HP StorageWorks Continuous Access XP Journal user guide for the XP12000/XP10000/SVS200 Part number: T1778-96005 Sixth edition: September 2007
Legal and notice information © Copyright 1999, 2007 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 . . . . . . . . . . . . . . . . . . . . . . . . . . Intended audience . . . . . . . Prerequisites . . . . . . . . . Related documentation . . . . . Document conventions and symbols HP technical support . . . . . . Subscription service . . . . . . HP websites . . . . . . . . . . Documentation feedback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Overview of HP copy solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Switching Continuous Access XP's master journal group Continuous Access XP Journal data volume pair status . . . Continuous Access XP Journal split types (PSUS) . . . Continuous Access XP Journal suspend types . . . . . Continuous Access XP Journal suspension conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4 Configuring arrays and logical paths . . . . . . . . . . . . . . . . Displaying the DKC Operation pane . . . . . . . . . . . . . . . . If Continuous Access XP Journal has not been started . . . . . . . If Continuous Access XP Journal has already been started . . . . . DKC Operation pane components . . . . . . . . . . . . . . . Displaying information about remote arrays . . . . . . . . . . . Displaying information about logical paths . . . . . . . . . . .
Displaying the History Operation pane . . . . . . . . . . If Continuous Access XP Journal has not been started . . If Continuous Access XP Journal has already been started History Operation pane components . . . . . . . . . Saving operation history to a text file . . . . . . . . . . . 9 Performing optional operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
12 Troubleshooting . . . . . . . . . . . . . . . . . . . . . . . . General troubleshooting . . . . Troubleshooting Continuous Access Continuous Access XP Journal error Calling HP . . . . . . . . . . . . . . . . . . . . XP Journal operations codes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figures 1 ..Continuous Access XP Journal components for Fibre Channel connections . . . . . . . 20 2 ..Continuous Access XP Journal components . . . . . . . . . . . . . . . . . . . . . 21 3 ..Connection configuration for multiple secondary arrays . . . . . . . . . . . . . . . 22 4 ..Corresponding consistency group and journal group numbers . . . . . . . . . . . . 27 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 5 ..Remote copy operations 6 ..
37 ..Trns. bus. tsks – DC1 W/O jrn. pr. to DC1 W/jrn. pr. (via del resy. ops.– 3DC mul-trg. conf.) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 38 ..Recovering from primary site and Continuous Access XP Synchronous secondary site failures 93 39 ..Transferring business tasks from Continuous Access XP Journal secondary sites to primary sites 94 40 ..Copying access attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 41 ..DKC Operation pane, DKC option . .
77 ..Typical SIM showing reference code and SIM type . . . . . . . . . . . . . . . . .
Tables 1 ..Document conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 ..Emulation types for journal volumes . . . . . . . . . . . . . . . . . . . . . . . . 14 24 3 ..Metadata information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 ..Continuous Access XP Journal data volume pair status . . . . . . . . . . . . . . . . 32 39 5 ..Continuous Access XP Journal split types (PSUS) 43 . . . . . . . . . . . . . . . . . . 6 ..
38 ..Config. def. file settings (HORCM_MON, HORCM_CMD) & specif. vals. in NAS Blade sys. 192 39 ..Example 1: RAID Manager XP configuration definition file (for instance number 16) . . . 193 40 ..Command execution that lists LUs constituting the file system sample . . . . . . . . . . 194 41 ..Example 2: RAID Manager XP configuration definition file (for instance number 16) . . . 194 42 ..Config. def. file settings (HORCM_INST) & specif. vals. in NAS Blade system . . . . . 194 43 ..
About this guide This guide provides information about: • HP StorageWorks Continuous Access XP Journal on HP StorageWorks XP12000 Disk Arrays • HP StorageWorks Continuous Access XP Journal on HP StorageWorks XP10000 Disk Array Intended audience This guide is intended for guide is intended for customers and HP-authorized service providers with knowledge: • Disk array hardware and software • Data processing and RAID storage subsystems and their basic functions • Basic Java classes and interfaces Prerequisit
Document conventions and symbols Table 1 Document conventions Convention Element Blue text: Table 1 Cross-reference links and e-mail addresses Blue, underlined text: http://www.hp.
• • • • • Technical support registration number (if applicable) Product serial numbers Error messages Operating system type and revision level Detailed questions Subscription service HP recommends that you register your product at the Subscriber's Choice for Business website: 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.
About this guide
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.
data at primary and/or secondary sites and provide maximum flexibility in data backup and duplication activities. See Combining Continuous Access XP Journal and Business Copy XP for more information. 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.
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 Continuous Access XP Journal components for Fibre Channel connections Journal obtain This function stores pre-existing primary volume data contnets as the base-journal in the primary site journal volume. This function then 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.
Components Continuous Access XP Journal operations involve local arrays at the primary and secondary sites, physical communications paths between these arrays, and Command View XP or XP Remote Web Console. Continuous Access XP Journal copies the original online data at the primary site to backup volumes at the secondary site through dedicated Fibre Channel remote copy connections using a journal volume.
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. The secondary array consists of the remote control unit (secondary array) and SVP.
Continuous Access XP Journal secondary data volumes). The secondary array continuously 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 array at the secondary site on a separate LAN. The secondary array 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.
Journal volumes When using Continuous Access XP Journal, updates to primary data volumes are stored in other volumes, which are called journal volumes. Updates (which are sometimes referred to as update data) that are stored in journal volumes are called journal data. Because journal data is stored in journal volumes, you can perform and manage highly reliable remote copy operations without suspension of remote copy operations.
CAUTION: You must register journal volumes in a journal group before creating a data volume pair in the journal group for the first time.
NOTE: If the metadata or journal data cannot be stored for a given length of time because the metadata or journal data areas are full with metadata or journal data that has not been discarded, the pair is suspended. Use a Command View XP management station or XP Remote Web Console to specify this timeout period (Data overflow watch) as a journal group option. This timeout period must be within the range of 0 to 600 seconds. For more information, see Changing options for journal groups.
Figure 4 Corresponding consistency group and journal group numbers Remote copy connections Remote copy connections are the physical paths that primary arrays use to communicate with secondary arrays. Remote copy connections enable communications between primary and secondary arrays. Primary and secondary arrays are connected through Fibre Channel interface cables. You must establish paths from the primary array to the secondary array, and also from the secondary array to the primary array.
NOTE: HP strongly recommends establishing at least two independent remote copy connections from the primary array to the secondary array and vice versa to provide hardware redundancy for this critical communications path. Initiator and RCU target ports Initiator and RCU target ports are required at the primary and secondary arrays. The initiator port at the primary array is connected to the RCU target port at the secondary array through a Fibre Channel interface.
When using Continuous Access XP Journal as a data migration tool, host failover is recommended but not required. When using Continuous Access XP Journal as a disaster recover tool, host failover is required to ensure effective disaster recovery operations. When a Continuous Access XP Journal pair is suspended due to an error condition, the primary array generates sense information that should be transferred to the remote site using the host failover software for effective disaster detection and recovery.
The primary array's base-journal data is sent to the secondary array journal volume and then the secondary data volume (in response to a read command) issued from the secondary array. The initial copy operation finishes when all base-journals have been destaged to the remote secondary volume.
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. This command transfers control parameters and FBA-format data for consecutive updated records in a track using a single write operation.
Table 3 Metadata information Type Description Journal type Journal type (for example, base-journal or update journal) LDEV No. (data) Number of the primary data volume that stores the original data Original data storing position Primary data volume's slot number, and start and end of sub-block number (data length) LDEV No.
