • Primary site for Continuous Access XP Synchronous and Continuous Access XP Journal • Secondary site for Continuous Access XP Synchronous • Secondary site for Continuous Access XP Journal Each site requires one local array. For Command View XP, a management station is also required. HP recommends installing RAID Manager XP in the host computers at each site.
For more information about configuring Continuous Access XP Journal journal groups, see ”Journal group operations” on page 30. For more information about configuring ports, see ”Configuring port attributes” on page 85. Recovering from primary site failures If a disaster or failure occurs only at the primary site in a 3DC multi-target system, use the secondary volume at the Continuous Access XP Synchronous secondary site to enable the secondary site to take over your business tasks.
Continuous Access XP secondary sites to primary sites (in 3DC cascading configurations)” on page 72.
2. If a pair resynchronization has not been performed to reverse the Continuous Access XP Synchronous pair’s primary/secondary relationship (that is, if the pair is suspended), use RAID Manager XP to resynchronize the pair, so the primary/secondary relationship is reversed.
Transferring business tasks from Continuous Access XP secondary sites to primary sites (in 3DC multi-target configurations) If you remove failures from the primary site and other locations and the system changes to a 3DC multi-target configuration (see”Recovering from primary site failures” on page 70 and ”Changing to 3DC multi-target configurations after recovering from primary site failures” on page 71), you can transfer your business tasks back to the primary site.
2. Use the volume at the Continuous Access XP Journal secondary site to resume your business tasks.
6. Use RAID Manager XP to resynchronize the Continuous Access XP pair. The system configuration changes to the original 3DC multi-target configuration (see the lower-right portion of Figure 35). Figure 35 Transferring business tasks from Continuous Access XP Journal secondary sites to primary sites Combining Continuous Access XP Journal and LUN Security XP Extension You can create a Continuous Access XP Journal pair using volumes to which LUN Security XP Extension has assigned an access attribute.
Copying access attributes If you create a Continuous Access XP Journal pair using volumes to which the VMA is set with LUN Security XP Extension, the access attribute recorded in the VMA can be copied from of the primary data volume to the secondary data volume. The VMA, which is necessary for copying the access attribute, can be set to OPEN-V volumes only. For more information, see the HP StorageWorks LUN Security XP Extension user guide.
Figure 36 shows the process of copying access attributes when you create a pair using volumes with VMA. Figure 36 Copying access attributes In Figure 36, if both primary and secondary data volumes have VMA, the primary data volume’s VMA is copied to the secondary data volume. In Figure 36, the secondary data volume’s access attribute is Read/Write when the pair status is SMPL.
NOTE: If the pair is suspended or deleted during COPY status, the primary data volume’s access attribute is not copied to the secondary data volume. To copy the primary data volume’s access attribute to the secondary data volume, ensure that the pair status is PAIR, and then suspend or delete the pair. NOTE: When the primary data volume is in COPY or PAIR status, two access attributes can be changed and cannot be changed according to the existing access attribute.
NOTE: If the LUN Security XP Extension license capacity is insufficient in the array at the secondary site, you might not able to create Continuous Access XP Journal pairs. For more information, see the HP StorageWorks LUN Security XP Extension user guide.
Continuous Access XP Journal operations requirements and restrictions
4 Configuring arrays and logical paths To introduce Continuous Access XP Journal into your system and configure your system for remote copy operations, you must consider what data should be copied from which array to which array. The array containing data to be copied to another array is referred to as the primary array. The array to which data is copied from another array is referred to as the secondary array. The location of the primary array is referred to as the primary site.
3. Select the DKC Operation tab. The DKC Operation pane appears. If Continuous Access XP Journal has already been started 1. Select the DKC Operation tab. The DKC Operation pane appears. NOTE: DKC is an acronym for disk controller, which controls an entire array. Continuous Access XP Journal panes use the word DKC to indicate an array. DKC Operation pane components Figure 37 DKC Operation pane, DKC option • Switch Display: Changes information displayed in the DKC Operation pane.
NOTE: Information on the DKC Operation pane is updated when you do one of the following: • Select another tab, and reselect the DKC Operation tab. • Click Refresh ( ) (top right of Command View XP or XP Remote Web Console pane). • Click Apply. • Select Modify mode when you are in View mode. • Open or close the DKC Status pane. • Click Refresh on the DKC Status pane. Displaying information about remote arrays 1. Display the DKC Operation pane. 2. In the Switch Display box, select DKC.
Displaying information about logical paths Logical paths are communications paths that connect local array ports and remote array ports. 1. In the Switch Display box, select DKC. 2. Select a remote array from below Subsystem in the tree. The list displays logical paths. Each row in the list indicates one logical path. The list displays the following: • Path Gr. ID: Path group ID. The icon indicates the status of the path.
• Tree: Lists channel adapters and ports on the local array. Channel adapter (Fibre Channel interface) Target port RCU target port Initiator port External port Port Initiator/External MIX mode • List: Displays ports on the local array. • Port: Port number. • Attribute: Port attribute (initiator, target, RCU target, external, or initiator/external). • PCB Mode: Port mode. The mode is either Standard, High, or MIX. • CU#: Currently, displays nothing. • Error Code: Error code.
• The secondary array cannot read data from journal volumes in the primary array. CAUTION: Before changing a Fibre Channel port to an initiator port, disconnect the port from the host, delete all affected data volume pairs, delete all logical paths from the initiator port to the remote array, and remove all channel paths to the port.
To associate the primary and secondary arrays and configure logical paths between the two arrays, the primary and secondary array administrators must perform the following procedure: 1. Ensure that the remote copy connections and ports are properly configured. Also, obtain the remote array’s serial number. NOTE: To get the remote array’s serial number, ask the remote array’s administrator. The serial number is displayed in the Command View XP or XP Remote Web Console main pane when logging into the array.
