HP StorageWorks Auto LUN XP user guide for the XP12000/XP10000 Part number: T1715-96008 Sixth edition: March 2008
Legal and notice information © Copyright 2005, 2008 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 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Supported storage platforms and firmware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Intended audience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Prerequisites. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Manual Migration tab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pane contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tree view . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table view . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Tables 1 2 3 4 5 6 7 8 9 10 11 12 13 Parity group table boxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Attribute tab tree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . History tab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
About this guide This guide provides information about the following: • ”Auto LUN XP for the XP12000/XP10000”on page 11 • ”Auto LUN XP features” on page 11 • ”Auto LUN XP tasks” on page 11 • ”Reserve volumes” on page 12 • ”Volume migration” on page 12 • ”Estimating usage rates” on page 13 • ”Automatic migration” on page 14 • ”Manual migration” on page 16 • ”Requirements and restrictions” on page 16 • ”Starting Auto LUN XP” on page 24 • ”Creating and executing migration plans” on page 30 • ”Troubleshooting A
• HP StorageWorks Performance Control XP user guide • HP StorageWorks XP Disk/Cache Partition user guide • Hitachi True Copy for z/OS® for the XP128/XP1024/XP10000/XP1200 • HP StorageWorks RAID Manager XP user guide You can find these documents at http://www.hp.com/support/rwc/manuals. 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.
• Operating system type and revision level • Detailed, specific questions For continuous quality improvement, calls may be recorded or monitored. Subscription service HP strongly recommends that customers register online using the Subscriber's choice web site: http://www.hp.com/go/e-updates. Subscribing to this service provides you with e-mail updates on the latest product enhancements, newest driver versions, and firmware documentation updates as well as instant access to numerous other product resources.
1 Auto LUN XP for the XP12000/XP10000 Auto LUN XP monitors resources in disk arrays and hosts connected to disk arrays. Auto LUN XP works with open system and mainframe volumes. You can analyze the information Auto LUN XP provides to optimize data storage and retrieval on disk arrays, resolve activity bottlenecks, and optimize volume allocation.
Reserve volumes The reserve volume function reserves target volumes for automatic and manual migration operations. After you reserve a number of target volumes, Auto LUN XP maintains this number of reserved volumes by swapping the reserve attribute after each migration operation (following the migration operation, the original source volume becomes a reserved volume).
Figure 2 Data Flow after an Auto LUN XP operation NOTE: Immediately after a source volume and a target volume replace each other, RAID level before the replacement might be displayed or an internal error might occur in the disk array. In this case, select Refresh to update the display. Estimating usage rates The estimate function calculates expected usage of source and target parity groups after a proposed volume migration.
group B will probably decrease and I/O response time for parity group A might increase, while the overall throughput might increase or decrease. From an open-system host, you can also use RAID Manager command to perform manual volume migration. To use RAID Manager to perform Auto LUN XP migration, install Auto LUN XP on the XP12000/XP10000. For more information, see ”Using RAID Manager for manual Auto LUN XP migrations” on page 44.
The auto migration function can move a high-usage volume to a higher HDD class group, forcing a low-usage volume out of that HDD class group. To do so, the auto migration function requires a minimum of 5% difference in estimated disk usage between the two volumes. If the difference is less than 5%, this migration is considered ineffective and the volume is not moved.
Auto migration sequence of events The following are the typical steps to set up an auto migration plan: 1. Analyze monitor data. You specify the monitor data term to be analyzed. 2. Make auto migration plan. You specify the auto migration plan’s parameters. 3. Perform auto migration plan. You specify when the auto migration plan is executed. 4. Analyze monitor data to confirm tuning results. Manual migration Use manual migration to select and migrate logical volumes under direct, manual control.
Table 2 Movability of volumes in pairs consisting of CV and normal values (continued) Source volume Target volume Normal volume (OPEN-V with VMA setting) Not movable Not movable Movable Not movable Not movable Not movable CV (not OPEN_V) Not movable Not movable Not movable Movable Not movable Not movable CV (OPEN-V without VMA setting) Not movable Movable Not movable Not movable Movable Not movable CV (OPEN-V with VMA setting) Not movable Not movable Movable Not movable Not mov
Universal Replicator: If the status of a Universal Replicator volume is COPY or PAIR, the volume cannot be used as a source volume. Note that if you specify a Universal Replicator volume as a source volume, you cannot specify an external volume as the target volume. ShadowImage for z/OS® and ShadowImage: Usage of ShadowImage for z/OS® and ShadowImage volumes as a source volume depends on the status and cascade configuration of the volume as described below.
