Overview
The Create EVM Job page appears when you click the New Job button that appears on the EVM Jobs View, EVM Network View, and EVM Storage View. The page is organized like a wizard to guide you through the steps of creating an EVM job. The information below will assist you in completing the steps. For information on a specific step, click the step number: 1 2 3 4 5 6.
Job Name - mandatory entry
Enter a name for your EVM job. The wizard will create the job file using this name. Choose a name that reflects the type of activity the job will perform.
Examples
backup_daily_sales
replicate_transactions_for_system_ migration_test
copy_west_region_for_offsite_disaster_recovery
Notes
When EVM saves the job, the prefix stor1_ will be added to the name.
The name can be up to 45 characters. You may include special characters that the host operating system allows in file names. When EVM saves the job, spaces are replaced with the underscore character.
Do not include a file extension. EVM will append the correct extension.
The Job Name cannot be changed after the wizard creates the job file. To rename a job, you must first delete the job, then create another job using the new name.
Created By - optional entry
Enter the name of the organization or person who is creating the EVM job. The name can be an organization, person, or a combination.
Examples
Cheryl_Jones
Accounting_Department_(Bill Moore)
West_Region_Manager
Notes
The Created By name cannot be changed after the wizard creates the job.
The name can be up to 45 characters. You may include spaces and special characters.
STEP 2 Unit Replication Method - mandatory selection
Select the unit replication method that is appropriate for your job.
Notes
IMPORTANT: Scheduled Jobs. The wizard presents replication methods that are available when you create the job. The method you select may not be available later due to changes in the configuration of the StorageWorks subsystem.
As a general guide, snapshot is the more appropriate method when you want to replicate a unit as quickly as possible. The clone method is preferable when you need fault tolerance more than fast replication.
The clone method is available on StorageWorks subsystems that have Array Controller Software version 8.5. The clone method cannot be used with any units that are configured as RAID 3/5.
The snapshot method is only available with certain versions of Array Controller Software. The snapshot method can be used with disks configured as RAID 3/5.
STEP 3 Units to Replicate - mandatory selection
Select one or more units that you want to replicate in this job.
Notes
IMPORTANT: Scheduled Jobs. The wizard presents units that are available when you create the job. The units that you select may not be available later due to changes in the configuration or utilization of the StorageWorks subsystem.
The wizard shows all units that are available for replication on the StorageWorks subsystem. Select only the units that are appropriate to this job.
Selecting too many units in a single job increases the chance that there may be insufficient resources on the StorageWorks subsystem to perform the replication. For more information, see EVM Job Rules.
Some units on the StorageWorks subsystem may not appear in the display. This can occur as a result of the configuration of the host computer and StorageWorks subsystem.
STEP 4 Application I/O Handler - optional selection
Default
The default selection is None. In this case, the EVM job will not communicate with an application to suspend and resume file I/O.
IMPORTANT: Using the Application I/O Handler. It is highly recommended that you enable this feature in circumstances where the selected units could experience file I/O during the replication process.
Enabling the Application I/O Handler Feature
To enable the feature, scroll through the list of available computers and select the relevant Application Server. This is the server that is performing file I/O with the units that you want to replicate in this job.
IMPORTANT: Running EVM Jobs. To run the EVM job with the application handling feature enabled, you must have already created the script files that you reference.
In the Pre-split Command box, enter the path and file name of the I/O handler that will suspend I/O just before the BCV units are split from the source. In most cases, the I/O handler is a script file that runs the required I/O commands. See notes below regarding path and file names.
Select the EVM Job Mode. The default is Wait. This instructs EVM to wait for the return of a completion code from the I/O handler before EVM proceeds with the split. To instruct EVM to proceed with the split without waiting, select Don't Wait.
In the Post-split Command box, enter the path and file name of the I/O handler that will resume I/O after the BCV units are split from the source. In most cases, the I/O handler is a script file that runs the required I/O commands. See notes below regarding path and file names.
Select the EVM Job Mode. The default is Wait. This instructs EVM to wait for the return of a completion code from the I/O handler before EVM proceeds with its next task. To instruct EVM to proceed without waiting, select Don't Wait.
Examples
/sales/suspend_transactions.sh
/sales/resume_transactions.sh
Notes
The file name must include the extension.
The entry may be up to 255 characters.
You may omit the path name and enter just the file name under the following condition: on the server you selected, the path to the file is included in the system’s Path variable.