1. The secondary array provides a status indicator number as management information to distinguish (by sequence number), those data entries that have successfully been received at the remote site. If the status is 1, that journal data has arrived at the secondary array. If the status is 0, journal data has not arrived yet. The secondary array determines whether or not journal data still needs to be settled according to this status indicator number.
• Maintains sequence numbers in periods of low host activities. Journal group operations Journal groups maintain update sequence consistency across a group of journal volumes. The primary and secondary data volumes of the pairs in a journal group must be located logically within one physical primary array and one physical secondary array (1-to-1 requirement).
Journal obtain in Continuous Access XP Synchronous secondary sites To perform the DC2-to-DC3 delta resync operation when a DC1 failure occurs, you must first obtain journal data in DC2 the Continuous Access XP Synchronous secondary site with a 3DC multi-target configuration.
For information about the delta resync operation that is performed when a failure occurs in the configuration shown in Figure 7, see Switching Continuous Access XP's master journal group. The contingency Continuous Access XP Journal pair must be in HOLD status when the failure occurs in order to perform a delta resync operation.
Figure 8 Delta resync activation in a 3DC multi-target configuration (after failure occurs) In Figure 8, because a failure occurs in primary site A (DC1), the horctakeover command is used at DC2 to change the former Continuous Access XP Synchronous secondary site (DC2) to be the primary site. If you perform a DC2-to-DC3 delta resync operation in this situation, the contingency DC2-to-DC3 Continuous Access XP Journal pair for the delta resync operation in Figure 7 becomes synchronized and usable.
volume is copied to the secondary data volume, or the delta resync operation finishes without any further processing. In other words, any contingency to avoid a full DC2-to-DC3 copy in the event of a DC1 problem will be abandoned if the difference between the DC2 and DC3 data get too large to store all the necessary delta data.
for Command View XP or XP Remote Web Console, this SIM results in an SNMP trap indicating the reason for the suspension. • Continuous Access XP Journal pair status The Continuous Access XP Journal Suspending and Deleting transitional states occur when a request to change the Continuous Access XP Journal pair status is accepted, but the requested status change (PSUS, PSUE, or SMPL) is not complete yet. These states are not reported to the host.
Pair status Description Primary data volume access Secondary data volume access PSUS (pair suspended-split) (see Table 5 for split types) Data volume pair is not synchronized because you split this pair (pairsplit-r) or deleted this pair from the secondary array (pairsplit-S). For Continuous Access XP Journal pairs, the primary and secondary arrays keep track of any journal data discarded during the pairsplit-r operation.
Pair status Description Primary data volume access Secondary data volume access PFUL Continuous Access XP Journal monitors the total amount of data in the journal volume. If the amount of data exceeds the threshold (80%), the pair status changes from COPY or PAIR to PFUL. The write data that inflows is monitored during the specified time (Data Overflow Watch). You can use Command View XP or XP Remote Web Console to set the monitoring time period (the default is 60 seconds).
Pair status Description Primary data volume access Secondary data volume access Deleting Pair is not synchronized and is in transition from PAIR, COPY, or PSUS/PSUE to SMPL. When the pairsplit-S operation is requested, the status of all affected pairs changes to Deleting. When the pairsplit-S operation is complete, the status changes to SMPL. Read / Write Read Only HOLD Contingency pair is prepared for a delta resync operation.
Table 5 Continuous Access XP Journal split types (PSUS) Type Applies to Description PSUS, P-VOL by Operator Primary data volume The secondary data volume's split type is PSUS-by MCU. PSUS, S-VOL by Operator Primary data volume, Secondary data volume You split the pair from the primary or secondary array using the secondary data volume option. PSUS, by MCU Secondary data volume The secondary array received a request from the primary array to split the pair.
Continuous Access XP Journal suspend types Table 7 describes the Continuous Access XP Journal suspend types, which indicate the reason for the suspension. Continuous Access XP Journal pairs can be suspended by the primary or secondary array. For more information, see Continuous Access XP Journal suspension conditions.
shared memory. When a Continuous Access XP Journal pair is suspended, tracks containing the following journal are marked in the differential bitmap as modified (to be copied during the pairresync operation): • Journal data the primary array has created, but has not sent to the secondary array yet After marking these primary data volume tracks as modified, the primary array discards the journal data.
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.
NOTE: Administrator or Continuous Access XP Journal write access to Command View XP or XP Remote Web Console is required to perform Continuous Access XP Journal operations. Users without Administrator or Continuous Access XP Journal write access can only view Continuous Access XP Journal information.
Table 10 lists the volumes and volume capacities that can be used for Continuous Access XP Journal data and journal volumes. NOTE: The capacity of a journal volume is not included in the accounting capacity.
Table 12 Journal volume specifications Item Support specifications Emulation type Same NOTE: Only OPEN-V can be used for journal volumes. Volume capacity Same or different amount of capacity is available Table 13 list the RAID-level combination of data and journal volumes in the journal group that can be used for Continuous Access XP Journal.
NOTE: If you use LUSE volumes, you must add 1 to the value calculated by the above formula. • Maximum number of pairs that can be created: Use the following formulas to calculate the maximum number of pairs you can create: • When one pair of extended shared cache memories is used Maximum number of pairs = ( ↓ (11,605 /Number of bitmap areas) ↓ ) The ↓…↓ symbols around a value indicate the value should be rounded down to the nearest integer.
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 memories1. NOTE: Only the HP service engineer can install shared memory options and apply High Performance mode.
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).
and you use the CL1-A port as an RCU target port in the Continuous Access XP Journal primary site, you must not use the CL1-A or CL5-A port as an RCU target port in the Continuous Access XP secondary site. As these examples indicate, you must avoid using illegal combinations of an RCU target port in the Continuous Access XP Journal primary site and a target port in the Continuous Access XP Journal primary site.
Location (channel adapter number) 1K (Add.4) 1L (Add.
Location (channel adapter number) Illegal port combinations Location (channel adapter number) Illegal port combinations CLB-L CLB-L, CLF-L CLB-C CLB-C, CLF-C CLD-L CL9-L, CLD-L CLD-C CL9-C, CLD-C CLF-L CLB-L, CLF-L CLF-C CLB-C, CLF-C CL9-M CL9-M, CLD-M CL9-D CL9-D, CLD-D CLB-M CLB-M, CLF-M CLB-D CLB-D, CLF-D CLD-M CL9-M, CLD-M CLD-D CL9-D, CLD-D CLF-M CLB-M, CLF-M CLF-D CLB-D, CLF-D Table 17 Illegal port combinations (for REAR CL2) Location (channel adapter number) 2R (Add.
Location (channel adapter number) 2W (Add.4) 2X (Add.
Location (channel adapter number) Illegal port combinations Location (channel adapter number) Illegal port combinations CLA-L CLA-L, CLE-L CLA-C CLA-C, CLE-C CLC-L CLC-L, CLG-L CLC-C CLC-C, CLG-C CLE-L CLA-L, CLE-L CLE-C CLA-C, CLE-C CLG-L CLC-L, CLG-L CLG-C CLC-C, CLG-C CLA-M CLA-M, CLE-M CLA-D CLA-D, CLE-D CLC-M CLC-M, CLG-M CLC-D CLC-D, CLG-D CLE-M CLA-M, CLE-M CLE-D CLA-D, CLE-D CLG-M CLC-M, CLG-M CLG-D CLC-D, CLG-D Installing hardware You and the HP representative in
Remote copy connections Figure 9 shows remote copy connection configurations for Continuous Access XP Journal operations. The primary and secondary arrays of each Continuous Access XP Journal pair must be connected using optical fiber cables. If you use multimode shortwave optical fiber cables, fibre cables up to 1.5 km in length and up to two switches are required for distances greater than 0.5 km.