• iSCSI volumes (for example, volumes to which paths are defined from iSCSI ports) • Volumes in abnormal or inaccessible conditions (for example, fenced) • Volumes Continuous Access XP Journal uses • Volumes specified with the Read Only or Protect attribute • Volumes that Volume Security disabled for use as secondary volumes • Volumes that LUN Security XP Extension specified as Read Only or Protect or disabled for use as secondary volumes • Volumes reserved by a migration program other than Auto LUN XP • Vo
Since this equation is true, you can create 20 migration plans of OPEN-3 volumes. Calculating differential tables for mainframe volume migration When you migrate mainframe volumes, use the following equation to calculate the total number of required differential tables per migration plan: ( X + Y ) × 15 ÷ Z = Total number of differential tables per migration plan , where: • X: The number of cylinders in the volume to be migrated.
Table 5 Control cylinders by emulation type (continued) 3390-MA 53 3390-MB 53 3390-MC 53 3390-LC 23 NF80-F 22 NF80-K 7 NF80-KA 7 NF80-KB 7 NF80-KC 7 Calculating differential tables for open-system volume migration When you migrate open-system volumes, use the information in Table 6 to calculate the total number of required differential tables per migration plan.
The following table shows the number of control cylinders according to the emulation type.
• Business Copy XP • ShadowImage for z/OS • Hitachi FlashCopy Mirroring • Hitachi FlashCopy Mirroring Version 2 • Snapshot XP • Flex Copy XP Progress value If a volumes with heavy and light workloads are migrated together, the migration might not complete (even though the progress value indicates 100%). If this happens perform one of the following tasks. • Reduce the workloads between hosts and the disk array and complete the migration.
Powering off disk arrays To turn the power off at the disk array, ensure that volume migration is finished. If migration is not finished, HP strongly recommends that you do not turn off the power until the migration is finished. If you turn the power off while migration is in progress, data migration stops and some of the data are not migrated. If you turn the power on later, Auto LUN XP resumes data migration.
NOTE: If a hyphen appears in monitoring data columns on the Auto LUN pane, statistics cannot be collected for that item. Tabs and selection trees When you click one of the tabs (WWN, Port-LUN, LDEV, or Physical) in the lower-left portion of the pane, the pane displays different trees of items you can click. The top left of the pane shows the status of the Short range Monitor and Long range Monitor. For information about these monitors, see ”Auto LUN Monitoring Options pane” on page 50.
• Port Control: System performance is controlled by the upper limits and threshold specified in the Port tab of the Performance Control pane • WWN Control: System performance is controlled by the upper limits and threshold specified in the WWN tab of the Performance Control pane • No Control: System performance is not controlled by Performance Control For further information, see HP StorageWorks Performance Control XP user guide. Graph section The graph shows statistics displayed in the table.
WWN, Port-LUN, LDEV, and Physical tabs Figure 6 WWN, Port-LUN, LDEV, and Physical tabs Click one of the tabs (WWN, Port-LUN, LDEV, or Physical) in the lower-left portion of the pane to view data about ports, LDEVs, or physical components. When you click a tab, the selection tree in the lower-left portion of the pane changes. You can then click entities to display information in the table and graph sections of the pane.
After replacing the micro program, monitoring data is not stored until a service technician releases the SVP from Modify mode. NOTE: If a host bus adapter’s WWN is displayed in red in the tree, the host bus adapter is connected to two or more ports, but Performance Control does not control traffic between the HBA and some ports. For information about controlling the traffic between the HBA and all connected ports, see HP StorageWorks Performance Control XP user guide. Port-LUN tab This tab displays ports.
If the parity group icon represents a single parity group 1-3, the text 1-3 is displayed on the right of the icon. If the parity group icon represents two or more connected parity groups, all connected parity groups are displayed to the right of the icon. For example, if parity group 1-3 is connected with parity group 1-4, the text 1-3[1-4] is displayed on the right of the parity group icon. All parity groups connected with 1-3 are enclosed by square brackets.
Table 11 Auto LUN pane, Physical tab icons (continued) Icon Description CHA-1E (Fibre 16Port)* a Fibre Channel port in High Speed mode. LUN security is applied to this port. CHA-1E (Fibre 16Port)* a Fibre Channel adapter in Initiator/External MIX mode CHA-1E (Fibre 16Port)* a Fibre Channel port in Initiator/External MIX mode. LUN security is applied to this port Long-wave Fibre Channel adapter in standard mode.
Manual Migration tab Use this tab to create and execute migration plans for manually migrating volumes. Figure 7 Manual Migration tab Pane contents This pane contains the following items: • Monitoring Term: Displays the monitoring period specified in the Auto LUN pane, analyzes disk usage information collected by Auto LUN XP during the monitoring period, and calculates estimated usage rates of the source and target parity groups after a proposed volume migration.