NOTE: If an error occurs, the right-most column of the Preset list displays the error code. To view detailed information about the error, right-click the error code, and select Error Detail. An error message appears providing detailed information about the error. Figure 39 Add DKC pane • DKC S/N: Remote array’s five-digit serial number. • Controller ID: Remote array’s controller ID (that is, array family ID). NOTE: The controller ID for an XP12000/XP10000 is 4. • Path Gr. ID: Enter the path group ID.
Changing options for logical paths and arrays Use the DKC Option pane to set logical path and array options. These options apply to the local array. NOTE: DKC is an acronym for disk controller, which controls an entire array. Continuous Access XP Journal panes use the word DKC to indicate an array. To change logical path and array options, the primary and secondary array administrators can perform the following procedure: 1. Ensure that Command View XP or XP Remote Web Console is in Modify mode.
If you need high performance at the local array, set the minimum number of paths to two or more, depending on the number of pairs the local array manages. • RIO MIH Time: RIO MIH timer value, which is the wait time until the data transfer from the local array to the remote array is complete. The RIO MIH time value must be from 10 to 100 seconds. The default setting is 15 seconds. NOTE: RIO MIH is an acronym for remote I/O missing interrupt handler.
NOTE: If an error occurs, the right-most column of the Preset list displays the error code. To view detailed information about the error, right-click the error code, and select Error Detail. An error message appears providing detailed information about the error. Figure 41 Add Path pane • Port: Select an initiator port on the local array. NOTE: When specifying a port, use the keyboard to enter the port number. When entering the port number, you can abbreviate the port number to two characters.
4. Do one of the following: • In the tree, select a remote array. • In the list, right-click a remote array, and select Edit Path(s). The list displays information about logical paths. 5. In the list, select the logical paths you want to delete. 6. Right-click the list, and select Delete Path. A confirmation message appears. 7. Click OK to close the message. 8. In the DKC Operation pane, locate the upper-right list and the Preset list, and check the settings you made.
• If a logical path is selected in the list, the DKC status pane displays only the status of the selected logical path. 5. Check the status of the logical paths, and click Close to close the DKC status pane. Figure 43 DKC Status pane • List: • No.: Serial numbers used for rows in the list. • Path Status: Logical path status (see Table 25). • Port: Port on the local array. • Pair-Port: Port on the remote array. • DKC S/N: Remote array’s serial number. • Path Gr. ID: Path group ID.
Table 25 Logical path status (continued) Status Remarks Initialization Failed An error occurred with the initialization of the connection between the local and remote array. Probable causes are: • • • No cable is connected to the local array. No cable is connected to the remote array. No cable is connected to the network device that comes between the local and remote arrays.
5 Using journal groups To introduce Continuous Access XP Journal in your system and configure your system for remote copy operations, the primary and secondary site administrators must configure journal groups, ports, and logical paths. Also, these administrators must register journal volumes in journal groups. During Continuous Access XP Journal remote copy operations, data in primary volumes is copied via journal volumes to secondary volumes.
Journal Operation pane components Figure 44 Journal Operation pane (XP12000) Figure 45 Journal Operation pane (XP10000) • Tree: Lists journal groups in the local array you are logged into. NOTE: The tree displays journal groups used with Continuous Access XP Journal, and does not display journal groups used with Universal Replicator for z/OS. • Journal Groups: Located at the top of the tree. When selected, the upper-right list displays journal groups in the local array.
When you double-click this item, the tree displays journal groups in which journal volumes are registered. When you select a journal group in the tree, information about the journal group appears in the upper-right list. The journal group icons are: Journal group in initial status. Initial Journal volumes are registered in this journal group, but no data volumes (primary or secondary data volumes) are registered in this journal group. Master journal group.
• Status: Status of a journal group in the local array. Journal group in initial status. Initial Journal volumes are registered in this journal group, but no data volumes (primary or data secondary volumes) are registered in this journal group. When creating a Continuous Access XP Journal volume pair, data volumes are registered in a journal group. The status of the journal group changes to Active. Either of the following: Active • • Initial copy is in progress.
• Path Gr. ID: Path group IDs, which are used to identify path groups (that is, groups of logical paths). Up to eight logical paths can be registered in one path group. This column is blank if the journal group’s attribute is neither Master nor Restore. • CLPR: Journal group’s CLPR number. • CTG: Number of a consistency group to which the mirror belongs. This column is blank if the journal group does not have any volume pair.
2. Display the Journal Operation pane (”Displaying the Journal Operation pane” on page 95). 3. In the tree of the Journal Operation pane, do one of the following: • To register new volumes to a journal group in which journal groups are already registered, select the journal group from below Registered. CAUTION: You can register volumes only when the journal group’s attribute is Initial or the journal group’s status is Active, Stop, Hold, or Hold(Failure).
NOTE: If an error occurs, the right-most column of the Preset list displays the error code. To view detailed information about the error, right-click the error code, and select Error Detail. An error message appears providing detailed information about the error. Figure 46 Edit JNL Volumes pane • JNL Volumes: Information about journal volumes. • Parity Group: Parity group where a journal volume belongs. NOTE: If a parity group begins with E, the journal volume is an external LU.
NOTE: If a parity group name begins with E, the volume is an external LU. • CU:LDEV: Volume’s CU and LDEV numbers. The CU number appears to the left of the colon (:). The LDEV number appears to the right of the colon. NOTE: If a volume ends with #, the volume is an external LU. • Capacity(GB): Volume’s capacity in gigabytes. • Emulation: Volume’s emulation type. • CLPR: Number and name of the CLPR where the volume belongs.
CAUTION: In general, you can delete journal volumes only when the journal group’s attribute is Initial or the journal group’s status is Stop or Hold(Failure). Table 26 Requirement to delete journal volumes (when one journal group uses two mirror IDs) Status of journal group Can journal volumes be deleted? Mirror ID 1 Mirror ID 2 Active Hold No Active Hold(Failure) No Stop Hold Yes Stop Hold(Failure) Yes 5.