Figure 11 Switch remote copy connections • Extender connections Figure 12 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.
The following factors affect the speed of writing to and reading from journal volumes, particularly when a temporary communications path failure occurs between the primary and the secondary arrays or when the amount of data transferred from hosts to the primary array increases: • • • • • RAID configuration of the RAID groups that will contain journal volumes Types of physical volumes in the RAID groups that will contain journal volumes Frequency of access to non-journal volumes in the RAID groups that will
A journal volume's data transfer speed depends on the data transfer speed of the RAID group to which the journal volume belongs. One RAID group can consist of one or more logical volumes, including journal volumes. Therefore, if frequent accesses are made to non-journal volumes in a RAID group, relatively fewer accesses can be made to journal volumes in the same RAID group, which can cause a drop in the journal volumes' data transfer speed.
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. The total capacity of the journal volumes in each journal group must exceed the value illustrated previously.
Figure 15 Data transfer speed between primary and secondary arrays Bandwidth of data transfer paths The bandwidth (that is, the data transfer speed) of data transfer paths between the primary and secondary arrays depends on your environment. The bandwidth must exceed the data transfer speed in “normal status” mentioned previously. Journal transfer speed of DKCs A DKC's journal transfer speed depends on the number of paths used for transferring journals.
NOTE: You must log in with Administrator or Continuous Access XP Journal write access, and you must be in Modify mode. If you are in the View mode, you can view the Continuous Access XP Journal pair status, but you cannot modify the settings. 3. Click the Continuous Access XP Journal button ( ). The Journal Operation pane appears. 4. On the DKC Operation pane, select Port, and set the initiator and RCU target ports (see Configuring port attributes). 5. On the DKC Operation pane, select DKC.
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? P-VOL that is also used as a Continuous Access XP P-VOL or Continuous Access XP S-VOL No. No. No. P-VOL (none of the above) Yes.1 Yes.1 No. S-VOL in PSUS status Yes.1 No. No. S-VOL (none of the above) No. No. No. Reserved volume No. No. No. P-VOL in COPY(RS-R) status No. No. No.
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? S-VOL in PAIR status Yes.1 No. No. S-VOL in PSUS status Yes.1 No. No. S-VOL in SSWS status Yes.1 No. S-VOL in PSUE status Yes.1 No. No. No. No. No. No. No. No. System volume No. No. No. User volume Yes. Yes. No. Volume with Read/Write attribute Yes. Yes. Yes. Volume with Protect attribute Yes. Yes. No.
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 to which a path is defined Yes. Yes. No. Volume to which no path is defined No. No. No. Volume to which LUN security is applied Yes. Yes. No.
Combining Continuous Access XP Journal and Business Copy XP You can use Continuous Access XP Journal and Business Copy XP together in the same array and on the same volumes to provide multiple copies of data at the primary and/or remote sites. Table 19 describes host pair status reporting for Continuous Access XP Journal volumes, Business Copy XP volumes, and Continuous Access XP Journal/Business Copy XP shared volumes.
Figure 16 Shared Continuous Access XP Journal secondary data volume and Business Copy XP P-VOL • Continuous Access XP Journal/Business Copy XP configurations that share the Continuous Access XP Journal secondary data volume and Business Copy XP P-VOL Figure 17 shows an example of a Continuous Access XP Journal secondary data volume that also functions as a Business Copy XP P-VOL.
Figure 18 Shared CA XP Jour. pri. data vol. & BC XP P-VOL, & CA XP Jour. sec. data vol. & BC XP P-VOL • Continuous Access XP Journal/Business Copy XP configuration where a Business Copy XP S-VOL in PSUS status is used as a Continuous Access XP Journal primary data volume In Figure 19, the Business Copy XP S-VOL in PSUS status also functions as a Continuous Access XP Journal primary data volume. This configuration allows Continuous Access XP Journal to make a remote backup copy of the Business Copy XP S-VOL.
Figure 21 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 22 Restoring Business Copy XP P-VOLs, step 3 4. Execute the RAID Manager XP pairsplit -r command on the Continuous Access XP Journal pair to suspend the pair. Figure 23 Restoring Business Copy XP P-VOLs, step 4 5.
Figure 25 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 26 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 27 Restoring Business Copy XP P-VOLs, step 8 9. Execute the RAID Manager XP pairresync command on the Continuous Access XP Journal pair to resynchronize the pair.
Figure 28 Restoring Business Copy XP P-VOLs, step 9 Combining Continuous Access XP Journal with Continuous Access XP (3DC cascading configuration) The local array allows Continuous Access XP Journal and Continuous Access XP Synchronous to be combined. This combination is intended to ensure that the response time against host I/Os is comparable, regardless of whether the distance between the primary and secondary sites is short or long.
Basic behavior This section explains the basic behavior of a 3DC cascading configuration under the following conditions: • The Continuous Access XP Synchronous pair status is PAIR. The Continuous Access XP Journal pair status is also PAIR. • A host in the primary site issues an I/O request to the Continuous Access XP Synchronous P-VOL As illustrated in Figure 30, if a host in the primary site issues an I/O request, data is stored in the Continuous Access XP Synchronous P-VOL.
• Local array at the primary site • Continuous Access XP Synchronous installed • Continuous Access XP Synchronous P-VOL • Local array at the Intermediate site • Continuous Access XP Synchronous installed • Continuous Access XP Journal installed • Volume used as a Continuous Access XP (synchronous) S-VOL and a Continuous Access XP Journal primary data volume • Continuous Access XP Journal master journal volume • Local array at the remote site • Continuous Access XP Journal installed • Continuous Access XP Jo
3. Wait until the Continuous Access XP Synchronous S-VOL in the intermediate site is suspended (in SSWS status) or changed to a Continuous Access XP Synchronous P-VOL. After the S-VOL is suspended or changed to a P-VOL, you can use the volume to resume your business tasks. For more information about using RAID Manager XP, see the HP StorageWorks RAID Manager XP user guide.
data volume, asynchronously with regards to writing from the host. Update sequence consistency is maintained with the Continuous Access XP Journal secondary data volume.
• Local array at the primary site • Continuous Access XP Synchronous installed • Continuous Access XP Journal installed • Volume used as a Continuous Access XP Synchronous primary volume and as a Continuous Access XP Journal primary data volume • Continuous Access XP Journal master journal volume • Local array at the Continuous Access XP Synchronous secondary site • Continuous Access XP Synchronous installed • Continuous Access XP Journal installed • Continuous Access XP Synchronous secondary volume NOTE: T
NOTE: If you delete the Continuous Access XP Synchronous pair created in step 2, the Continuous Access XP Journal delta resync contingency pair created in step 6 is deleted as well. In addition, if you delete the Continuous Access XP Journal pair created in step 6, the secondary data volume of the Continuous Access XP Journal pair created in step 6 is deleted. In this case, you can only perform the delete operation on the remaining primary data volume.
NOTE: Right after you create the Continuous Access XP Journal pair for delta resync operations or if a failure occurs at the primary site right after the recovery of a Continuous Access XP Synchronous or Continuous Access XP Journal pair, only part of the differential data between the primary and secondary data volume might be stored in the master journal group. When any of these requirements is not fulfilled, an error usually occurs and delta resync operations fail.