• Source LDEV columns: • LDEV: Logical volume ID • Emulation: Emulation type of the logical volume • Capacity: Capacity of the logical volume • RAID: RAID type • PG: Parity group ID (frame and group number) • HDD: Type of hard disk drive • CLPR: The numbers and names of CLPRs corresponding to each parity group to which the logical volume belongs • Target LDEV columns: • LDEV: Logical volume ID • RAID: RAID type • PG: Parity group ID (frame and group number) • HDD: Type of hard disk drive Buttons • Set: Add
7. In the tree, double-click a parity or external volume group. A list of logical volumes in that group appears. Logical volumes that can be migrated are indicated by the icon. 8. From a volume with the icon, click the source volume (logical volume to migrate), and click the S button. The selected volume is defined as the source volume. 9. In the tree, double-click the parity group to which you want to migrate the source volume. A list of logical volumes in that parity group appears.
NOTE: Auto migration cannot use external volumes or volumes reserved by another program. Figure 8 Auto Migration tab Pane contents This pane contains the following items: • Monitoring Term: Displays the monitoring period specified in the Auto LUN pane, analyzes disk usage information collected by Auto LUN XP during the monitoring period, and calculates estimated usage rates of the source and target parity groups after a proposed volume migration.
Auto Plan parameter section • Auto Migration Function: If you click ON, logical volumes are automatically migrated. If you click OFF, they are not. • Sampling Term: The auto migration function analyzes resource usage within the SAMPLING TERM and creates auto migration plans based on that analysis. Use Sampling Term to narrow resource usage statistics to be analyzed. • Date and Time: Specifies a date/time frequency for analyzing resource usage.
10. Click Close. Deleting auto migration plans • To delete all auto migration plans, click Delete All. • To delete one auto migration plan, click the plan in the table, and click Delete. Remaking auto migration plans If auto migration operations do not produce enough improvements in disk access performance, discard the current migration plans and create new auto migration plans. 1. In the Auto LUN pane, click Physical. 2. Click Plan. The Manual Migration tab appears. 3. Click Auto Migration.
Auto LUN XP uses collected usage statistics to estimate usage rates of parity groups after a proposed volume migration. • Attribute tree: Lists HDD classes. When you double-click an HDD class icon, a list of parity groups in that HDD class appears. • Apply button: Applies settings in the Attribute tab to the disk array. • Reset button: Discards changes in the Attribute tab and restores the original settings. • Close button: Closes this pane.
• CLPR: Number and name of the CLPR that corresponds to the parity group to which the logical volume belongs, in the format CLPR number:CLPRname. For more information about CLPRs, see the HP StorageWorks XP Disk/Cache Partition user guide. • Owner: Program that reserved the volume. If Auto LUN XP reserved this volume, USP is displayed. If another program reserved this volume, Other[XX] is displayed, where [XX] is the program ID.
5. Right-click the parity group you want to fix. Figure 12 Attribute tab tree 6. Click Fixed PG. The parity group becomes fixed and is marked with an icon. 7. Click Apply. Releasing (unfixing) parity groups To move logical volumes in a fixed parity group to another parity group, change the fixed parity group to a normal parity group. To do this, go to the Attribute tab, right-click the fixed parity group, and click Normal PG.
History tab Use this tab to display information about automatic and manual migration operations that occurred in the past. You can find out when migration operations took place and whether they finished successfully.
Viewing migration history logs The History tab displays logs of automatic and manual migration operations. 1. In the Auto LUN pane, click Physical. 2. Click Plan. The Manual Migration tab appears. 3. Click the History tab. The History tab appears. To view logs of auto operations, look at Auto Migration History. To view logs of manual migration operations, look at Migration History. The migration logs may display the following messages.
Migration was stopped (other reason): Migration Stopped (CU:LDEV->CU:LDEV) yyyy/mm/dd hh:min:sec (XXXXXXXXXXXX) (a) (b) (c) (d) (a) Source volume (b) Target volume (c) Canceled time (d) Reason for cancellation: • No reserve volume. No reserved volume is set. Please make another plan. • Reserve volume emulation is different. The reserved volume's emulation type is different. • Utilization check. The usage rate is over the limit, or there is no monitor data. • Migration failed. Error code: XXXX.