5. Click Close to close the JNL Group Detail pane. Figure 47 JNL Group Detail pane • JNL Group: Journal group’s number. • Attribute: Journal group’s attribute. • Initial: Journal group in initial status. Journal volumes are registered in this journal group, but no data volumes (primary or secondary data volumes) are registered in this journal group. • Master: Master journal group. Journal groups and primary data volumes are registered in this journal group. • Restore: Restore journal group.
• Data Overflow Watch (sec): Time (in seconds) for monitoring whether metadata and journal data are full. NOTE: Data Overflow Watch displays nothing when one of the following conditions is satisfied: • Inflow Control is No. • The journal group is a restore journal group. • Copy Pace: Pace for an initial copy activity for one volume: High, Medium, or Low. NOTE: • By default, the pace for an initial copy activity is Low. • Copy Pace displays nothing if the journal group is a restore journal group.
• Status: Status of a journal group (or a mirror) in the local array. Journal group in initial status. Initial Journal volumes are registered in this journal group, but no data volumes (primary or secondary data volumes) are registered in this journal group. When creating a Continuous Access XP Journal volume pair, data volumes are registered in a journal group. The status of the journal group changes to Active. Either of the following: Active • • Initial copy is in progress.
• Volume: Number of data volumes registered in the mirror. • Capacity: Total capacity of data volumes registered in the mirror. • Previous: Displays detailed information about the previous journal group. • Next: Displays detailed information about the next journal group. • Close: Closes the JNL Group Detail pane. Changing options for journal groups If you are logged onto an XP10000, you can perform this operation only on journal groups that have a number below 10.
NOTE: If an error occurs, the right-most column of the Preset list displays the error code. To view detailed information about the error, right-click the error code, and select Error Detail. An error message appears providing detailed information about the error. Figure 48 Change JNL Option pane • Meta/Data Ratio: Ratio of metadata area to journal data area. For example, if you select 16, the ratio of metadata area to journal data area is 1:16.
To specify Medium, ensure that the amount of update I/Os (that is, write requests from hosts to primary data volumes) is 10 MB/s or less per parity group. If it exceeds 10 MB/s, data volume pairs might become split (suspended). • High: Speed of initial copy activity is faster than Low and Medium. To specify High, ensure that update I/Os (that is, write requests from hosts to primary data volumes) do not occur. If update I/Os occur, data volume pairs might become split (suspended).
• Forward Path Watch Time: Specifies if the master journal group’s Path Watch Time value is forwarded to the restore journal group. If the Path Watch Time value is forwarded from the master journal group to the restore journal group, the two journal groups have the same Path Watch Time value. • Yes: Path Watch Time value is forwarded to the restore journal group. • No: Path Watch Time value is not forwarded to the restore journal group. No is the default.
• If necessary, repeat step 3 through step 5 to specify other journal groups. 6. Click Apply. NOTE: If an error occurs, the right-most column of the Preset list displays the error code. To view detailed information about the error, right-click the error code, and select Error Detail. An error message appears providing detailed information about the error. Splitting mirrors (Pairsplit-r: suspending copy operations) A mirror is a combination of a master journal group and a restore journal group.
• Enable: Hosts can issue read and write I/Os to the secondary data volume after you split the mirror. • Disable: Hosts cannot issue read and write I/Os to the secondary data volume after you split the mirror. This is the default. This parameter takes effect only when the selected volume is a master journal group. • Range: Split range. This is always set to Group and cannot be changed. • Suspend Mode: Specify how to handle update data that is not applied to the secondary data volume.
NOTE: The Range column indicates the unit used for restoring the mirror, and always displays Group. Deleting data volumes from mirrors (Pairsplit-S: ending copy operations) A mirror is a combination of a master journal group and a restore journal group. If you delete data volumes from a mirror, data copying from the master journal group to the restore journal group ends. To delete volumes from a mirror, the primary and secondary array administrators can perform the following procedure: 1.
Using journal groups
6 Performing Continuous Access XP Journal pair operations To perform remote copy operations with Continuous Access XP Journal, you must create pairs of a primary data volume and a secondary data volume. A primary data volume is a data volume to be copied. A secondary data volume is a data volume that is the copy destination for a primary data volume. To manipulate pairs of these data volumes, use the Pair Operation pane of Continuous Access XP Journal.
Pair Operation pane components Figure 50 Pair Operation pane • Tree: Lists ports in the local array ( ). Host groups or iSCSI targets ( ) appear below each port. If you select a port, host group, or iSCSI target, the list displays volumes for the port, host group, or iSCSI target. You can select only one port, host group, or iSCSI target at one time and cannot select two or more simultaneously.
• Port-G-ID-LUN: Logical volumes in the local array. The icons are: Volume is not paired with any other volume. Volume is not a primary or secondary data volume. Volume is a primary data volume. Volume is a secondary data volume. To the right of the icon appears information about an LU path, which is a path from hosts to the logical volume. Figure 51 Information in Port-G-ID-LUN column on the Pair Operation pane 1 Port that receives read and write requests from hosts.
This column is blank if the volume in the local array is not a primary nor secondary data volume. NOTE: The controller ID for an XP12000/XP10000 is 4. NOTE: This column can be blank while the pair is in transition to the SMPL status. To display the latest information in this column, refresh the screen. • Pair Port-G-ID-LUN: Remote array’s data volume. This column indicates a primary data volume if the remote array is a primary array.
In the following cases, this column is blank: • When the volume in the local is not a primary or secondary data volume. • CLPR: Number and name of the CLPR where the data volume belongs. • Used Volume: Size of used volumes and licensed capacity. For example, if 12.34 (15.00) (TB) is displayed, the licensed capacity for Continuous Access XP Journal is 15.0 terabytes, and 12.34 terabytes of volumes are used. • Total Pairs: Current total number of data volume pairs.