Access XP Journal pair's secondary data volume must be the secondary data volume at the Continuous Access XP Journal secondary site. b. If the Continuous Access XP Synchronous pair's primary/secondary relationship is reversed (that is, the copy direction is reversed as illustrated in the lower-right portion of Figure 32( because the primary site is no longer operational), re-create the Continuous Access XP Journal pair deleted in step 1 via full copy.
Figure 32 Recovering from primary site disasters Recov. from prim. site fail. via delta resync (50-08-xx or greater FW) (prior to 50-08-xx FW) If a disaster or failure occurs only in the primary site in a 3DC multi-target system, you can use the secondary volume in the Continuous Access XP Synchronous secondary site to enable the secondary site to take over your business tasks.
XP Synchronous secondary site, the status of the Continuous Access XP Journal pair's data volumes changes as explained in Table 21. Table 21 Changes to CA XP Jour. pair status by delta resync ops.
Figure 33 Recovering from primary site disasters (when performing delta resync operations) Changing back to a 3DC multi-target configurations after recovering from primary site failures If you follow the instructions in Recovering from primary site failures and the system does not change to a 3DC cascading configuration due to primary site failures, you have the ability to change the system to a 3DC multi-target configuration after removing failures from the primary site and other locations.
Figure 34 Changing to a 3DC multi-target configuration after recovering from primary site failures Trans. business tasks from CA XP sec. sites to pri. sites (in 3DC casc. configs.) If you follow the instructions in Recovering from primary site failures and the system changes to a 3DC cascading configuration, you can transfer your business tasks back to the primary site after removing failures from the primary site and other locations. To transfer your business tasks back to the primary sites: 1.
Figure 35 Transferring business tasks from CA XP second. sites to prim. sites (in 3DC cascading configs.) Transfer. business tasks from CA XP sec. sites to prim. sites (in 3DC multi-target configs.
Figure 36 Transfer. business tasks from CA XP sec. sites to prim. sites (in 3DC multi-target configs.) Trans. bus. tsks – DC1 (W/O a jur. pr.) to DC1 W/ jur. pr. (del. res. op. in 3DC multi-tar. conf) If you remove failures from the primary site and other locations and change the system to a 3DC multi-target configuration (see Recovering from primary site failures (when performing delta resync operations)), you can transfer your business tasks back to the primary site.
Table 22 Chgs to CA XP Jrn. pr. stat. by del. resyn. ops. (trs. bus. tsks. btw. CA XP 2nd to pri.
No. If the Status of the Pair in the Primary Site is and the Status of the Pair in the UR Secondary Site is 3 PSUE or PSUS HOLD Continuous Access XP Journal operations requirements and restrictions Perform the Following Operation before Transferring Business Tasks Back to the Primary Site 1. Remove the Continuous Access XP Journal pair from the primary site. 2. Make sure that all the pairs belonging to the journal group in the primary site are removed. 3.
No. If the Status of the Pair in the Primary Site is and the Status of the Pair in the UR Secondary Site is Perform the Following Operation before Transferring Business Tasks Back to the Primary Site 4 HOLD HOLD The pair is in normal condition. You can transfer business tasks as described earlier in this section. 5 HLDE HOLD Change the status of the HLDE pair back to HOLD. For information about how to do this, see section 6.6. 6 SMPL HOLD 1.
Figure 37 Trns. bus. tsks – DC1 W/O jrn. pr. to DC1 W/jrn. pr. (via del resy. ops.– 3DC mul-trg. conf.) Recovering from primary site and Continuous Access XP Synchronous secondary site failures If a disaster or failure occurs in the primary site and Continuous Access XP Synchronous secondary site in a 3DC multi-target configuration, you can resume your business tasks using the secondary volume in the Continuous Access XP Journal secondary site.
Figure 38 Recovering from primary site and Continuous Access XP Synchronous secondary site failures Transferring business tasks from Continuous Access XP Journal secondary sites to primary sites If you follow the instructions in Recovering from primary site and Continuous Access XP Synchronous secondary site failures and remove failures from the primary site and Continuous Access XP Synchronous secondary site, you can transfer your business tasks back to the primary site.
Figure 39 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. However, you cannot specify a volume with the S-VOL Disable attribute as the pair's secondary data volume.
Table 24 Access attribute and availability as a Continuous Access XP Journal pair volume Access attribute Availability as a Continuous Access XP Journal pair volume Primary data volume Secondary data volume Read/Write OK OK Read Only OK OK Protect OK OK S-VOL Disable OK NO Read Only and S-VOL Disable OK NO Protect and S-VOL Disable OK NO 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 a
Figure 40 Copying access attributes In Figure 40, if both primary and secondary data volumes have VMA, the primary data volume's VMA is copied to the secondary data volume. In Figure 40, the secondary data volume's access attribute is Read/Write when the pair status is SMPL. After the primary data volume's VMA is copied to the secondary data volume, the secondary data volume's access attribute becomes Read Only, which is the same as the primary data volume's access attribute.
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.
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.
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 the user name and password, and click OK. The Command View XP or XP Remote Web Console main pane appears. For more information, see the HP StorageWorks Command View XP user guide for XP Disk Arrays or the HP StorageWorks XP Remote Web Console user guide for XP12000/XP10000/SVS200. 2.
• • • • • Detailed information about remote arrays (see Displaying information about remote arrays) • Detailed information about logical paths (see Displaying information about logical paths) • Detailed information about ports on the local array (see Displaying information about ports on local arrays) Preset: Displays changes made in the pane. When you change settings in the pane, the changes are displayed in the Preset list before the changes are applied to the arrays.
NOTE: In the current version, the path group ID is set to the default value. • M-R Path: Channel type of the logical path between the local and remote arrays. This column always displays Fibre. • Status: Indicates whether logical paths failed. • Normal: No failure occurred at the logical paths • Failed: All logical paths failed • Warning: Some logical paths failed • Num of Path: Number of logical paths.
Figure 42 DKC Operation pane, Port option • 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.
• Target: A target port is a Fibre Channel port used to connect an array and a host. When the host issues a write request, the request is sent to a volume through a target port on the array. You must configure target ports on primary arrays for Continuous Access XP Journal operations. • Initiator: An initiator port is a Fibre Channel port that sends commands to a remote array. You must configure initiator ports on primary and remote arrays for Continuous Access XP Journal operations.
2. Display the DKC Operation pane (see Displaying the DKC Operation pane). 3. In the Switch Display box, select Port. The tree displays channel adapters in the local array and port attributes. 4. Do one of the following: • Select a channel adapter from the tree. • Select a port attribute (that is, target, RCU target, or initiator) from the tree. 5. Right-click the port you want to configure. 6. Select the port type (that is, initiator, RCU target, or target).
• Remote array's serial number. • Remote array's controller ID. NOTE: The controller ID for an XP12000/XP10000 is 4. • Path group ID. The path group ID is within the range of 1-FF (hexadecimal). Up to 64 path group IDs can be registered per array. NOTE: In the current version, the path group IDs always take the default values and cannot be changed. 8. In the Add DKC pane, select a primary and remote array port to configure a logical path. 9. Click Option. 10.
Figure 43 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. Path group IDs are used for identifying groups of logical paths. One path group can contain up to eight logical paths. Path group IDs are within the range of 0-FF (hexadecimal). If you select Default, the default path group IDs is set.