Failed making an auto migration plan: yyyy/mm/dd hh:min : MakePlan is terminated by error: XXXXXXXXXX (a) (b) (a) End time (b) Reason for failure: • Cannot make proper migration plan: The program could not make a proper migration plan. Check the number of the reserved volumes and their locations. Check the maximum disk usage and change if necessary. • Failed to get XXXXXXXXXX (information or data: The program failed to get the information or data to make a plan.
The migration logs may display the following messages: Table 12 Migration log messages Message Meaning Migration Start Migration operation started Migration Complete Migration operation completed successfully Migration Cancel User canceled the migration operation Migration Failed Migration operation failed Migration Give Up Migration operation was canceled by Auto LUN XP (for example, volume usage rate exceeded specified maximum) Migration Canceled by Controller The migration operation was canc
The following is an example of the procedure for using RAID Manager commands to perform manual Auto LUN XP migrations. In this example, group1 is the group name in the RAID Manager configuration file, and pair1 is the name of the volume pair that is a target of the operation. 1. Start RAID Manager. 2. Type the following command for an SMPL pair to begin the Auto LUN XP migration: paircreate -g group1 -d pair1 -m cc -vl When Auto LUN XP migration starts, the status of the pair changes to COPY. 3.
• When you try Auto LUN XP migrations or migration cancellations using RAID Manager, EX_CMDRJE might appear and the command might be refused, depending on the condition in the disk controller. Table 13 shows error codes, causes, and solutions for this case. Table 13 Errors during Auto LUN XP migrations or migration cancellations using RAID Manager Error code Cause Solution 2051 Migration target volume is being used with LUN Security XP Extension.
Table 13 Errors during Auto LUN XP migrations or migration cancellations using RAID Manager Error code Cause Solution 2342 Migration target volume is already specified as an Auto LUN XP target volume. Release the Auto LUN XP volume, and perform the volume migration. 2344 Volume specified by the migration cancellation is not a migration target volume. RAID Manager is trying to perform the migration cancellation for a pair that is not migrating. Check the configuration definition file.
Table 13 48 Errors during Auto LUN XP migrations or migration cancellations using RAID Manager Error code Cause Solution 238C Migration target volume is a Business Copy XP reserved volume. Cancel the Business Copy XP reservation, and perform the volume migration. 239D Migration target volume is an ENAS system volume. Volume migration cannot be executed. Check the configuration definition file.
2 Auto LUN/Performance Control Base Monitor Auto LUN XP statistics Disk arrays automatically collect statistics twice a day (in the morning and afternoon). Auto LUN XP monitors disk arrays and obtains usage statistics about resources such as front-end and back-end processors, hard disk drives, and logical volumes every 15 minutes. Auto LUN XP displays statistics collected for the last three months. Statistics over three months old are discarded.
period. The pane also displays a graph indicating how the write pending rate changed within that period. • Usage statistics about access paths. An access path is a path through which data and commands are transferred within a disk array. In a disk array, channel adapters control data transfer between hosts and cache memory. Disk adapters control data transfer between cache memory and disk drives. Data transfer does not occur between channel adapters and disk adapters.
NOTE: Settings in the Monitor Options pane work with settings for Continuous Access XP and Continuous Access XP Journal. Therefore, changes you make to one pane affect the settings in the other panes. XP12000/XP10000 disk arrays NOTE: To begin monitoring the disk array, Command View XP or XP Remote Web Console must be in Modify mode. A yellow background color signals Modify mode. A gray background color signifies View mode. You cannot change in the Monitoring Options pane in View mode.
Viewing parity group usage statistics Auto LUN XP monitors parity groups, and displays average and maximum usage rates for the specified period. Auto LUN XP also displays a graph illustrating changes in parity group usage within that period. 1. In the Auto LUN pane, click Physical, and click the Parity Group folder. The table displays usage statistics about the parity group.
NOTE: If an exclamation mark (!) appears before a usage rate, that parity group usage rate might be inaccurate because the configuration has changed (for example, volumes have been moved by a migration plan or changed by VSC). 3. To display a graph illustrating changes in usage rates for parity groups, click the parity groups in the table, and click Draw. The table displays the following items: • Graph column: The check mark icon indicates the graph is currently illustrating data for that item.
1. In the Auto LUN pane, click Physical, and click the CHA folder. The table displays usage statistics about channel adapters (CHAs). Figure 18 CHA usage statistics 2. To display a graph illustrating changes in usage rates for channel adapters, click the channel adapters in the table, and click Draw. The table displays the following items: • Graph column: The check mark icon indicates the graph is currently illustrating data for that item. • ID: Channel processor ID numbers. • Usage: The Ave.