The path and file name entry cannot be changed after the wizard creates the job. If you change them, you must first delete the EVM job, then create a job using the new path and file name.
The wizard does not validate script files or commands for compliance with the application’s syntax rules. It is recommended that you thoroughly test the script files before they are run from an EVM job.
When creating an EVM job that will be run later, you may specify the path and file name before creating them. However, the preferred practice is to create and thoroughly test the script files prior to creating the EVM job.
STEP 5 BCV Mounting - optional, but recommended selection
Default
The default selection is None. In this case, the BCV units that EVM creates will only be recognized by the StorageWorks subsystem. They will not be mounted on a server or be recognized by host operating systems or application programs.
IMPORTANT: Mounting BCV Units. If you do not choose to mount the BCV units as part of this job, they can be mounted later. However, to mount the BCV units later will require use of StorageWorks programs, such as CLI Window, and a reboot of the server on which they are being mounted.
To Mount BCV Units on a Server
To mount the BCV units, scroll through the list of available computers and select the one you want. If you are creating this job to perform a tape backup, the computer that you select should be the tape backup server.
In the mount as box, enter a mount point (directory) and slice (partition) for each unit that you are replicating in this job. Enter the mount points in the same sequence that the selected units appear in Step 3.
IMPORTANT: Selecting Existing Mount Points (Directories). The selected mount points must not be in use by other devices when the EVM job is run. If a mount point is busy when EVM attempts to mount a BCV unit, it will cause the EVM job to fail.
IMPORTANT: Selecting Slices (Partitions). The slice number of the BCV unit must be the same as the slice number of the source unit. For further information, see Slices (Partitions).
Examples
/usr/d1_bcv:6 (BCV of unit D1:slice 6)
/home/daily_sales_bcv:7 (BCV of daily_sales:slice 7)
You have selected StorageWorks units D1 (slice 6), D2 (slice 5) and D100 (slice 7) to replicate. (They appear in that order in Step 3). You want the BCV duplicate of unit D1 to appear on the server as directory "/usr/d1_bcv", the duplicate of D2 as directory "/opt/d2_bcv" and the duplicate of D100 as directory "/usr/d100_bcv". You would enter: "/usr/d1_bcv:6, /opt/d2_bcv:5, /usr/d100_bcv:7" in the mount point box.
Notes
When entering a series of mount points, separate them with commas or spaces.
When the job is run, EVM will check the selected server to see if any of the mount points are being used by other devices. When one or more mount points are being used by other devices, EVM will not proceed with the job.
The wizard only presents servers that are correctly configured for the EVM Network.
IMPORTANT: Scheduled Jobs. The wizard presents servers that are available when you create the job. The one that you select may not be available later due to configuration changes in the EVM Network.
STEP 6 Backup Handler - optional selection
Default
The default selection is None. After BCV units are created and mounted on a server, EVM will not issue a command to a backup application to start a tape backup.
Enabling the Backup Feature
In the Backup Command box, enter the path and file name of the backup handler. In most cases, the backup handler is a script file that runs the required backup application job. See notes below regarding path and file names.
Select the EVM Job Mode. The default is Wait. This instructs the EVM job to wait for the return of a completion code from the backup handler before it ends. To instruct the EVM job to end without waiting, select Don't Wait.
IMPORTANT: Running EVM Jobs. To run the EVM job with the backup handling feature enabled, you must have already created the script files that you reference.
Examples
/sales/backup_daily_sales.sh
/west_inventory/daily_backup.sh
Notes
The file name must include the extension.
The entry may be up to 255 characters.
You may omit the path name and enter just the file name under the following condition: on the server you selected, the path to the file is included in the system’s Path variable.
The path and file name entry cannot be changed after the wizard creates the job. If you change them, you must first delete the EVM job, then create a job using the new path and file name.
The wizard does not validate script files or commands for compliance with the application’s syntax rules. It is recommended that you thoroughly test the script files before they are run from an EVM job.
When creating an EVM job that will be run later, you may specify the path and file name before creating them. However, the preferred practice is to create and thoroughly test the script files prior to creating the EVM job.
For EVM to run the backup application, you must specify the correct server and mount points in Step 5.
Save
To save the EVM job, click Save. The Confirm and Save EVM Job page will appear. You can confirm that your choices are correct and then save the job.
Back (clear entries)
To clear all entries and return to the EVM Network View, click Back.
Notes
To view information on other EVM pages without clearing your entries, use the navigation line at the top of the EVM page or use the browser navigation.