• If you change settings in the Display Filter pane, the changes remain in effect during the Continuous Access XP Journal session until you reset the settings. Figure 53 Display Filter pane • Port: Port to be displayed. To display all ports, select ALL. NOTE: You can specify the port number with two characters. For instance, you can abbreviate CL1-A to 1A. You can also enter the port number in both lowercase and uppercase characters. • G-ID: Group number of the host group or iSCSI target to display.
NOTE: If the number of pairs exceed approximately 10,000, the snapshot function is too large to save on a floppy disk. If this happens, save the snapshot file on a large-capacity disk. The following is an example of a snapshot file.
Creating a pair of data volumes (Paircreate) Using the Paircreate pane, the administrator of the primary array must create pairs of primary and secondary data volumes. When creating a pair, the administrator can specify options for initial copy activities. Before creating one or more pairs, the administrator must ensure that: • Volumes to be used as secondary data volumes are offline to all hosts. • Ports have already been configured (see ”Configuring port attributes” on page 85).
12.Click Set to close the Paircreate pane. 13.In the Preset list on the Pair Operation pane, check the settings you made. • To modify a setting, right-click the setting, and select Modify. • To cancel a setting, right-click the setting, and select Cancel. 14.Click Apply to start the pair creation and initial copy operations. NOTE: If an error occurs, the right-most column of the Preset list displays the error code.
NOTE: You can select only one secondary data volume even when two or more primary data volumes are selected in the Pair Operation pane. Continuous Access XP Journal automatically specifies the other secondary data volumes. For more information, see the explanation for the Select Other S-VOL(s) option on page 124. NOTE: If you selected more than one primary data volume, select the secondary data volume for the primary data volume being displayed.
NOTE: R-JNL does not display journal group numbers used by Universal Replicator for z/OS. • CT Group: Consistency group. NOTE: You cannot change his option after a pair is created. • DKC: Secondary array’s serial number and controller ID. You can also specify the path type (for example, channel type). NOTE: The secondary array must be the same for all pairs being created during one operation. • Initial Copy: Specifies whether the initial copy operation is started after the volume pair is created.
NOTE: You cannot select more than one pair. 3. Select Pairdisplay. 4. In the Pairdisplay pane (Figure 55), view detailed information about the pair. 5. Click Close. Figure 55 Pairdisplay pane • Status: Status of the pair. If the pair is split (or suspended), the column also displays the suspend type. If the pair is waiting for initial copy, the column also displays Queuing. • Alternative Path: Alternate path. • Prog./Sync.
• P-VOL: Primary data volume. The first line displays the primary data volume’s port number, G-ID, and LUN. The G-ID is the group number of a host group or iSCSI target. If the primary data volume exists in the local array, the first line also displays the CLPR number and name. The second line displays the device emulation type. The third line displays the volume capacity. NOTE: If a logical volume is an external LU, # appears after the LDEV number.
NOTE: • The Display Filter settings (see ”Filtering information in the list on the Pair Operation pane” on page 119) can affect how Previous or Next is recognized. • The list displays a maximum of 1,024 rows at once. The Previous and Next buttons on the Pairdisplay pane can only be used for the currently displayed 1,024 rows. • Refresh: Updates the pair status information. • Close: Closes the Pairdisplay pane.
• Group: If a failure occurs with a pair, all pairs in the consistency group where the pair belongs are split. • LU: If a failure occurs with a pair, only the pair is split. When the Change Pair Option pane appears, the pane displays the current option value. • Set: Applies settings to the Preset list in the Pair Operation pane (Figure 50). • Cancel: Discards the settings. Splitting a pair of data volumes After the initial copy operation finishes, use the Pairsplit-r pane to split the pair of data volumes.
• Enable: Hosts can write data to the secondary data volume while the pair is split. This option is available only when the selected volume is a primary data volume. • Range: Split range. If you select two or more pairs in the same consistency group, the default is LU. Otherwise, the default is Group. • LU: Only the specified pairs are split.
3. If any pair was suspended due to an error condition (use the Pairdisplay pane to view the suspend type), ensure that the error condition has been removed. The primary array will not resume the pairs until the error condition has been removed. 4. In the tree, select Subsystem or a ( ). 5. In the list, right-click the pair you want to restore. The result of this operation differs according to the selected mirror’s status. • When you select a pair in PSUS or PSUE status, the split pair recovers. 6.
NOTE: If Range is Group, you cannot change Priority. • DKC: Array. • Resync Mode: Processing after recovery of the pairs. • Normal: A split pair whose status is PSUS or PSUE is recovered. • Return to standby: The pair status is recovered from HLDE to HOLD. • Error Level: Range used for splitting a pair when a failure occurs. • Group: If a failure occurs with a pair, all pairs in the consistency group where the pair belongs are split. • LU: If a failure occurs with a pair, only the pair is split.
NOTE: If a pair is deleted, the deletion might take effect on the secondary data volume later than it takes effect on the primary data volume. Therefore, an error might occur if you reference a pair’s status or manipulate a pair immediately after RAID Manager XP deletes the pair. If such an error occurs, check the secondary data volume’s status and retry referencing the pair status or manipulating the pair. Figure 59 Pairsplit-S pane • Range: Delete range.
Creating a Point-in-Time copy of data volumes Use Continuous Access XP Journal to make Point-in-Time (PiT) duplicates of groups of volumes. The term Point-in-Time (PiT) refers to the process of copying data immediately when you want to obtain it. Use Continuous Access XP Journal to create a PiT copy of data volumes belonging to a journal group. 1. Stop all write I/Os from hosts to primary data volumes in the journal group.
7 Monitoring remote copy operations Use the Usage Monitor pane to monitor remote copy usage for the connected local array. The usage monitoring function collects I/O statistics for all volumes on the connected array. Displaying the Usage Monitor pane The Usage Monitor pane (Figure 60) displays user-selected remote copy I/O statistics and status of remote copy usage monitoring, allowing you to monitor remote operations.