In the Auto LUN pane, click Physical, and click the ACP folder. The table displays a list of disk adapters. Figure 20 DKA usage statistics The table displays the following items: • Adapter: Disk adapter ID numbers. • Cluster-1: If the Cluster-1 column displays 0 and the Cluster-2 column displays a hyphen, the disk adapter is located in Cluster-1. • Cluster-2: If the Cluster-2 column displays 0 and the Cluster-1 column displays a hyphen, the disk adapter is located in Cluster-2.
2. Click the DRR below the disk adapter. The table displays usage statistics about DRRs in the disk adapter. Figure 22 DRR usage statistics 3. To display a graph illustrating changes in usage rate for DRRs, click the DRRs in the table, and click Draw. The table displays the following items: • Graph column: The check mark icon indicates the graph is currently illustrating data for that item. • ID: DDR ID numbers. • Usage: The Ave. (Total) column displays the average usage rate in the specified period.
2. Do any of the following: • To check usage statistics about paths between adapters (CHAs and DKAs) and the cache switch, click Adapter-CSW below the Access Path Usage folder. Figure 24 Usage statistics about paths between adapters and the cache switch • To check usage statistics about paths between adapters (CHAs and DKAs) and shared memory, click Adapter-SM below the Access Path Usage folder.
4. Click Apply. Auto LUN XP starts monitoring the disk array. To stop monitoring the disk array, complete the following instructions. If you stop monitoring, Auto LUN XP stops collecting statistics. 1. In the Auto LUN pane, click Monitoring Options. The Monitoring Options pane appears. 2. Click Disable for the Current Status option. 3. Click Apply. Viewing disk drive workload statistics Use the LDEV tab of the Auto LUN pane to check workloads on physical disk drives (parity groups) or logical volumes.
The table displays the following: • Graph column: The check mark icon indicates the graph is currently illustrating data for that item. • Group: Parity group ID. If the ID starts with the letter E, logical volumes in the parity group are external LUs. • LDEV: Logical volume ID. If the ID ends with the symbol #, logical volume is an external LU. • Type: Emulation type. • IO Rate (IOPS): Number of I/O requests to the parity group (or logical volume) per second.
• To view traffic statistics about host ports in a host group, click the host group. Figure 32 Traffic statistics about host ports in a host group • To view traffic statistics about LU paths, click the LUN icon. Figure 33 Traffic statistics about LU paths NOTE: The traffic statistics reported for an LU is aggregated across all LU paths defined for an LU. • I/O rate is the sum of I/Os across all LU paths defined for an LU. • Transfer rate is the total transfer rate across all LU paths defined for an LU.
Viewing HBA/port traffic statistics If Performance Control XP is enabled, Auto LUN XP monitors paths between host bus adapters (HBAs) in host servers and ports on disk arrays. You can view both I/O and transfer rates between HBAs and ports. 1. In the Auto LUN pane, click WWN. The tree view displays a list of PFC groups ( groups. ). The Not Grouped item appears below the PFC • If you double-click a PFC group, host bus adapters ( ) in the PFC group are displayed.
• Max.: Maximum I/O or transfer rate for the specified period. • Response Time: Time between when the DKC receives the read/write command and when the DKC responds to the command. The value is average response time over one minute. • Attribute: Priority of each host bus adapter. Prio. indicates a high-priority HBA (prioritized WWN). Non-Prio. indicates a low-priority HBA (non-prioritized WWN). When calculating I/O rates displayed in the All Prio. row, Auto LUN XP uses I/O rates of other prioritized ports.
Index A G access path usage statistics 50, 56 Attribute tab 36 audience, documentation 7 Auto LUN pane 24 auto migration execution 22 history 40 parameters 14 planning 22 powering off disk arrays 24 reserving target volumes 38 settings 33 Graph section 26 C calculating migration plans 19 cascaded volumes 18 channel adapters (CHAs) usage statistics 49, 53 channel processors (CHPs) usage statistics 49, 54 CHAs (channel adapters) usage statistics 49, 53 CHPs (channel processors) usage statistics 49, 54 con
logs 41 manual 16, 22, 31 operations 12 plans 30 powering off disk arrays 24 RAID Manager, using 44 releasing parity groups 39 target volumes 38 Monitoring Data section 25 Monitoring Options pane 50 Monitoring Term section 25 optimizing performance 11 access path usage 56 channel adapters (CHAs) 53 channel processors (CHPs) 54 data recovery and reconstruction processors (DRRs) 55 disk adapters (DKAs) 54 disk array resources 51 disk processors (DKPs) 55 HBA/port traffic 61 logical volume usage 52 Monitorin
Figures 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Data Flow during an Auto LUN XP Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Data Flow after an Auto LUN XP operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Auto migration function example 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