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.
an error occurs, the local array suspends all affected Continuous Access XP Journal (and Universal Replicator for z/OS) pairs to prevent remote copy operations from adversely affecting performance due to an inadequate number of paths. If the local array contains Continuous Access XP Journal pairs that contain critical data for disaster recovery, set the minimum number of paths to one, so Continuous Access XP Journal operations continue even if there is only one path to a remote 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 45 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.
1. Ensure that Command View XP or XP Remote Web Console is in Modify mode. For more information, see the HP StorageWorks Command View XP user guide for XP Disk Arrays or the HP StorageWorks XP Remote Web Console user guide for XP12000/XP10000/SVS200. 2. Display the DKC Operation pane (Displaying the DKC Operation pane). 3. In the Switch Display box, select DKC. 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).
Figure 46 Logical paths list and Preset list (delete path operations) Viewing the status of logical paths The DKC Status pane displays the status of logical paths. To view the status of logical paths, the primary and secondary array administrators can perform the following procedure: 1. Display the DKC Operation pane. 2. In the Switch Display box, select DKC. 3. In the upper-right list, right-click a remote array or logical path. • To display remote arrays in the list, select Subsystem from the tree.
Figure 47 DKC Status pane • List: • No.: Serial numbers used for rows in the list. • Path Status: Logical path status (see Table 28). • 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. • M-R Path: Type of channel interface between the local and remote arrays. This column displays fibre. • Minimum Paths: Minimum possible number of paths between the local and remote arrays.
Table 28 Logical path status Status Remarks Normal This path was successfully established and can be used for Continuous Access XP Journal remote copy activities. Nothing An operation for configuring or deleting this logical path is in progress. 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.
9. In the Preset list, check the settings you made. • The Preset list displays the specified remote arrays. One row indicates one remote array. • To cancel removal of the relationship, right-click the remote array in the Preset list, and select Cancel. 10. Click Apply to remove the relationship.
Configuring arrays and logical paths
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.
If Continuous Access XP Journal has already been started • Select the Journal Operation tab. The Journal Operation pane appears. Journal Operation pane components Figure 48 Journal Operation pane (XP12000) Figure 49 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. • Registered: When selected, the upper-right list displays journal groups in which journal volumes are registered. When you double-click this item, the tree displays journal groups in which journal volumes are registered.
One row in this list represents one mirror (or one journal group). • JNLG: Number of a journal group in the local array. • Attribute: Attribute of a journal group in the local array. 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.
Initial Active Journal group in initial status. 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: • Initial copy is in progress. The primary and secondary data volumes are not synchronized. • Initial copy is finished.
• 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: Displays the CLPR number for the journal group. If more than one JNL volume is registered in JNL group, the CLPR number of the first registered JNL volume is displayed.
To register journal volumes, the primary and secondary array administrators must perform the following procedure: 1. Ensure that Command View XP or XP Remote Web Console is in Modify mode. For more information, see the HP StorageWorks Command View XP user guide for XP Disk Arrays or the HP StorageWorks XP Remote Web Console user guide for XP12000/XP10000/SVS200. 2. Display the Journal Operation pane (Displaying the Journal Operation pane). 3.
7. Click Set to close the Edit JNL Volumes pane. 8. In the Preset list, check the settings you made. • To add volumes to register, right-click a volume in the Preset list, and select Modify. A pane appears where you can add volumes. • To cancel registering a volume, right-click the volume, and select Cancel. • If necessary, repeat steps Step 3 through Step 8 to add volumes to other journal groups. 9. Click Apply to register journal volumes.
• Emulation: Journal volume's emulation type. • CLPR: Number and name of the CLPR where the journal volume belongs. • Operation: Displays one of the following: • Blank: This column is usually blank. • Add: Volume to be added to a journal group. • Delete: Volume to be deleted from a journal group. • Add: Registers volumes in a journal group. When registering volumes, select the volumes from Free Volumes, and click Add to add the volumes to JNL Volumes. • Delete: Deletes volumes from a journal group.
• CU: Displays volumes belonging to a CU. If selected, select a CU from the list to the right. The Free Volumes list displays volumes in the selected CU. Figure 52 The Free Volumes List and the CU Button in the Edit JNL Volumes Panel • • • • • JNL Group: Journal group's number. Current: Number and capacity of journal volumes that currently exist in the journal group. Total: Estimated number and capacity of journal volumes after addition or deletion.
Table 29 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. In the JNL Volumes list of the Edit JNL Volumes pane (Figure 50), select the volumes you want to delete. In the JNL Volumes list, one row represents one volume. 6. Click Delete. The volumes you want to delete appear in blue italics.
Figure 53 JNL Group Detail pane • JNL Group: Journal group's number. • Attribute: Journal group's attribute. When the journal group uses multiple mirror IDs, Attribute indicates the attribute of the data volume in the journal group whose mirror ID is not Hold or Hold(Failure). • 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.
NOTE: Inflow Control displays nothing if the journal group is a 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.
• CU:LDEV: Journal volume's CU and LDEV number. The CU number appears to the left of the colon (:). The LDEV number appears to the right of the colon. • Capacity (GB): Journal volume's capacity in gigabytes. • Emulation: Journal volume's emulation type. • CLPR: Number and name of the CLPR where the journal volume belongs. • Mirrors: List of mirrors. • Mirror ID: Mirror ID. This column is blank if the journal group's attribute is neither Master nor Restore. • Attribute: Mirror ID'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. 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. Active Either of the following: • Initial copy is in progress. The primary and secondary data volumes are not synchronized. • Initial copy is finished.
• Path Watch Time (min): Time for monitoring blockade of paths to the remote array. If the status of a mirror where secondary data volumes in the local array are registered is Hold or Hold(Failure), this column is blank. • 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.
• To modify a setting, right-click the setting in the Preset list, and select Modify. • To cancel a setting, right-click the setting, and select Cancel. 6. Click Apply to apply the settings. 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.
• Data Overflow Watch (0 to 600 sec): Time (in seconds) for monitoring whether metadata and journal data are full. This value must be within the range of 0 to 600 seconds. NOTE: If Inflow Control is No, Data Overflow Watch does not take effect and does not display anything. • Copy Pace: Pace (speed) for an initial copy activity for one volume. The default is Low. • Low: Speed of initial copy activity is slower than Medium and High.
CAUTION: If you want to downgrade the DKCMAIN microprogram version to lower than 50-09-00-00/00, you must set the Path Watch Time within the range of 1 to 59 minutes before downgrading DKCMAIN, or set it within the range of 1 to 60 minutes after downgrading DKCMAIN. If the Path Watch Time is outside the range of 1 to 59 minutes before you downgrade DKCMAIN, the Path Watch Time might not be displayed after DKCMAIN is downgraded. Use the JNL Group Detail panel to check the Path Watch Time.
• Use: Journal data is stored in cache. NOTE: When there is insufficient space in cache, journal data is also stored in the journal volume. • Not Use: Journal data is not stored in cache. • Blank: The current Use of Cache setting remains unchanged. CAUTION: This setting does not take effect on master journal groups. However, if you use the RAID Manager XP horctakeover command to change a master journal group to a restore journal group, this setting takes effect on the journal group.
5. In the Preset list, check the journal groups you want to delete. • To cancel deleting a journal group, right-click the journal group, and select Cancel. • 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.
Figure 55 Pairsplit-r pane • S-VOL Write: Specifies whether hosts can issue read and write I/Os to the secondary data volume after the mirror is split. • 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.