NOTE: When monitoring is stopped, the usage monitor graph is closed. The usage monitor graph only appears when monitoring is running. • Gathering Interval: Data collection interval for usage monitoring between 1 and 15 minutes in one minute increment. The default is 1. NOTE: When monitoring is stopped, the default (1) appears in the Gathering Interval box. • Update: Displays the most recent data sample time on the graph.
maximum value of the statistical data displayed. If the value on the y-axis exceeds 10,000,000, the value is displayed in exponential notation (for example, 1E7 = 1x107 = 10,000,000; 2E8 = 2x108 = 200,000,000). 1. Ensure that usage monitoring is running. The usage monitor graph is only displayed when monitoring is on. 2. Right-click the graph area on the Usage Operations pane, and click Display Item. The Display Item pane (Figure 61) appears. 3.
• Stop the usage monitoring function by selecting Disable in the Monitoring Switch box, and clicking Apply. Figure 61 Display Item pane Table 28 Remote copy I/O statistics Statistic Host I/O Initial Copy Description Read Record Count Number of read I/Os per second. Read Hit Record Count Number of read hit records per second. Write Record Count Number of write I/Os per second. Write Hit Record Count Number of write hit records per second. Read Transfer Rate Amount of data read per second.
Table 28 Remote copy I/O statistics (continued) Statistic Asynchronous Copy M-JNL R-JNL Description M-JNL Asynchronous RIO count Number of asynchronous remote I/Os per second at the primary array. M-JNL Total Number of Journal Number of journals at the primary array. M-JNL Average Transfer Rate Average transfer rate for journals in the primary array. Unit is kilobytes per second. M-JNL Average RIO Response Remote I/O process time on the primary array.
Monitoring remote copy operations
8 Viewing history of data volume pair operations Use the History Operation pane (Figure 62) to view a history of operations for data volume pairs. For example, you can check the date and time data volume pairs were created or deleted. You can also save operation history in a text file. Displaying the History Operation pane Use the History Operation pane display history of operations for data volume pairs, such as the date and time when data volume pairs are created or deleted.
History Operation pane components Figure 62 History Operation pane • Status: Current status of operation history. • No history file exists.: Operation history does not exist. • Reading a history file failed.: An attempt to read operation history failed. • Updating ... n (%).: Operation history update is in progress, where n (%) indicates the progress. • Complete.: Operation history update is complete.
• Operation: Operation performed. • Pair definition: Data volume pair was defined. • Paircreate Start: Creation of data volume pair was started. • Paircreate Complete: Creation of data volume pair was finished. • Pairresync Start: Restoring data volume pair was started. • Pairresync Complete: Restoring data volume pair was finished. • Pairsplit-r Start: Splitting (Suspending) data volume pair was started. • Pairsplit-r Complete: Splitting (Suspending) data volume pair was finished.
• The History Operation pane does not display history information older than seven days. If the number of operations exceeds 65,535, the oldest operations are deleted in chronological order to keep the number at 65,535, even if the operations occurred within the last week. The history file always contains the most recent operations up to a maximum of 65,535 operations. The history information for an entire week might not always be available.
9 Performing optional operations Use the Optional Operations pane to perform the following procedures: • ”Displaying the Optional Operation pane” on page 145 • ”Changing array options” on page 146 Displaying the Optional Operation pane Use the Optional Operation pane to set options. If Continuous Access XP Journal has not been started 1. Use your Web browser to display the storage device list. In the storage device list, select the array to log into. Enter a user name and password, and click OK.
• Preset: The number to the left of the slash (/) indicates the number of items displayed in the Preset list. The number to the right of the slash indicates the maximum number of items that can be displayed in the Preset list. • Apply: Applies settings in the Preset list to the array. • Cancel: Cancels settings in the Preset list. NOTE: Information on the Optional Operation pane is updated when you do one of the following: • Select another tab, and reselect the Optional Operation tab.
• Max Initial Copy: Number of volumes that can be copied concurrently during an initial copy operation. Figure 65 System Option pane • Maximum Initial Copy Activities: Number volumes that can be copied concurrently during an initial copy operation. A value within the range of 1 to 128 can be specified. The default is 64 volumes. Maximum Initial Copy Activities can impact the performance of the primary array, depending on amount of I/O activity and number of pairs being registered at the same time.
Performing optional operations
10 Disaster recovery operations Preparing for disaster recovery The type of disaster and the status of the Continuous Access XP Journal volume pairs determines the best approach for disaster recovery. Unfortunately, some disasters are not “orderly” and involve intermittent or gradual failures occurring over a longer time period. You should anticipate and plan for all types of failures and disasters. The major steps in preparing for disaster recovery are: 1.
Transferring operations back to the primary site After the disaster recovery procedure finishes and your business applications are running at the secondary site, the next activity is to restore the primary site and make arrangements for copying data from the secondary site back to the primary site. The following procedures explains how to use RAID Manager XP to copy data from the secondary site to the primary site. 1.
5. If no error occurs in step 4, wait until suspension finishes. After suspension finishes, check whether there is a secondary data volume (which was originally a primary data volume) whose pair status is other than PSUS. If such a pair exists, remove the error cause and go back to step 1 after resuming your business task at the secondary site. 6.
Disaster recovery operations
11 Continuous Access XP Journal operations with NAS Blade This chapter describes how to use Continuous Access XP Journal with the TagmaStore USP NAS Blade. This chapter assume you: • Are familiar with Linux or UNIX computers. • Have read all of the previous chapters and fully understand the overview, requirements, and restrictions for Continuous Access XP Journal. • Have read the HP StorageWorks RAID Manager XP user guide and have a thorough understanding of RAID Manager XP operations.
Scope of Continuous Access XP Journal functions with NAS Blade Scope related to Continuous Access XP Journal • Volume type You can use only user LUs as Continuous Access XP Journal primary data volumes (P VOLs) and secondary data volumes (S VOLs) in a NAS Blade system. You cannot use NAS system LUs (NAS OS LU, Cluster Management LU, and dump LU) as Continuous Access XP Journal P-VOLs or S-VOLs.