5. From the mirrors, select and right-click the mirrors you want to restore. The result of this operation differs according to the status of the selected mirror. • When selecting a mirror in Stop status, data copying is restarted after the recovery. • When selecting a mirror in Hold status, a delta resync operation is performed after the recovery. • When selecting a mirror in Hold(Failure) status, the status changes to Hold after the recovery. 6. Select JNL Groups and Pairresync. 7.
• If necessary, repeat step 3 through step 7 to specify other mirrors. NOTE: The Range column indicates the unit used for splitting the mirror, and always displays Group. The Mode column indicates how data volumes are deleted. 8. Click Apply to delete data volumes from the mirrors. 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.
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.
Figure 56 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.
To the right of the icon appears information about an LU path, which is a path from hosts to the logical volume. Figure 57 Information in Port-G-ID-LUN column on the Pair Operation pane 1 Port that receives read and write requests from hosts. In Information in Port-G-ID-LUN column on the Pair Operation pane, read and write requests are sent to the logical volume via the CL3-E port. 2 Group number of the host group or iSCSI target.
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. This column indicates a secondary data volume if the remote array is a secondary data volume. The following is an example of an entry in this column.
• If the volume pair is split and, therefore, is in PSUS or PSUE status, this column usually displays the synchronization rate (that is, concordance rate) between the secondary data volume before it was split and the secondary data volume after it was split. For example, the synchronization rate is 100 percent if the contents of the secondary data volume is the same before and after the volume pair was split.
NOTE: Information on the Pair Operation pane is updated when you do one of the following: • Select another tab, and reselect the Pair 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. Filtering information in the list on the Pair Operation pane Use the Display Filter pane to filter information in the list on the Pair Operation pane. The list displays only volumes that satisfy certain conditions. 1.
• G-ID: Group number of the host group or iSCSI target to display. To display all host groups and iSCSI targets, select ALL. • JNL Group: Journal group to be displayed. To display all journal groups, select ALL. • Mirror: Mirror to be displayed. To display all mirrors, select ALL. • P-VOL/S-VOL: Type of volumes to be displayed. • To display all primary data volumes, select P-VOL. • To display all secondary data volumes, select S-VOL. • To display all primary and secondary data volumes, select ALL.
1 Port in the local array 2 Group number and name of the host group or iSCSI target in the local array (host group name is limited to a maximum of 16 characters) 3 LU number in the local array 4 CU and LDEV numbers in the local array 5 Pair status 6 Journal group number in the local array 7 Mirror ID 8 Remote array's serial number and controller ID 9 Port number in the remote array 10 Group number of the host group or iSCSI target in the remote array 11 LU number in the remote array 12
After all these points are confirmed, the administrator of the primary array can use the following procedure to create a pair of a primary and secondary data volume. The administrator of the secondary array cannot perform this operation. 1. Ensure that Command View XP or XP Remote Web Console is in Modify mode. For more information, see the HP StorageWorks Command View XP user guide for XP Disk Arrays or the HP StorageWorks XP Remote Web Console user guide for XP12000/XP10000/SVS200. 2.
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: • P-VOL displays only one primary data volume even when two or more primary data volumes are selected in the Pair Operation pane. P-VOL only displays the primary data volume that has the smallest LUN (LU number). • If a logical volume is an external LU, # appears after the LDEV number. For more information about external LUs, see the HP StorageWorks External Storage XP user guide. • S-VOL: Select a secondary data volume.
NOTE: If you select and right-click more than one volume in the Preset list and select Modify, the Paircreate pane appears, and the Select Other S-VOL(s) list displays No Change. To change the setting with this list, select another item from the list. • Mirror ID: Assigns a master journal group to the primary data volume, assigns a restore journal group to the secondary data volume, and assigns a mirror ID to the volume pair. • M-JNL: Assigns a master journal group to the primary data volume.
NOTE: If a timeout error occurs during the Paircreate operation, the copy operation might not execute correctly in the order you set with the Priority parameter. The timeout error might be caused by the CU configuration or a remote copy path error. Review the error, delete the pair with the error, and retry the Paircreate operation. • Error Level: Range used for splitting a pair when a failure occurs.
Figure 61 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.: If the volume in the local array is a primary data volume, this column displays an initial copy operation's progress.
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.
• Other Information: Displays the following: • Established Time: Date and time the volume pair was created. • Updated Time: Date and time the volume pair status was last updated. • Refresh the Pair Operation tab after this window is closed: If selected, the Pair Operation pane is updated when the Pairdisplay pane closes. • Previous: Displays the pair status information for the previous pair in the list (the pair in the row above).
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 and gives detailed information about the error. Figure 62 Change Pair Option pane • 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.
10. In the list on the Pair Operations pane, verify that the pair was split successfully. If the pair was split, the pair status is PSUS. NOTE: To restore a pair that has already been split, use the Pairresync command (see Restoring a pair of data volumes). Figure 63 Pairsplit-r pane • S-VOL Write: Specifies whether hosts can write data to the secondary data volume. The default is Disable (that is, do not permit). • Disable: Hosts cannot write data to the secondary data volume while the pair is split.
When the secondary array receives a request for splitting a pair, all journal data (that is, update data) that the pair retains is written to the secondary data volume. After that, the pair's status changes from Suspending to PSUS if the pair does not retain any journal data for a certain time period. • Purge: When splitting the pair, update data is not copied to the secondary data volume. If you restore the pair later, update data is copied to the secondary data volume.
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. 11. In the list on the Pair Operations pane, verify that the pairs were successfully restored. • If the pair was successfully restored, the pair status is COPY or PAIR.
• Normal: A split pair whose status is PSUS or PSUE is recovered. • Delta: A delta resync operation is performed. For information, see Requirements for performing delta resync operations. • 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.
To monitor the progress of deleting pairs, click Refresh to update the information in the list, or use the Pairdisplay pane (see Displaying detailed information about a pair of data volumes) to monitor the detailed status of each pair. NOTE: To restore a pair that was deleted from the secondary array, delete the pair from the primary array, and restore the pair using the appropriate initial copy option.
NOTE: If the following conditions are satisfied and you click Apply, a warning message appears and processing cannot continue: • The Preset list contains two or more pairs belonging to the same consistency group. • The Range column displays Group for at least one of the above pairs. To continue processing, do one of the following: • Ensure that the Range column displays LU for all pairs in the same consistency group.
Performing Continuous Access XP Journal pair operations
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 66) displays user-selected remote copy I/O statistics and status of remote copy usage monitoring, allowing you to monitor remote operations.
Figure 66 Usage Monitor pane • Monitoring Switch: Usage monitor operation (see Starting and stopping usage monitoring). • Enable: Monitoring is on. • Disable: Monitoring is off. 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: If you set 1 minute for Gathering Interval, the sampling data is held one day. If you set 15 minutes for Gathering Interval, the sampling data is held 15 days. When you change Gathering Interval, data obtained before the change is deleted. 4. Click Apply. Stopping usage monitoring 1. Select the Usage Monitor tab. 2. In the Monitoring Switch box, select Disable. 3. Click Apply. NOTE: Depending on the SVP's load status, you might not be able to stop monitoring.
NOTE: If you specify the unmounted LU, the graph is not displayed. NOTE: When a graph displays, the CU and LDEV numbers appear above the graph. If the logical volume is an external LU, # appears after the LDEV number. For more information about external LUs, see the HP StorageWorks External Storage XP user guide. 4. In the Monitor Data box, select the I/O statistics data you want to display on the graph. You must select at least one box. Table 31 describes I/O statistics data. 5. Click Set.
Figure 67 Display Item pane Table 31 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. Unit is kilobytes per second. Write Transfer Rate Amount of data written per second. Unit is kilobytes per second.
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.
8 Viewing history of data volume pair operations Use the History Operation pane (Figure 68) 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 68 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.
• VOL: Volume manipulated during the operation. This volume is located in the local array. The number to the left of the colon (:) is the CU number. The number to the right of the colon is the LDEV number. • Paired VOL: Volume paired with the manipulated volume. This volume is located in the remote array. The number to the left of the colon (:) is the CU number. The number to the right of the colon is the LDEV number. • Operation: Operation performed. • Pair definition: Data volume pair was defined.
NOTE: • History information (that is, rows in the list) might not appear in chronological descending order. To sort the information in descending (or ascending) order, click a header in the list. • The History Operation pane can display history information for up to 65,535 operations. However, the pane can only display up to 16,384 operations at one time. To display the remaining operations, click Previous or Next. • The History Operation pane does not display history information older than seven days.
1 Data and time operation finished. 2 Time taken for copying (from start to end of copying). This is displayed only when the operation is Paircreate Complete or Pairresync Complete. 3 Journal group number. 4 Mirror ID. 5 CU and LDEV numbers of volume in the local array. CU number is on the left of the colon (:). LDEV number is on the right of the colon. 6 CU and LDEV numbers of volume in the remote array. CU number is on the left of the colon (:). LDEV number is on the right of the colon.
Viewing history of data volume pair operations
9 Performing optional operations Use the Optional Operations pane to perform the following procedures: • Displaying the Optional Operation pane • Changing array options 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.
Figure 69 Optional Operation pane • Tree: Select Subsystem. • List: Displays array option settings (see Changing array options). • Preset: When changing settings in the Optional Operation pane, the Preset list displays the changes. Changes are not applied to the array until you click Apply. • Operation: Operation that occurs when you click Apply. • Change system option: Changes array options. • Blank: Nothing occurs when you click Apply.
4. Do one of the following: • Right-click Subsystem in the tree, and select Change System Option. • Right-click information in the list, and select Change System Option. 5. In the System Option pane, change the array options. 6. Click Set. 7. In the Preset list, verify the changes you made. • To modify a change, right-click the change, and select Modify. A pane appears where you can modify the change. • To cancel a change, right-click the change, and select Cancel. 8. Click Apply.
Figure 72 Array option settings in the Optional Operation pane • Max Initial Copy: Number of volumes that can be copied concurrently during an initial copy operation. • Error Code: Displays an error code if an error occurs when you click Apply.
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.
secondary data volumes and splits journal groups that will become a Continuous Access XP Journal pair. This ensures consistency of secondary data volumes and makes the secondary data volumes usable. If possible, the horctakeover command attempts to restore pairs to reverse primary and secondary data volumes. If the horctakeover command executes successfully, your business tasks will be taken over to (or your business application will be started at) the secondary site, using secondary data volumes.
Access XP Journal terminology is now reversed: the original primary data volumes are now secondary data volumes, and the original secondary data volumes are now primary data volumes. 1. At the primary and secondary sites, ensure that all Continuous Access XP Journal components are fully operational and are free from failures. 2. Ensure that pair status of primary and secondary data volumes in all Continuous Access XP Journal pairs is PAIR.
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.
Figure 73 Using Continuous Access XP Journal in NAS Blade systems 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.
Scope related to RAID Manager XP • Relationships of failover and failback on RAID Manager XP and NAS Blade system RAID Manager XP is not subject to failover and failback by the NAS Blade system. Prepare the same configuration file as the main node in the standby node except for ip_address in the HORCM_MON section. If a failure occurs in the NAS package while RAID Manager XP is operating and failover occurs, start RAID Manager XP in the standby side and continue operating Continuous Access XP Journal.
You should not operate Business Copy XP settings by the Snapshot feature using RAID Manager XP commands. You should also not operate Business Copy XP set by the Snapshot feature using Command View XP or XP Remote Web Console. Operate Business Copy XP using the GUI screen in NAS Backup Restore or the command in the Snapshot feature.
• NAS Sync Image (Required when copying the file system managed by NAS Sync Image using Continuous Access XP Journal) Notes on operations • UID for user registration and GID for group registration in primary/secondary sites With Continuous Access XP Journal, the file owner information (UID and GID in the primary site) and quota and ACL information for each user (UID in the primary site) and group (GID in the primary site) included in the source file system are copied to the secondary site.
When the host name is specified in the RAID Manager XP configuration definition file and you change the following system configurations, you must change the configuration definition file: • Editing the /etc/hosts file (when resolving the host name using the /etc/hosts file) • Changing registration information on NIS servers, or changing settings for NIS servers (when resolving the host name using NIS) • Changing registration information on DNS servers, or changing settings for DNS servers (when resolving th
you must prepare a total of four RAID Manager XP configuration files, assuming one instance per NAS package. For two instances, you need a total of eight configuration files. In NAS Blade, you can define one or two RAID Manager XP instances per NAS package. To operate only the Continuous Access XP Journal pairs in RAID Manager XP, you need one instance. To operate pairs where Continuous Access XP Journal and Business Copy XP are cascaded, you need two instances.
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.sh command, you can continue using the file system in the primary site. Alternatively, you can delete it and create a file system with the same configuration using the same LU in Create file systems in Continuous Access XP Journal P-VOLs.
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 38, change each entry of ip_address in HORCM_MON to an appropriate value.
Table 40 Command execution that lists LUs constituting the file system sample $ sudo horc_devlist | grep 11( 17) 64 OPEN-V(3.906GB) 12( 18) 12 OPEN-V(3.906GB) 13( 19) 66 OPEN-V(3.906GB) ':sample$' -- -- Normal Own File:sample -- -- Normal Own File:sample -- -- Normal Own File:sample The device file and device names of the LU entry (line) you want to control using RAID Manager XP are changed to an appropriate name.
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.
You can check the device file and LDEV numbers for device files constituting a file system by issuing the horc_devlist command in the NAS package in which the P-VOL is connected (see Table 46). Compare this with Table 45. For information about the horc_devlist command, see the NAS Backup Restore User's Guide. Table 46 Checking device file and LDEV numbers of P-VOLs $ sudo horc_devlist | grep 11( 17) 64 OPEN-V(3.906GB) 12( 18) 12 OPEN-V(3.906GB) 13( 19) 66 OPEN-V(3.
backing up the file system copied from the Continuous Access XP Journal P-VOL to S-VOL periodically using Business Copy XP. Figure 75 shows an example of a cascade configuration of Continuous Access XP Journal and Business Copy XP.
Table 50 Example of RAID Manager XP configuration definition file at primary site (instance 16) HORCM_MON #ip_address service poll(10ms) timeout(10ms) 123.456.78.
Table 51 Example of RAID Manager XP configuration definition file at secondary site (instance 16) HORCM_MON #ip_address service poll(10ms) timeout(10ms) 123.456.80.
Table 52 Example of RAID Manager XP configuration definition file at secondary site (instance 17) HORCM_MON #ip_address service poll(10ms) timeout(10ms) 123.456.80.
ssh {-1|-2} nasroot@fixed-IP-address-of-the-NAS-package 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.
Figure 76 Overview of remote copy operations Starting CA XP Journal ops. and creating CA 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.
Resynchronizing Continuous Access XP Journal volume pairs You must delete the target file system using the enas_fsdelete command before resynchronizing the volume pair. Deleting the target file system, however, clears only the file system's management area. Resynchronizing the volume pair is not the same as performing an initial copy.