NOTE: Create a shell script on the OS the system administrator uses. For more information about creating shell scripts, see the appropriate documentation for that OS. • Special file names given from standard input to the RAID Manager XP command When giving special file names from standard input to the raidscan, inqraid, or mkconf.sh commands, specify cat/home/nasroot/horc_devfile instead of ls/dev/sd*.
• HP StorageWorks Continuous Access XP Journal user guide (this document) • HP StorageWorks RAID Manager XP user guide • NAS Backup Restore User’s Guide Software prerequisites All of the following program products (PPs) must be installed in the NAS Blade connecting to the Continuous Access XP Journal P-VOL or S-VOL, and their licenses must be valid.
• Unmount and mount the source file system during the splitting of a Continuous Access XP Journal pair volume. • Connect the target file system to the NAS package. • Unmount and delete the target file system before the Continuous Access XP Journal pair is resynchronized.
To complete the RAID Manager XP configuration definition file: 1. Add the HORCM_DEV and HORCM_INST sections to a template for the RAID Manager XP configuration definition file using the mkconf.sh command in RAID Manager XP. 2. Edit the file and create the RAID Manager XP configuration definition file. You must perform these operations in each NAS package in CL1 and CL2 where the Continuous Access XP Journal P-VOL is connected and in each NAS package in CL1 and CL2 that uses the S VOL.
To execute the mkconf.sh command, the LU constituting the file system to be paired must be attached to the NAS OS. The LU configuration (size and number) in the file system to be paired must be exactly the same in the primary and secondary sites. To create the correct RAID Manager XP configuration definition file, HP recommends creating a file system to be temporarily paired before executing the mkconf.sh command. After creating the RAID Manager XP configuration definition file using the mkconf.
NOTE: You can specify the host name instead of the fixed IP address if the fixed IP address and the corresponding host name are registered in /etc/hosts, an NIS server, or a DNS server. See the NAS Blade Manager User’s Guide for information about registering the fixed IP address and the corresponding host name in /etc/hosts and how the NAS Blade system can search the host name using NIS or DNS. Based on Table 29, change each entry of ip_address in HORCM_MON to an appropriate value.
For example, in Figure 76, 11, 12, and 13 are LUs constituting the file system sample. Their LDEV numbers are 64, 12, and 66. $ sudo 11( 12( 13( horc_devlist | grep ':sample$' 17) 64 OPEN-V(3.906GB) -- -18) 12 OPEN-V(3.906GB) -- -19) 66 OPEN-V(3.
NOTE: You can specify the host name instead of the fixed IP address if the fixed IP address and the corresponding host name are registered in /etc/hosts, an NIS server, or a DNS server. See the NAS Blade Manager User’s Guide for information about registering the fixed IP address and the corresponding host name in /etc/hosts and how the NAS Blade system can search the host name using NIS or DNS. HORCM_MON #ip_address 123.456.78.
Compare this with Figure 80. For information about the horc_devlist command, see the NAS Backup Restore User’s Guide. $ sudo 11( 12( 13( horc_devlist | grep ':sample$' 17) 64 OPEN-V(3.906GB) -- -18) 12 OPEN-V(3.906GB) -- -19) 66 OPEN-V(3.
Figure 84 shows an example of a cascade configuration of Continuous Access XP Journal and Business Copy XP. Figure 84 Example of a cascade configuration of Continuous Access XP Journal and Business Copy XP By preparing the RAID Manager XP configuration definition file as shown in Figure 85, Figure 86, and Figure 87, you can operate the cascade configuration of Continuous Access XP Journal and Business Copy XP from RAID Manager XP.
the cascade configuration, you cannot operate the cascaded Business Copy XP using the GUI or CLI of the Snapshot feature. HORCM_MON #ip_address 123.456.78.51 service 20331 poll(10ms) 1000 HORCM_CMD #dev_name dev_name #UnitID 0 (Serial# 62486) /dev/sdf HORCM_DEV #dev_group # /dev/sdt VG_TC # /dev/sdu VG_TC # /dev/sdv VG_TC # /dev/sdw VG_TC HORCM_INST #dev_group VG_TC VG_TC VG_SI VG_SI timeout(10ms) 3000 dev_name SER = VG_032 SER = VG_033 SER = VG_034 SER = VG_035 dev_name ip_address 123.456.80.51 123.
HORCM_MON #ip_address 123.456.80.51 service 20331 poll(10ms) 1000 HORCM_CMD #dev_name dev_name #UnitID 0 (Serial# 62490) /dev/sdf HORCM_DEV #dev_group # /dev/sdt VG_TC # /dev/sdu VG_TC # /dev/sdv VG_TC # /dev/sdw VG_TC # /dev/sdt VG_SI # /dev/sdu VG_SI # /dev/sdv VG_SI # /dev/sdw VG_SI HORCM_INST #dev_group VG_TC VG_TC VG_SI VG_SI timeout(10ms) 3000 dev_name SER = VG_032 SER = VG_033 SER = VG_034 SER = VG_035 SER = VG_014 SER = VG_015 SER = VG_015 SER = VG_016 dev_name ip_address 123.456.78.51 123.
HORCM_MON #ip_address 123.456.80.51 service 20332 poll(10ms) 1000 HORCM_CMD #dev_name dev_name #UnitID 0 (Serial# 62490) /dev/sdf HORCM_DEV #dev_group # /dev/sdt VG_TC # /dev/sdu VG_TC # /dev/sdv VG_TC # /dev/sdw VG_TC # /dev/sdan VG_SI # /dev/sdao VG_SI # /dev/sdap VG_SI # /dev/sdaq VG_SI HORCM_INST #dev_group VG_TC VG_TC VG_SI VG_SI timeout(10ms) 3000 dev_name SER = VG_032 SER = VG_033 SER = VG_034 SER = VG_035 SER = VG_014 SER = VG_015 SER = VG_016 SER = VG_017 dev_name ip_address 123.456.78.
Enter the command shown in the following example to check the result of setting up the environment variable: $ sudo horc_printenv NAS Blade system allows you to set the following user environmental variables. • HORCMINST • HORCC_MRCF • HORCC_SPLIT • HORCC_RSYN • HORCC_REST Settings for disaster recovery For disaster recovery operations, perform the settings described in ”Disaster recovery operations” on page 149.
• ”Deleting Continuous Access XP Journal volume pairs” on page 172 Figure 88 Overview of remote copy operations Starting Continuous Access XP Journal operations and creating Continuous Access XP Journal pairs If the Continuous Access XP Journal S-VOL contains a file system, you must delete the file system using the enas_fsdelete command in NAS Blade Manager before starting Continuous Access XP Journal operations.
Splitting Continuous Access XP Journal volume pairs When splitting a Continuous Access XP Journal volume pair, you can choose either offline or online backup. With offline backup, you can split the pair by unmounting the P-VOL. With online backup, you can split the pair by temporarily saving updates to a file system without unmounting the P-VOL. During offline backup, a Continuous Access XP Journal volume pair is split after completely stopping access from clients by deleting CIFS/NFS shares.
NOTE: When the S-VOL is the file system managed by NAS/Sync Image, mount the differential-data snapshots using the syncmount command, and create the CIFS/NFS shares using the enas_nfscreate and enas_cifscreate commands, if necessary. 11.At the secondary site, start the program that accesses the S-VOL. Online backup To split the Continuous Access XP Journal volume pair with online backup: 1.
To resynchronize the Continuous Access XP Journal volume pair: 1. At the secondary site, reserve the device file to be used as a target file system • Non-LVM: ?sudo horc_svol_define -d Device-file-number • LVM: ?sudo horc_svol_vmdefine -d Device-file-number, ... 2. At the primary site, resume the Continuous Access XP Journal volume pair. sudo pairresync {-g group name|-d volume name} 3. At the primary site, verify that the Continuous Access XP Journal volume pair has been resumed.
NOTE: When the S-VOL is the file system managed by NAS Sync Image, delete the CIFS/NFS shares under the differential-data snapshots using the enas_nfsdelete and enas_cifsdelete commands, unmount the differential-data snapshots using the syncumount command, and release the differential-data storage devices using the syncstop command. 3. At the secondary site, delete the S-VOL file system using the enas_fsdelete command. 4. At the primary site, delete the Continuous Access XP Journal volume pair.
• Non-LVM: sudo horc_svol_import -f Target-file-system-name -d Device-file-number • LVM: sudo horc_svol_vmimport -f Target-file-system-name -d Device-file-number, ... 3. Mount the S-VOL using the enas_fsmount command, and create CIFS/NFS shares using the enas_nfscreate and enas_cifscreate commands. Recover the file system when mounting the S-VOL. 4. Start the program that accesses the S-VOL.
10.At the secondary site, prevent NAS Sync Image from performing operations on the old S VOL. sudo horc_pvol_freeze -f Source file system name 11.At the secondary site, split the Continuous Access XP Journal volume pair. sudo pairsplit {-g group name|-d volume name} -rw 12.At the secondary site, enable operations from NAS Sync Image on the old S-VOL. sudo horc_pvol_unfreeze -f Source file system name 13.At the primary site, connect the target file system to the NAS package.
Manager XP, or a command name in case of the RAID Manager XP command); and TIME shows the core file’s creation time. Table 32 Format of RAID Manager XP log files in the NAS Blade system Log file classification Log file name Log file format RAID Manager XP logs under operation RAID Manager XP starting log /home/nasroot/log*/curlog/horcm_HOST.log Command log /home/nasroot/log*/horcc_HOST.log RAID Manager XP error log /home/nasroot/log*/curlog/horcmlog_HOST/horcm.
Notes on operations of RAID Manager XP log files Since RAID Manager XP log files are output to NAS OS LU, if Continuous Access XP Journal operations still show an error, log files become large and available space on NAS OS LU decreases. You should periodically check the capacity of log files (except trace files) to which RAID Manager XP restricts capacity, stop RAID Manager XP, and delete log files using the horc_logremove command before exceeding 1 MB.
If you delete the Continuous Access XP Journal volume pair when the S-VOL is not connected to the NAS package, release the device file used in the target file system using the horc_svol_delete (non-LVM) or horc_svol_vmdelete (LVM) command.
12 Troubleshooting General troubleshooting You are responsible for the operation and normal maintenance of the Command View XP management station. For troubleshooting information on Command View XP, see the Troubleshooting chapter of the HP StorageWorks Command View XP user guide for XP Disk Arrays. The following are guidelines for troubleshooting the Command View XP management station: • Check the cabling and the LAN. Verify that both the computer and LAN cabling are firmly attached.
Table 33 General troubleshooting information for Continuous Access XP Journal (continued) Error Corrective action A Continuous Access XP Journal error message is displayed on the Command View XP management station. Remove the error cause, and retry the Continuous Access XP Journal operation. The status of a logical path is not normal. Check the path status in the DKC Status pane, and see Table 34. The pair status is suspended. Check the pair status in the Pairdisplay pane, and see Table 35.
Table 34 Troubleshooting problems with status of logical paths (continued) Pair status Description Corrective action Resource Shortage (Port) Array rejected the request for establishing a logical path because all resources in the local and remote arrays are being used for other connections. Delete the failed path and all paths and remote arrays not currently in use. The local array can be connected to up to 64 remote arrays with up to eight paths to each remote array.
Table 34 Troubleshooting problems with status of logical paths (continued) Pair status Description Corrective action Pair-Port Number Mismatch Specified port in the remote array is physically disconnected from the local array. Ensure that you specified the correct remote array port. Correct the port number if necessary. Ensure that the local and remote arrays are connected by cables to each other. If the arrays are not by cables, and retry the operation.