NOTE: If you start the split pair operation with offline backup, but have not yet connected the target file system to the NAS package at the secondary site, finish connecting the target file system to the NAS Package, and then start the previously described operations.
• Non-LVM: ?sudo horc_svol_delete -d Device-file-number • LVM: ?sudo horc_svol_vmdelete -d Device-file-number, ... 4. At the primary and secondary sites, stop RAID Manager XP. sudo horcmshutdown.sh (1-instance configuration) sudo horcmshutdown.sh 16 17 (2-instances configuration) Disaster recovery operations This section describes procedures for disaster recovery operations. You might also want to refer to Settings for disaster recovery.
ssh {-1|-2} nasroot@fixed-IP-address-of-the-NAS-package 4. At the primary and secondary sites, start RAID Manager XP. sudo horcmstart.sh (1-instance configuration) sudo horcmstart.sh 16 17 (2-instances configuration) 5. At the primary site, delete CIFS/NFS shares in the old P-VOL, unmount the old P-VOL, and delete the file system in the old P-VOL.
• Non-LVM: sudo horc_svol_define -d Device file number • LVM: sudo horc_svol_vmdefine -d Device file number, ... 16. At the primary site, enter the following command to reverse resynchronization of data from the primary site to the secondary site. sudo pairresync {-g group name|-d volume name} -swaps 17. At the primary site, mount the new P-VOL using the enas_fsmount command, and create CIFS/NFS shares using the enas_nfscreate and enas_cifscreate commands.
Table 53 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.log RAID Manager XP trace log /home/nasroot/log*/curlog/horcmlog_HOST/horcm_PID.trc Command trace /home/nasroot/log*/curlog/horcmlog_HOST/horccc_PID.
NOTE: If you read RAID Manager XP log files using Windows, you must ensure that the text viewer you are using can display text containing the line feed code LF (Line Feed). 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.
If 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. 2. Confirm all Continuous Access XP Journal volume pair statuses that are using the external volume with problems using the pairdisplay command. If the status of the Continuous Access XP Journal volume pair is PSUE, delete the volume pair using the pairsplit -S 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 54 General troubleshooting information for Continuous Access XP Journal Error Corrective action The Command View XP management station hangs, or Continuous Access XP Journal operations do not function properly. Ensure that the problem is not caused by the management station or Ethernet hardware or software, and restart the management station. Restarting the Command View XP management station does not affect Continuous Access XP Journal operations in progress.
Table 55 Troubleshooting problems with status of logical paths Pair status Description Corrective action Initialization Failed Link initialization procedure for the remote array failed. Ensure that the local and remote arrays are connected correctly via cables. Ensure that you entered the correct serial number (S/N) of the remote array and the correct path parameters (for example, local array port, remote array port, and controller ID).
Pair status Description Corrective action Invalid Port Specified port is not configured as an initiator port, or this path already exists. Ensure that the topology settings of the local and remote array ports are correct. Ensure that you entered the correct remote array S/N and path parameters (for example, local array port, remote array port, and controller ID). After that, delete the failed path. You might need to change the minimum paths setting or delete the remote array to delete the path.
Pair status Description Corrective action Logical Blockade Failure occurred at a local array port. Repair the port status on the local array. Failure occurred at a remote array port. Repair the port status on the remote array. After that, delete the path using the Delete Path or Delete DKC command, and add the path and remote array again using the Add Path or Add DKC command. Failure occurred in the path relay equipment. Repair the path relay equipment.
Table 56 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 57 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. As a result, one of the following could not complete: primary-secondary array communications, journal creation, copy operation, restore operation, staging process, or de-staging process.
• 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 recent service information messages (SIMs) 220 Troubleshooting
13 Power management for arrays and network relay devices This appendix explains power management for arrays and network relay devices during remote copy operations.
When powering off primary arrays To power off the primary array 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. CAUTION: To power off the primary array, contact your HP account support representative and ask them to power off the array.
NOTE: To power off the primary and secondary arrays at the same time, see When powering off primary and secondary arrays at the same time.
Power management for arrays and network relay devices
14 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
15 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 58 Typical SIM showing reference code and SIM type Reference code Severity Description SIM generated by SVP log file Byte 22 Byte 23 21 80 Moderate Logical path is blocked due to a failure. Secondary array SIM Log 21 81 Service Logical path is restored. Secondary array SSB Log 21 82 Moderate Line failure reported by the extender. Secondary array SIM Log DC 0X Serious Volume being used by the primary data volume is suspended. Path recovery is impossible.
Index Symbols 3DC configuration cascading, 74 multi-target, 77 A Add DKC pane, 105 applications, combining with, 65 audience, 13 B bandwidth, 64 base-journals, 29, 31 bitmap areas, number of, 50 bitmaps, difference, 29, 31 Business Copy XP, 17, 69 C cable length, maximum, 27, 47, 59 cache pinned track recovery, 225 requirements, 53 cascading configuration, 74 CKD-to-FBA conversions, 31 combining with other applications, 65 Command View XP troubleshooting management station, 213 Command View XP or XP
FBA-to-CKD conversions, 31 Fibre Channel connections, 17, 27, 47, 59 file recovery, 181 filtering Pair Operation pane list, 146 G graph, Usage Monitor, 167 graphical user interface, 28 H hardware installation, 58 help obtaining, 14 History Operation pane components, 172 displaying, 171 saving information as snapshot file, 174 horctakeover command, 182 host failover software, 21, 28 HP technical support, 14, 219 HP Performance Control XP, 68 I initial copy operations, 29 initiator ports, 28, 103 instal
mirrors deleting data volumes from, 139 restoring, 138 splitting, 137 modes, optional, 22 monitoring displaying Usage Monitor pane, 165 graph, displaying, 167 saving data, 170 starting, 166 stopping, 167 multi-target configuration, 77 multiple secondary arrays, 21 multiple secondary volumes, 48 multivolume files, 48 N network relay devices, powering off, 221, 221, 223 nonvolatile storage (NVS), 53 NVS (nonvolatile storage), 53 O obtain function, 20 one-to-one copy operations, 48 operational parameters,
primary arrays Command View XP or XP Remote Web Console, 28 components, 22 configuring, 64 control units, 22 copy operations, 21 copying journals, 32 defined, 99 Fibre Channel interface connections, 27 initial copy operations, 29 initiator ports, 28 journal copy function, 20 journal data, storing, 32 journal group operations, 34 journal groups, 23 logical paths adding, 109 configuring, 105 deleting, 110 options, 108 removing relationships, 114 status, viewing, 112 NVS (nonvolatile storage), 53 pair status,
saving history information as snapshot file, 174 pair status as snapshot file, 147 Usage Monitor data, 170 secondary arrays Command View XP or XP Remote Web Console, 28 components, 22 configuring, 64 control units, 22 copy operations, 21 copying journals to, 32 defined, 99 Fibre Channel interface connections, 27 file recovery, 181 initial copy operations, 29 initiator ports, 28 journal copy function, 20 journal data, storing, 32 journal group operations, 34 journal groups, 23 logical paths adding, 109 confi
T target ports, 28, 103 technical support HP, 14 service locator website, 15 technical support, HP, 219 temporary path failures, 62 text symbols, 14 three data center (3DC) configuration cascading, 74 multi-target, 77 timeout errors, 214 transfer speed, 60 troubleshooting calling HP Support, 219 Command View XP management station, 213 Command View XP or XP Remote Web Console, 213 error codes, 219 pair status, 214 powering off arrays, 221 timeout errors, 214 TrueCopy for Mainframe, 18 234 U update copy o