Table 35 Troubleshooting suspended Continuous Access XP Journal pairs Suspend type Applies to Description Corrective action PSUE, by RCU Primary data volume Primary array suspended a pair because the primary array detected an error condition at the secondary array. The suspend type for the secondary data volume is by MCU. Clear the error condition at the secondary array or secondary data volume. To access the secondary data volume, delete the pair from the primary array.
Table 36 provides troubleshooting instructions for Continuous Access XP Journal pair suspension. Hardware failures affecting cache memory and shared memory of the primary or secondary array might also cause Continuous Access XP Journal pairs to be suspended. Table 36 Resolving Continuous Access XP Journal pair suspension Classification Causes of suspension SIM Recovery procedure Primary array hardware or secondary array hardware Hardware redundancy was lost due to some blockade condition.
Calling HP If you need to call you HP account support representative, provide as much information about the problem as possible, including: • Command View XP or XP Remote Web Console configuration information saved on disk using the FD Dump Tool • Circumstances surrounding the error or failure • Exact content of any error messages displayed on the host systems • Continuous Access XP Journal (or other) error codes displayed by Command View XP or XP Remote Web Console • Reference codes and severity levels of
Troubleshooting
A Power management for arrays and network relay devices This appendix explains power management for arrays and network relay devices during remote copy operations.
1. Power on the primary array. 2. If the primary array is ready to resume remote copy operations, restore data volume pairs that were split, and confirm that the status of the data volume pairs is COPY or PAIR. This operation must be performed at the primary site. After taking these steps, remote copy operations resume. The primary array can respond to read journal commands issued by the secondary array.
When powering off network relay devices To power off a network relay device during remote copy operations, first ensure that all data volume pairs or mirrors (that is, pairs of journal groups) are split and the status of all data volume pairs is PSUS. If data volume pairs are split and the network relay device is powered off as mentioned above, use the following procedure to resume remote copy operations: 1. Power on the network relay device. 2.
Power management for arrays and network relay devices
B Pinned track recovery for Continuous Access XP Journal volumes If a hardware failure occurs when the array is in operation, some data in the array’s cache memory might not be written to disk drives and might remain in cache memory. Data remaining in cache memory due to such a reason is referred to as a “pinned track. If a pinned track occurs on a Continuous Access XP Journal primary or secondary data volume, the primary array suspends the data volume pair (SIM reference code = D41x, D42x, DB1x, DB2x).
Pinned track recovery for Continuous Access XP Journal volumes
C SIM reporting The local array generates service information messages (SIMs) to notify you of possible array service requirements. SIMs can be generated by the array’s channel and storage path microprocessors and service processor (SVP). The SVP reports all SIMs related to Continuous Access XP Journal processing. Each time a SIM is generated, the amber Message LED on the array’s control panel (under the Ready and Alarm LEDs) turns on as an additional alert.
Table 37 Typical SIM showing reference code and SIM type (continued) Reference code Byte 22 Byte 23 DC 1X Severity Description SIM generated by SVP log file Serious Volume being used by the primary data volume is suspended. Primary array SIM Log Primary array SIM Log Primary array SIM Log Primary array SIM Log Primary array SIM Log Secondary array SIM Log Failure detected in the primary array. DC 2X Serious Volume being used by the primary data volume is suspended.
Index Numerics 3DC configuration cascading 65 multi-target 67 A Add DKC pane 86 applications, combining with 57 audience, documentation 11 B bandwidth 55 base-journals 26, 28 bitmap areas, number of 41 bitmaps, difference 26, 28 Business Copy XP 15, 60 C cable length, maximum 24, 39, 51 cache pinned track recovery 191 requirements 44 cascading configuration 65 CKD-to-FBA conversions 28 combining with other applications 57 Command View XP troubleshooting management station 179 Command View XP or XP Remote
firmware, supported 11 G graph, Usage Monitor 136 graphical user interface 25 H hardware installation 50 help, obtaining 12, 13 History Operation pane components 142 displaying 141 saving information as snapshot file 144 horctakeover command 149 host failover software 19, 26 HP storage web site 13 Subscriber’s Choice for Business web site 13 technical support 12, 185 HP Performance Control XP 60 I I/O statistics 138 initial copy operations 26 initiator ports 25, 85 installing hardware 50 integrity, data
mirrors deleting data volumes from 113 restoring 112 splitting 111 modes, optional 20 monitoring displaying Usage Monitor pane 135 graph, displaying 136 I/O statistics 138 saving data 139 starting 136 stopping 136 multiple secondary arrays 20 multi-target configuration 67 multivolume files 40 N network relay devices, powering off 187, 189 nonvolatile storage (NVS) 44 NVS (nonvolatile storage) 44 O obtain function 18 operational parameters 20 operations overview 15, 17 optional modes 20 Optional Operations
troubleshooting 180 NVS (nonvolatile storage) 44 pair status 31 pinned track recovery 191 ports, configuring 85 powering off 130, 187, 188 read and write I/O operations 27 remote copy connections 51 remote, displaying information about 83 secondary arrays, functioning as 21 serial numbers 21 settings 146 storing journals 28 suspension conditions 35, 36 switching back to after disaster recovery 150 target ports 25 temporary path failures 54 update copy operation 27 primary site resuming normal operations 150
troubleshooting 180 multiple 20 NVS (nonvolatile storage) 44 pair status 31 pinned track recovery 191 ports, configuring 85 powering off 130, 187, 188 primary arrays, functioning as 21 read and write I/O operations 27 remote copy connections 51 remote, displaying information about 83 restoring journals 29 serial numbers 21 settings 146 suspension conditions 35, 36 target ports 25 temporary path failures 54 update copy operation 27 write operations 44 secondary site, switching to 149 secondary volumes Busine
Usage Monitor pane components 135 displaying 135 graph, displaying 136 I/O statistics 138 saving data 139 starting monitoring 136 stopping monitoring 136 V VMA (volume management area) 41 W web sites HP documentation 12, 13 HP storage 13 HP Subscriber’s Choice for Business 13 200
